Brill's Companion to Seneca - Free Download PDF (2023)

Descripción: Brill's Companion to Seneca...

Brill’s Companion to Seneca

Brill’s Companion to Seneca Philosopher and Dramatist

Edited by

Gregor Damschen Andreas Heil With the assistance of

Mario Waida

LEIDEN • BOSTON 2014

Cover illustration: MS Hunter 231, folio 276. By permission of University of Glasgow Library, Special Collections. Library of Congress Cataloging-in-Publication Data Brill's companion to Seneca, philosopher and dramatist / edited by Gregor Damschen, Andreas Heil ; with the assistance of Mario Waida. pages cm Includes bibliographical references and index. ISBN 978-90-04-15461-2 (hardback) : alk. paper) – ISBN 978-90-04-21708-9 (e-book) 1. Seneca, Lucius Annaeus, approximately 4 B.C.-65 A.D.–Criticism and interpretation. I. Damschen, Gregor. II. Heil, Andreas, 1969PA6675.B75 2013 878'.01–dc23 2013014809

This publication has been typeset in the multilingual “Brill” typeface. With over 5,100 characters covering Latin, IPA, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities. For more information, please see www.brill.com/brill-typeface. ISSN 1872-3357 ISBN 978-90-04-15461-2 (hardback) ISBN 978-90-04-21708-9 (e-book) Copyright 2014 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Global Oriental, Hotei Publishing, IDC Publishers and Martinus Nijhoff Publishers. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. This book is printed on acid-free paper.

CONTENTS

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi PART ONE

LIFE AND LEGACY Imago suae vitae: Seneca’s Life and Career . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Thomas Habinek

The Works of Seneca the Younger and Their Dates . . . . . . . . . . . . . . . . . . . . . 33 C.W. Marshall Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Rolando Ferri Seneca and Senecae: Images of Seneca from Antiquity to Present Seneca the Philosopher . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Matthias Laarmann Seneca the Dramatist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Werner Schubert PART TWO

PHILOSOPHY Context: Seneca’s Philosophical Predecessors and Contemporaries . . . . 97 John Sellars Works De providentia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 R. Scott Smith De constantia sapientis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 R. Scott Smith

contents

De ira . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Maria Monteleone Consolatio ad Marciam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Jochen Sauer De vita beata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Fritz-Heiner Mutschler De otio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 R. Scott Smith De tranquillitate animi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Fritz-Heiner Mutschler De brevitate vitae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 R. Scott Smith Consolatio ad Polybium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Jochen Sauer Consolatio ad Helviam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 Jochen Sauer De clementia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Ermanno Malaspina Naturales quaestiones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Gareth D. Williams Epistulae morales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Aldo Setaioli De beneficiis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Mario Lentano Lost and Fragmentary Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Anna Maria Ferrero Epistulae Senecae ad Paulum et Pauli ad Senecam. . . . . . . . . . . . . . . . . . . . . . . 213 Alfons Fürst Topics Ontology and Epistemology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Mireille Armisen-Marchetti

contents

vii

Ethics I: Philosophy as Therapy, Self-Transformation, and “Lebensform” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239 Aldo Setaioli Ethics II: Action and Emotion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Margaret R. Graver Ethics III: Free Will and Autonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277 Aldo Setaioli Ethics IV: Wisdom and Virtue. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 Jula Wildberger Ethics V: Death and Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 Catharine Edwards Physics I: Body and Soul . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343 R. Scott Smith Physics II: Cosmology and Natural Philosophy . . . . . . . . . . . . . . . . . . . . . . . . . 363 Bardo Maria Gauly Physics III: Theology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379 Aldo Setaioli PART THREE

TRAGEDY Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405 Wolf-Lüder Liebermann Works Hercules furens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425 Margarethe Billerbeck Troas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435 Wilfried Stroh Phoenissae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449 Marica Frank Medea. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459 Wolf-Lüder Liebermann

viii

contents

Phaedra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 475 Roland Mayer Oedipus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483 Karlheinz Töchterle Agamemnon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493 Christoph Kugelmeier Thyestes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 501 Chiara Torre Dubious Works Hercules Oetaeus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 515 C.A.J. Littlewood Octavia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521 Rolando Ferri Topics Space and Time in Senecan Drama . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 531 Ernst A. Schmidt Vision, Sound, and Silence in the “Drama of the Word” . . . . . . . . . . . . . . . . . 547 Andreas Heil The Chorus: Seneca as Lyric Poet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561 Giancarlo Mazzoli The Rhetoric of Rationality and Irrationality . . . . . . . . . . . . . . . . . . . . . . . . . . . 575 Gottfried Mader Characters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 593 G.W.M. Harrison Themes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 615 G.W.M. Harrison Greek and Roman Elements in Senecan Tragedy . . . . . . . . . . . . . . . . . . . . . . . 639 Sander M. Goldberg Philosophical Tragedy? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 653 François-Régis Chaumartin

contents

PART FOUR

APOCOLOCYNTOSIS Apocolocyntosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 673 Renata Roncali PART FIVE

OTHER WORKS Epigrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 689 Joachim Dingel De vita patris . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695 Michael Winterbottom PART SIX

SYNTHESIS Seneca’s Language and Style . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 699 Michael von Albrecht Systematic Connections between Seneca’s Philosophical Works and Tragedies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 745 Susanna E. Fischer List of Journal Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 769 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 771 Andrea Balbo and Ermanno Malaspina Editions of Seneca’s Works (Since Haase’s Opera Omnia) . . . . . . . . . . . . . . . 861 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 865 General Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 873

PREFACE

There are already quite a few introductions to Seneca on the market; yet still lacking is a well-ordered, concise presentation which places the philosophical works and the tragedies on an equal footing and deals with them accordingly. The principal aim of “Brill’s Companion to Seneca” is to fill this gap. The subtitle “Philosopher and Dramatist” indicates and emphasizes the express intention of taking seriously both Seneca the philosopher and Seneca the poet and playwright. Therefore, the philosophical works are examined first and foremost in the light of philosophy, and this entails employing all the systematic methods which are at the disposal of specialists in philosophy. By analogy, the tragedies of Seneca are not viewed merely as valuable comparative sources for Seneca’s philosophy but recognised as valid forms of expression in their own right. The second aim is to create a valuable standard work for the purposes of international Seneca research. To this end the volume contains a comprehensive survey of each genuine, doubtful and wrongly attributed work of Seneca (“Works”). In the style of concise handbook articles each individual work is dated and briefly described with regard to content; included is a discussion of the most important philological and philosophical issues as well as an account of the reception history. In addition, the volume offers fuller presentations of the most important problem areas within the philosophic and tragic corpus of Seneca (“Topics”). It also contains a historical section in which Seneca’s life and posthumous influence is dealt with (“Life and Legacy”) as well as two studies in which the contemporary preconditions for the philosophical works and the tragedies are presented with due reference to the history of philosophy and the history of culture (“Context”). The quality of a handbook or collective work is of course dependent on the quality of the contributions. Therefore, our foremost debt of gratitude is to the authors who breathed life into the scheme. We are especially grateful to our teacher, Michael von Albrecht, who not only contributed, but encouraged us in the first place to take up this vast project. Aldo Setaioli was of great help to us in a time of crisis providing some of his contributions at short notice. We owe a lot to Andrea Balbo and Ermanno Malaspina for compiling the comprehensive bibliography. The Dresden University of Technology and the University of Lucerne provided valuable resources. Last but not least, we would like to thank the staff at Brill Publishers, particularly Michiel

xii

preface

Klein-Swormink, Irene van Rossum and Caroline van Erp, for their careful guidance as well as for their never ending patience. Finally, we have to ask pardon for the delay in publication. The contributors are not to be blamed in any case. Although some of the circumstances responsible were beyond our control, we, the editors, take full responsibility. The editors, Dresden and Lucerne, Spring 2013.

PART ONE

LIFE AND LEGACY

IMAGO SUAE VITAE: SENECA’S LIFE AND CAREER

Thomas Habinek

Introduction Something about Seneca prompts general studies. I wrote these words in 1985,1 and they remain true to this day. Biographies or “holistic” approaches to Seneca continue to appear on a regular basis and to outnumber such treatments of other Greek and Roman writers.2 The reason is not far to seek: the slipperiness of this figure, his complexity, versatility, and contrariety all challenge much-cherished ideals of stable, autonomous selfhood that lie at the heart of modern bourgeois civilization. Scholars seek to turn the fragmentary biographical details and the massive literary output of Seneca into a comprehensible life precisely because they resist such easy formulation. Indeed, Seneca is among the most aggressive proponents of the idea that the self is a social being, unlimited by the boundaries of an individual body or mind. To write a biography of Seneca is thus to risk doing violence to the historically contingent experience attested by the literary texts and historical fragments that cluster around the name “Seneca.” And yet, paradoxically, some sort of biographical framework is necessary if we are to begin to comprehend the ideas of self and society to which Seneca’s career, reputation, and writings provide access. Such a framework is best imposed not from the outside (i.e., modern ideas of the shape and structure of a life, modern conceptions of selfhood or subjectivity), but from the inside, by drawing on ancient, Stoic, and more specifically Senecan notions of the structuring principles of the world and its subcomponents.

Habinek 1985: 103. E.g., Griffin 1976, a characteristically British empiricist attempt to separate fact from fiction; Sørensen 1984, emphasizing Seneca’s humanism; Rozelaar 1976, which applies psychoanalytical categories to Seneca’s behavior; Maurach 1991 (1996), who differentiates his work from earlier biographies on the grounds that he considers the treatises of Seneca as well as the life; and Veyne 2003, which considers Seneca’s life through the lens of his Stoicism. Also relevant is Volk and Williams 2006, which purports to “see Seneca whole” by discussing him in parts. 2

thomas habinek

Two terms are crucial to such an endeavor and provide points of reference for this essay, namely ratio and societas.3 Senecan thought and Stoic philosophy, more generally, employ these terms to describe the conditions and highest aspirations of human existence. The dialogical tension between the terms and between the concepts they convey encapsulates a dynamic worldview that pays equal respect to humanity’s desire for order and control and its experience of chaos and dependence. Ratio, which translates and amplifies the Greek term logos, is best understood as ‘account’ or ‘accounting’. Ratio for the Stoics is the human apprehension of natura, or the order of the universe. (In this respect it resembles, and perhaps inspires, the evangelist John’s representation of Jesus as logos, or human understanding, of a God who personifies the principle of being.) Ratio, as a Latin term, implies the ability to count, analyze, describe, summarize, and otherwise render comprehensible and communicable to others the particulars of any aspect of experience.4 We capture the flavor of the term when we speak, for example, of “rationalizing” a law code, or providing an “account” of our behavior. While a modern notion of reason might allow for explanations that are symbolic, mathematical, or visual in form, ratio, especially when used as a Latin translation of Greek logos, inclines toward the verbal and narratival. Thus, to gather the pieces of evidence for Seneca’s birth, death, and intervening activities, to evaluate their reliability, and to organize them into some sort of narrative is to engage in activities appropriate to a historically based understanding of the term ratio and thus, I would argue, to a historically legitimate representation of Seneca’s life. Such an accounting, however, is only one part of the story, or only one version of the life of Seneca. For Seneca and the Stoics, humans are both rational and social beings. Their exercise of reason and their sociability

3 Uses of the terms are too numerous to recount here. Representative passages include Sen. epist. 79.9f., in which ratio is presented as the best and defining characteristic of human beings; Sen. epist. 121.3 f., where we are taught that humans and gods, as opposed to animals and plants, have reason; Sen. clem. 3.1.2, where Stoics are said to believe that “man is a social animal born for the common good” (hominem sociale animal communi bono genitum); Sen. benef. 4.18.1, where we are taught that ratio and societas are the god-given characteristics of human beings; Sen. dial. 3 (= de ira 1).15.2, dial. 4 (= de ira 2).31.7, clem. 1.12.3, 1.14.3, where the relationship between humans and society is likened to that between organs and body. In the light of the latter group of passages (on social interactions) it is hard to understand Veyne’s assertion that Seneca and the Stoics “did not perceive anything called society intervening between natural law and the individual” (Veyne 2003: 141). Seneca’s emphasis onsocietas is a manifestation of his Romanness as well as of his commitment to Stoic teachings. On societas and communitas in Roman law and thought, see Daube 1938. 4 See, in particular, Moatti 1997.

imago suae vitae: seneca’s life and career

toward one another differentiate them from other species and mark them for a higher and more god-like destiny. 5 Even animals that we might regard as social in nature are understood by the Stoics to be lacking in the modes of communication and intersubjective representation that ground human experience and implicate the self in social interactions.6 As Seneca puts it in his essay On Benefits, “God has given mankind two characteristics that transform it from vulnerable to supremely strong, namely reason and sociability […] it is sociability that grants mankind dominion over all other creatures” (duas deus res dedit, quae illum obnoxium validissimum facerent, rationem et societatem […] societas illi dominium omnium animalium dedit : benef. 4.18.1). The context of the quotation makes clear that Seneca is speaking not just about the relative power of human beings over other species, but also about the merit and value of human social existence. To understand a human being, at least from the perspective of a Roman Stoic like Seneca, one must situate him or her in a social context. The point may seem obvious, but in fact it flies in the face of most modern philosophical and scientific systems that take for granted an isolated “I,” who chooses the extent to which he or she will involve himself or herself in an external world. For the Stoics, as for a new generation of neuroscientists, the self is to be understood as a node in the network of intersubjective relations.7 Without others, there is no human self. The very characteristics that make us human (language, thought, ritual, etc.) are all social constructs. Roman Stoics, building on what seem to be folk understandings of the self, press this insight further, drawing the implication

5 On human sociability, human difference from animals, and human association with gods, see Cic. fin. 3.62–68 (= Long and Sedley 57F). Especially striking is the reference there to Chrysippus’s claim that “everything else was created for the sake of men and gods, but these for the sake of community (communitas) and society (societas).” Origenprinc. 3.1.2f. (= Long and Sedley 53A) explains that while humans share sensation and impulse with animals, animals lack reason. 6 Sen. epist. 121.6–15 (= Long and Sedley 57A); S. Emp. adv. math. 8.275f. (= Long and Sedley 53T). 7 For a biologically based understanding of the inevitability of human intersubjectivity and sociability, see, for example, Donald 2001. For the early (i.e., pre-Senecan) Stoic view, see Reesor 1989, esp. p. 8, where she summarizes Stoic thought thus: “It is inconceivable, therefore, that a man’s individuality can be realized in any other society, with any other group of individuals, or in any other period than that in which he is actually living.” The grounding of the self in society has received less attention by recent commentators than it should due to a (misplaced, in my view) eagerness to understand the Stoic self as predecessor of the modern autonomous self rather than to situate it in the broader system of Stoic thought (including physics) and the historical context of its production.

thomas habinek

that the boundaries of the self can expand and contract over a lifetime and that at any given moment some selves may be bigger than others, indeed may encompass others, such as family, dependents, and slaves.8 If, then, we are to comprehend Seneca’s life in terms that are appropriate to his era, we must understand it as a node in the network of social exchange. It is only through encounters and interactions with other human beings and participation in the broader social institutions that structure those interactions—such as a changing political culture; economic expansion; reliance on slaves; systems of patronage, friendship, and dependence; patterns of love and desire; and so forth—that a life of Seneca begins to take shape. Or, to put it in a more positive vein, the dearth of reliable, unambiguous evidence pertaining to the activities and intentions of one Lucius Annaeus Seneca does not prevent and may even facilitate the recovery and representation of “his” “life”. Biographical Account Seneca was born in Córdoba, capital of the Roman province of Baetica in the south of Spain, probably between 4 and 1 bc. The locale is attested by references in the writings of his father and his younger contemporary Martial,9 while the date is a plausible inference from his own allusions to childhood experiences10 and from Nero’s reference to the prematurity of his request to retire in ad62, at least as reported by Tacitus.11 Seneca’s father, whose surviving treatises on the history and practice of declamation provide strategically placed details of family background and connections, was of equestrian rank, and although of Spanish birth, seems to have spent a good

8 Linguistic and legal assumptions make the slave part of its master’s person: Reay 2003 and 2005. 9 Sen. suas. 2.18 refers to Statorius Victor, whom other sources identify as Córdoban (see RE 2.3.2230–2232), as municeps meus (‘my fellow townsman’); at suas. 6.27 he refers to a probable Córdoban as municeps nostrum (‘a townsman of ours’); Mart. 1.61.7f. identifies Seneca the Elder, Seneca the Younger, and the poet Lucan as Córdobans. Seneca the Younger may refer to his birthplace in a fragment from de matrimonio (Haase 88) and again in epigram 19 (= Anth. Lat. 405 Shackleton Bailey); but the attribution of these passages to Seneca has been challenged. Griffin 1972: 17 notes Seneca the Younger’s “general reticence about himself”; Sørensen 1984: 69 observes that “[Seneca] never refers to himself as Spaniard or shows any trace of Spanish national feeling.” 10 E.g., dial. 9 (= tranq.).17.7, nat. 1.1.3, epist. 108.22. For analysis, see Griffin 1976: 34–36, Abel 1981b, and Griffin 1984: 14–16. 11 Tac. ann. 14.56.1, discussed by Griffin 1976: 35 f.

imago suae vitae: seneca’s life and career

portion of his life in Rome.12 Seneca’s mother, Helvia, was from a prominent Baetican family, as indicated by the nomenclature on a dedicatory inscription from the town of Urgavo (mod. Arjona) about eighty kilometers east of Córdoba.13 Scholars have often assumed that the Annaei were Hispanienses, i.e., Spanish residents of Italian descent, rather than Hispani, of Spanish descent, based on the occurrence of similar names in Italy.14 Paul Veyne has recently suggested, however, based on demographic patterns, that it is more probable that Seneca’s roots were in fact native Iberian rather than transplanted Italian.15 As a child, Seneca was taken to Rome by his mother’s stepsister,16 a woman whose marriage to the eventual prefect of Egypt, C. Galerius,17 enabled Seneca’s youthful convalescence in that Roman province and his later entry into Roman politics.18 The members of Seneca’s family were, in effect, provincials on the make, eager to exploit the opportunities for political and social mobility provided by the Augustan settlement. In addition to his uncle Galerius, the prefect of Egypt, his father’s friend (and eventual adoptive father of Seneca’s older brother), the senator L. Iunius Gallio, may be supposed to have granted or extended access to high levels of the Roman political and economic elite.19 Seneca the Elder’s ambitions for his sons, as advertised in the preface to book 2 of the Controversiae,20 were to a large extent realized: the older brother, Annaeus Novatus, later L. Iunius Gallio Annaeanus, became a senator, suffect consul, and proconsul of Achaia,21

Griffin 1972 on his life. CIL II.2115, Vassileiou 1973, Gleason 1974, and Pflaum 1977, n. 438. 14 E.g., Syme 1964, appendix 80, followed by Griffin 1972: 2. 15 Veyne 2003: 1 f. with notes. 16 Sen. dial. 12 (= cons. Helv.).19.2. 17 For Galerius see RE 19 (1910): 598. 18 Sen. dial. 12 (= cons. Helv.).19.4. Seneca maintained connections with Egypt, including ownership of an estate there, into his later years: see Turcan 1967, Rostovtzeff 1998: 2.671, and Browne 1968: 17–24. To what extent Seneca’s youthful sojourn in Egypt shaped his later literary and political activities is an open question. 19 On L. Iunius Gallio see PIR 1.756, RE 19 (1910): 1035–1039, and Griffin 1976: 45, 48. 20 Sen. contr. 2 pr. 4. While scholars often read the Controversiae and Suasoriae of Seneca the Elder as the relatively insignificant reminiscences of a very old man, Bloomer 1997a is surely right to see the author as strategically placing his sons in a long tradition of Spanish learning and achievement. That tradition is, in turn, subtly represented as analogous to the traditions of the Roman elite: thus Seneca the Elder, in the course of writing to and instructing his sons, alludes to Cato the Elder’s instruction of his son (contr. 1 pr. 9f.) and to Asinius Pollio’s role in the education of his grandson (contr. 4 pr. 2–4). 21 Dittenberger 1915: 2.801, Acts 18: 12, and RE 1.2 (1900): 2236–2239. 13

thomas habinek

while his younger brother, Annaeus Mela, became an imperial procurator 22 and father of the poet Lucan (Marcus Annaeus Lucanus). As a Roman citizen of high rank, Seneca would have been expected to undergo training with a grammaticus, or teacher of literature; a rhetorician, or teacher of public speaking; and perhaps a philosopher.23 Seneca refers to his grammaticus once,24 and his easy familiarity with a wide range of Roman authors, as evinced in all of his writing, attests to the impact, direct or indirect, of early literary studies. As teacher of rhetoric and philosophy he names Papirius Fabianus, a man he (and his father) admired as both polymath and Latin stylist. 25 Unlike many young Romans, whose rhetorical training led almost immediately to public life, including pleading before tribunals, government service, and possible elected office, Seneca seems to have held back, for in a text written no earlier than ad 37 his father describes him and his older brother as still “preparing for the forum and public office.”26 Yet within a few years Seneca was sufficiently renowned for his eloquence to have earned the notice—and antipathy—of the emperor Gaius (also known as Caligula), who described his style of speaking as “sand without lime” and “drunken revelry”27 and, according to another ancient anecdote, intended to have him killed until an unnamed mistress observed that he was likely to die soon anyway of lung disease.28 Whether or not the latter story obscures Seneca’s involvement with the conspiracy of Lepidus and Gaetulicus against Caligula, perhaps via a connection with the emperor’s sisters as some have proposed, it is certainly the case that Seneca’s evolving career

Tac. ann. 16.17.1–6, and RE 1.2 (1900): 2236. On the aims and impact of such training see Habinek 2005a: 60–78. 24 Sen. epist. 58.5. 25 Sen. contr. 2 pr. 1–4; Sen. epist. 40.12–14. In the latter passage, as well as in epist. 58.6 and epist. 100.1–9 Seneca discusses him in relationship to Cicero. Other references to Fabianus include Sen. epist. 52.11 and 100.1–9. 26 Sen. contr. 2 pr. 4. Griffin 1976: 44 overinterprets the father’s language to suggest that it provides a precise date for Seneca’s quaestorship. 27 The Latin phrases are harena sine calce and comissationes meras (Suet. Cal. 53). The latter contains a pun on the adjective merus, -a, -um which means ‘pure’ or ‘unadulterated’ but often refers to wine that has not been diluted with water. Seneca’s rhetoric, according to Gaius, is like the homeward procession that follows a party where everyone has drunk too much strong wine. 28 Dio 59.19. Griffin 1976: 53–56 expresses undue skepticism for the hypotheses of Stewart 1953 and Lana 1955: 106–110, 115 that Seneca earned Gaius’s wrath because of his association with Gaetulicus’s conspiracy against him. Lana’s emphasis on the possible involvement of Gaius’s sisters and, in particular, of Seneca’s association with them, fits the better-documented pattern of Seneca’s later association with Livilla and Agrippina. 23

imago suae vitae: seneca’s life and career

illustrates the prospects and pitfalls of both senatorial and court politics.29 In ad 41, at the urging of Messalina, wife of the new emperor Claudius, Seneca was relegated to the isle of Corsica on a charge of adultery with Gaius’s sister Livilla,30 eventually to be recalled by another sister of Gaius, namely Agrippina, the new wife of Claudius and mother of soon-to-be emperor Nero.31 Of Seneca’s life in exile on Corsica we know only what he tells us in his various writings dating to that period, namely that he spent a great deal of time on his studies.32 Like Cicero and Ovid before him, Seneca used his literary productivity as a way to maintain a presence in Rome during forced separation from it. So much is clear from both his Consolation to his mother Helvia (interestingly enough, on her grief over the loss of him) and his Consolation to Polybius, Claudius’s powerful freedman, whose brother had recently died. The former work can be understood as reinforcing Seneca’s ties with his family and their associates, the latter as an honorific offering to an important figure in the imperial household in the hope of a return benefit, presumably in the form of recall from exile. Such was in time obtained not through Polybius, but thanks to the removal of Messalina, who was eventually executed for adultery and, at least in Tacitus’s account, general recklessness, and the ascendancy of Agrippina, who, although the niece of Claudius, was married to him with the fulsome approval of the Senate and people.33 According to Tacitus and others, Seneca was recalled by Agrippina to serve as tutor of her son Nero,34 whom she was grooming as heir to the throne in place of Messalina’s son Britannicus. We may reasonably surmise that Agrippina would not have effected this arrangement without at least some prior familiarity with Seneca’s loyalty and trustworthiness. Reliance on such private connections need not be seen as contradictory to the publicly announced privileging of Seneca due to the “renown of his intellectual

29 On the conspiracy see Simpson 1980, Barrett 1989, chapter 6, and Rowe 2002: 168f. Seneca’s friend and later addressee, C. Lucilius Iunior, was also caught up in the paranoia surrounding the conspiracy, at least as Seneca tells it (nat. 4 pr. 15f.). 30 Relegation, which is the precise term for Seneca’s removal (schol. Iuv. 5.109), was generally a milder punishment than exile and may have allowed Seneca to retain a significant portion of his wealth. On the various types of punitive removal from Rome under the early Empire, see Brunt 1961: esp. 202–204. On Messalina’s role see Dio 60.8.6. 31 Tac. ann. 12.8.2. 32 Sen. dial. 12 (= cons. Helv.).1.2 and 7.9. 33 Tac. ann. 12.1.1–7.3 on the betrothal and marriage of Claudius and Agrippina. 34 He is called magister or ‘teacher’ in Tac. ann. 12.8.2 and praeceptor in 15.62.2.

thomas habinek

pursuits” (claritudo studiorum), 35 a motif that recurs in ancient accounts of Seneca’s lifetime and in later reflections on his cultural significance.36 With his return from exile, Seneca also assumes a leading role in the ancient historiographical tradition, which is centered on the lives of emperors and their relationships to courtiers, aristocrats, and military commanders. In recognition of his special relationship to Nero, he comes to be identified as ‘friend of the prince’ (amicus principis),37 a role that compensated for its lack of institutional definition with an array of distinguished precedents. Hellenistic kings had had “friends” who were, in effect, advisers and public relations specialists,38 and Roman figures like Agrippa and Maecenas had occupied much the same position, even when they also held legally recognized posts, from early in the Augustan regime onward.39 Agrippina would have publicized Seneca’s relationship to her twelve-year-old son Nero as part of her project of grooming the latter to succeed Claudius. Indeed, associating Nero’s public pronouncements with Seneca was a pattern that persisted throughout Nero’s reign, even after Agrippina’s death: we hear of Seneca’s composition of Nero’s funeral oration for his predecessor Claudius in ad54;40 the surviving Senecan treatise, De Clementia (Concerning Mercy), dated to ad 55 or 56, seems designed in part to reassure the dominant class that Nero’s murder of Britannicus is the end, not the beginning of bloodshed;41 and, with tragic irony, Seneca was compelled to compose a letter to the Senate justifying the murder of Agrippina, the chief booster of his meteoric ascent.42

35 For the expression claritudo studiorum, see Tac. ann. 12.8.2. For its meaning and significance, see Habinek 2000. 36 Evidence gathered and discussed in Habinek 2000. 37 Tac. ann. 14.54.3 has Seneca refer to himself as among Nero’s seniores amici. Seneca also held the official position of suffect consul for six months in 55 or 56ad, which in turn entitled him to be regarded as one of the leading members of the Senate. 38 E.g., Savalli-Lestrade 1998. Tacitus has Seneca himself cite the precedents of Agrippa and Maecenas for his relationship to Nero: see ann. 14.54.3. Seneca’s description of them in ben. 6.33.3f. suggests that in his view they did not always fulfill the role of “friend” appropriately: they would have dissimulated in the case of Augustus’s daughter’s scandalous behavior, rather than speaking candidly, as was the historical expectation of friends. 39 Tacitus has Nero invoke L. Vitellius, in his relationship to Claudius, as precedent for Seneca’s position: ann. 14.56.1. 40 Tac. ann. 13.3.1–3 discusses the funeral oration; ann. 13.11.2 refers to Seneca’s “frequent” composition of speeches delivered by Nero. 41 On the context and aims of Seneca’s treatise de clementia see, among others, Griffin 1976: 133–171, and Lana 2001a (with reference to his extensive earlier work on the topic). 42 Tac. ann. 14.10.3–11.3 discusses the letter in some detail.

imago suae vitae: seneca’s life and career

Historians both ancient and modern disagree about the extent and nature of Seneca’s contribution to imperial governance under Nero. The discussion is complicated by the fact that ancient writers on politics tended to think concretely in terms of personal connections and reactive decisions rather than abstractly in terms of broad goals, agendas, and strategies. Unfortunately, this limitation in the representation of politics is too easily taken to correspond to a limitation in the actual performance and intentions of political actors. Thus the failure of the ancient sources to say that Nero and Seneca had a broad strategic understanding of the needs of the Empire does not mean they lacked one. Indeed, at least two themes recur throughout the Senecan ascendancy, one being a return to civilian governance after the Claudian swerve toward military autocracy, the other being a continuation of the Claudian interest in economic infrastructure and the rationalized use of urban space. With respect to the first strategic goal, we may note Seneca’s partnering with Burrus, the newly appointed prefect of the Praetorian Guard, whose prior experience was chiefly financial rather than military in nature.43 Through such an arrangement the Neronian regime sought (ultimately unsuccessfully) to integrate the praetorians into civilian lines of authority and prevent their persistence as, in effect, kingmakers—a role they had played in the elevation of both Gaius and Claudius. Moreover, Nero’s deep engagement in diplomacy and his failure to take on a distinct military role for himself stand in contrast to Claudius’s muchderided leadership of an expedition to Britain.44 Indeed, it is striking to note the extent to which Tacitus’s account of Nero’s reign is replete with diplomatic missions, meetings with ambassadors, etc. in contrast to the highly militarized narratives of earlier emperors. The remark Tacitus attributes to unnamed members of the population, namely that “more was accomplished

43 Veyne 2003: 20 following Syme 1963: 2.591. Claudius had shown special favor toward the praetorians: Dio 59.2.3, 60.12.2, Suet. Claud. 10.4, and Griffin 1984: 203. 44 Suet. Claud. 17.1–3 reports that Claudius undertook this ‘modest’ (modica) enterprise for the specific purpose of being able to celebrate a triumph. He refers to the same expedition in strikingly different terms in Vesp. 4.1f. Dio 60.23.1–6 contrasts the shortness of time Claudius spent in Britain with the elaborate nature of the celebrations upon his return to Rome. It is in this context that Dio also notes Claudius’s usurpation for himself and his military of initiatives in foreign policy that had traditionally been the prerogative of the Senate and people. Sen. apocol. 12 presents a chorus as singing a mock-heroic dirge celebrating Claudius’s military achievements and his ability to discern the truth in trials without even hearing evidence. In Tac. ann. 14.55.3 Nero contrasts the militarized youth of Augustus with Seneca’s and his own lack of participation in military activities. Griffin 1984: 231 notes that “whereas Gaius and Claudius each felt impelled to take the field personally after two years in office,” Nero waited at least nine, possibly eleven years.

thomas habinek

by auspices and counsel than by weapons and brawn”45 seems to summarize foreign policy during the ascendancy of Seneca and Burrus. Even the talented general Cn. Domitius Corbulo, who held command in the East during much of Nero’s reign, achieved as much by shrewd manipulation of his enemies’ fears and interests as by actual combat.46 As for economic policy and other plans for infrastructure, we must read between the lines of the scandal-driven narratives of the ancient historians to discern the role of the imperial household in managing affairs of state. Yes, Seneca was phenomenally wealthy, but that wealth manifested itself (and was no doubt in part produced by) up-to-date agrarian methods and a more or less new mode of international banking, 47 not to mention imperial bequests. During the period of Seneca’s close association with Nero financial reforms improved the health of the imperial treasury,48 thereby providing the wherewithal for the public works projects that characterized Nero’s regime both before and after the infamous fire of July 64.49 A recent careful reading of references to city life throughout Seneca’s literary career has identified changes in the city of Rome as visualized by 45

Tac. ann. 13.6.4. In letter 73, a treatise clearly intended for the eyes or ears of Nero, Seneca gives thanks for a leader (gubernator) who removes the necessity of bearing arms, serving on night watches, guarding the city walls, etc. (epist. 73.9). Also relevant is Seneca’s promulgation of an ideal of securitas—freedom from anxiety—as political and personal goal: see the excellent discussion in Lana 2001b. 46 For background on Corbulo and a discussion of his career see Syme 1970; on battles and diplomatic activities in and about Armenia see Tac. ann. 13.34.2–13.42.3; Griffin 1976: 223–229, 462–466. 47 Following Veyne 2003: 9–15, this seems the best way to understand Dio’s description of Seneca’s loans to British chieftains (Dio 62.2.2). Tacitus may allude to this type of activity as well when he has Seneca mention his “extensive involvement in lending at interest” (tam lato faenore: ann. 14.53.5) as a prerogative he owes to Nero—i.e., lending that “extends” beyond the close circle of Roman friends. Presumably Seneca’s fortune was made available to finance the “development” of Britain, a process that had to be revoked when it became more important to fund other endeavors. Griffin 1976: 246, with reasoning characteristic of that of many modern historians, assigns Seneca’s withdrawal of the loans from Britain to “prudence not malice.” But why should either character trait be relevant? Seneca was acting as the government’s banker and did what was thought to be best for the government. See also Levick 2003. On the relationship between public and private property, very different in antiquity than today, see Seneca’s concise remarks in benef. 7.3.2 f. 48 Tac. ann. 13.50.1–51.2. Cizek 1984: 135–139 suggests that the Senate’s ultimate refusal to follow through on Nero’s desire to move from indirect to direct taxation (and thus significantly reduce the role of intermediary corporations run by rich equestrians and by or for senatorial backers) was the turning point in Nero’s relations with that institution. The use of Seneca’s private fortune as a basis for investment in Britain can plausibly be interpreted as one aspect of the Neronian regime’s strategy of centralizing finances in the imperial household. 49 It should be noted, however, that by the time Galba succeeded Nero in ad 69, the treasury was bankrupt.

imago suae vitae: seneca’s life and career

Senecan prose: from crooked narrow streets to sweeping boulevards, largescale baths, and theaters large enough to hold the entire citizenry.50 Seneca’s moralizing tone rejects the distractions that both forms of urbanization create for the elite soul in search of serenity, but the changing background conditions taken for granted by the language of the ethical treatises can only have been created by impetus and direction from the highest levels of government, in which Seneca was a key participant.51 Yet the historiographical tradition, fixated as it was and is on the tense relations between emperor and elites, prefers stories of court intrigue to debates over policy, investment, and social goals. In the context of such stories, Seneca emerges as an exceptionally important player, one who worked with Burrus and others to strike a balance between Nero and Agrippina, to reassure the upper classes of Nero’s good intentions, to settle old scores against remnants of Claudius’s rule, and, finally, to deny legitimacy to Nero once he began to behave in a way that offended too large a segment of elite and popular opinion. Two episodes can stand for many. According to Tacitus, when Nero began to distance himself from Agrippina due to her overbearing involvement in personal and public affairs, she responded by making incestuous advances toward him.52 Such behavior might have shocked even jaded Romans; more importantly, if successful in uniting mother and son it would have reduced or eliminated the influence of other counselors, such as Seneca. And so the latter recruited the freedwoman Acte, of whom the princeps was already enamored,53 to inform Nero of the dangers of his behavior, especially the risk of losing support among the troops. When the hostility Nero felt toward Agrippina turned murderous, Seneca and Burrus, having lost the ability to triangulate with the emperor’s mother, experienced a sharp decline in influence. Burrus’s death in ad 62 further weakened Seneca’s position, as did Nero’s maturation (he was now 25 years old) and growing comfort with the exercise of power. Seneca sought to remove himself from the heights of power in a conversation with Nero presented by Tacitus as a kind of tragic agon, the only such episode in the surviving Annals.54 Nero neither released Seneca nor

Sommella 2000. It may also be worth noting in this context that as procurator, Seneca’s younger brother Mela had responsibility for revitalization of the capital city: see Cizek 1982: 104. 52 Tac. ann. 14.2.2. 53 Tac. ann. 13.12.1. 54 Tac. ann. 14.53.1–56.3. 51

thomas habinek

solidified his position. Seneca retreated to the countryside, refused to greet clients and petitioners, and set about re-positioning himself, as he did during the period of his relegation, through literary activity. In these final years of his life he composed a massive work on Natural Questions (interestingly, a return to a type of inquiry that had occupied him during his previous removal from the seat of power)55 and a series of quasi-private letters addressed to a friend and associate, C. Lucilius, a younger contemporary who held various equestrian procuratorships under Claudius and Nero.56 Just as the Consolations written from Corsica sought to manage relations with friends and foes alike, so too the Letters to Lucilius or Epistulae Morales, as they are also known, play a double game. On the one hand, they seek to make clear to the princeps that his semi-retired minister is engaged in philosophical, not conspiratorial activity.57 The composition of the letters, as if reflecting the actual day-to-day experience of their writer, serves as documentation of political withdrawal. Yet at the same time, the letters speak of preparation for death, of the possibility of undergoing savage violence at the instigation of “one more powerful” (potentioris: epist. 14.3). The sage, according to Seneca, shuns a power out to get him (nocituram potentiam: epist. 14.8) while taking care that the shunning not be obvious. “An important component of security,” he writes, “is not to admit one is seeking it” (pars enim securitatis et in hoc est, non ex professo eam petere: epist. 14.9). In effect, Seneca both displays his retirement and explains why he is doing so in the same collection of letters.58 He speaks simultaneously to Nero (or at least to those who would denounce him to Nero) and to others who are potentially sympathetic to his plight. Ultimately, Seneca addresses his readership as he does himself, arguing that when the time comes it is always easy to take one’s own life (epist. 70). A letter that exhorts the neophyte Stoic to accept responsibility for the conditions and mode of living (by acknowledging the ease of dying) in effect exhorts its own writer to face his dire predicament with serenity.

Sen. dial. 12 (= cons. Helv.).20.1 f. RE 13.2 (1927): 1645, C. Lucilius Iunior. Lucilius is also the addressee of De providentia (Sen. dial. 1) and Naturales quaestiones. In epist. 34.2 Seneca declares to Lucilius: “I attach you to myself; you are my project” (adsero te mihi; meum opus es). Such language, combined with the fact that Lucilius is the diminutive of Lucius (Seneca’s own praenomen), as A. Corbeill reminds me, suggests a different conceptualization of the boundary between self and other than the one that prevails in modern societies. 57 This is the interpretation of Cizek 1972: 155 followed by Veyne 2003: 160–164. 58 Cf. the similar assessment of Veyne 2003: 157–167. 56

imago suae vitae: seneca’s life and career

This dual enterprise—steering clear of the emperor while making it obvious how and why he was doing so—served Seneca well right up to his death. Implicated in the Pisonian Conspiracy against Nero in April of ad 65, Seneca (along with his brothers and nephew) was, in effect, invited to put an end to his own life. Even as he denied complicity in the plot against Nero, we learn (from Tacitus and his pro-Senecan source59) that some within the conspiracy were plotting to kill not only Nero, but also Calpurnius Piso, and place Seneca himself upon the throne.60 The plot within a plot caught the imagination of later generations of Romans, who saw in Seneca’s distancing of himself from Nero the possibility of an emperor selected on the basis of cultural competence rather than birth or military achievement.61 So compelling is Seneca’s renown that it is the military men within the conspiracy who are said to favor his elevation.62 While some dismiss the notion of Seneca as emperor as “appealing historical dream,” it is not clear that Nero saw the possibility as so fantastic.63 For what other reason did he insist on Seneca’s death? The latter had already ceded his immense wealth to the emperor,64 and, indeed, had graciously thanked the emperor—in person, according to Tacitus,65 and certainly in published form, as letter 73 makes clear—for making it possible for him and other philosophers to pursue their studies unimpeded.66 Griffin’s answer—“to remove the last goad to [Nero’s] flagging conscience”—seems a bit fanciful and, if nothing else, reveals her acquiescence in the one-sided view of Nero as irrational, guiltridden monster presented by senatorial historiography.67 The most likely reason is that Nero regarded Seneca as he, or his mother, had previously

59 For Tacitus’s reliance on the martyrology of Seneca’s friend Fabius Rusticus, see Townend 1964 and Champlin 2003: 40–50. 60 Tac. ann. 15.65.1 identifies the military tribune Subrius Flavus and unnamed centurions as those rumored to be planning to make Seneca emperor. 61 Habinek 2000, esp. 278–284. 62 Indeed, even Dio thinks that L. Faenius Rufus, co-prefect of the Praetorian Guard (together with Ofonius Tigellinus, see Tac. ann. 14.51.2f.) was Seneca’s chief collaborator in the plot to place Seneca on the throne. 63 The phrase is that of Veyne 2003: 168, who makes the equally puzzling claim that “Nero was […] too deprived of political sense to kill in a calculated manner” (2003: 167). 64 Dio 62.25.3. 65 Tac. ann. 14.53.1–3. 66 Veyne 2003: 160 endorses Cizek’s understanding of this letter as indicating, despite its fawning tone, that Seneca has “‘no intention of becoming involved with the newly adopted policies of Nero’s regime’ ” (Cizek 1972: 155). 67 Griffin 1976: 367. In her book on Nero, Griffin describes him as “young, vain and insecure” while also considering “difficulties inherent in the political system of the Principate” (1984: 185).

thomas habinek

regarded Britannicus, Iunius Silanus, Rubellius Plautus, and even Octavia, and as he would come to regard Corbulo—namely, as rivals or rallying points for potential opposition.68 Two plausible but strikingly different accounts of Seneca’s death survive. The early third-century historian Dio, a Roman senator writing in Greek, tells us nothing of the order from Nero, but focuses on four key points: Seneca wanted his wife Paulina to die with him, he attended to last-minute literary affairs, his death was hastened by soldiers, and he died in spite of his earlier attempts to appease Nero.69 Tacitus, writing about fifty years after the events, but generally understood to rely on the account of Seneca’s contemporary and friend, Fabius Rusticus,70 creates a more elaborate and detailed scenario in which Seneca seeks to revise his will for the benefit of his friends, but is forbidden; dictates final words to his scribes; argues against, but ultimately accepts his wife’s own determination to join him in suicide; has his veins cut, but also imbibes (to no avail) the hemlock he has stored against such an eventuality; and finally suffocates to death, having been placed in a warm bath—but not before remarking that the splash of water from the tub was a libation to Jupiter the Liberator.71 Dio’s version emphasizes the relentlessness of Nero, whose antipathy grinds down even the most pragmatic and selfinterested of fellow Romans, while Tacitus makes of Seneca a martyr for political freedom and hero of the Stoic process of self-transformation.72 Despite the best efforts of scholars, attempts to separate fact from fiction in the narratives are probably futile: as recent scholarship has shown, even the relatively recent and significantly better-attested death of the American president Abraham Lincoln prompted conflicting eye-witness accounts of the assassin’s words, the size of the room in which Lincoln died, and the language (and ideological substance) of the death announcement by Lincoln’s friend and Secretary of War Edwin Stanton.73 Whether we attribute such confusion to the high emotions of the occasion or to the calculated attempts of survivors to lay claim to the legacy of the deceased hero, or both, the truth of the event resides not in unverifiable “facts” but in the constellation of issues and problems raised by conflicting claims. 68 To her credit, in a later study (1984: 189–196) Griffin does consider the challenge posed to Nero by the proliferation of potential rivals, especially the increasing number of descendants. 69 Dio 2.25.1–3. 70 See note 59 supra. 71 Tac. ann. 15.62.1–64.4. 72 The best discussion is that of Veyne 2003: 157–172. 73 A. Gopnik, “Angels and Ages: Lincoln’s Language and its Legacy,” The New Yorker, May 28, 2007 (www.newyorker.com accessed on July 16, 2007).

imago suae vitae: seneca’s life and career

In Lincoln’s case, the debate is to a large extent about the secular versus religious significance of his legacy (e.g., did Stanton say “he belongs to the ages” or “he belongs to the angels”?), and something similar might be said of the contrast between the accounts of Dio and Tacitus. Neither writer finds it necessary to comment on the underlying pragmatism of the choice to commit suicide (they, like their readers, would have understood that in so doing Seneca protected the terms of his will),74 but in one case that pragmatism is all there is, in the other it is subordinated to the more idealistic aim of exemplifying (as Tacitus had said of his own father-in-law75) how to be a great man under a bad emperor.76 The Social Self The variant death scenes presented by Dio and Tacitus illustrate the ultimate indistinguishability of biography from social concerns and constraints— exactly as Seneca’s own Stoicism would have predicted. Instead of amassing indecisive arguments for or against the “truth” of a given version,77 we would do better to probe more deeply into the issues highlighted by ancient accounts of Seneca’s death as well as of his life. While such an inquiry can move in an almost unlimited number of directions, six aspects of Seneca’s relationship to Roman society will concern us here: the use of writing to extend the self beyond the boundaries of time and space; changing gender roles as illustrated by his relationship with his wife; the psychological and ideological effects of continuing reliance on slaves in all aspects of life; the tension between competing models of economic exchange made manifest in the stories and accusations concerning Seneca’s wealth; the deployment of Stoic philosophy as an authoritative generalizing discourse at the expense of the proliferating efforts of specialists; and the reorientation of Roman life from past to future as a distinguishing characteristic of the late Julio-Claudian

Veyne 2003: 166 following Tac. ann. 6.29. [P]osse etiam sub malis principibus magnos viros esse (Tac. Agr. 42). 76 I am not persuaded by the view that Tacitus undercuts the authority of Seneca’s death scene by describing the suicide of Petronius, Nero’s arbiter elegantiae, who seems to have staged his death as, at least in part, a parody of Seneca’s (Tac. ann. 16.18.1–19.3). Different need not mean better or worse; and in any event, Petronius’s death was as appropriate for an Epicurean under duress as Seneca’s was for a Stoic facing comparable pressure: see Veyne 2003: 172. 77 See, for example, Griffin 1976: 371, who says there is “sufficient justification for adopting [Tacitus’s] version of the facts.” 75

thomas habinek

period. By examining each of these topics, however briefly, we can begin to understand the significance of Seneca’s life for his contemporaries and for ourselves. Writing To start with the dual death scenes recounted above, we notice that the importance of writing and the problem of Paulina (will she die with Seneca or not?) figure in the narratives of both Dio and Tacitus.78 What might these features signify more generally? Tacitus tells us not only that Seneca composed death-bed remarks that survived to his own day, but that he referred, in effect, to his large corpus of philosophical treatises in seeking to regulate the emotional response of friends who witnessed his death. As Tacitus puts it, “partly by conversation and partly in the more intense role of a reprimander (coercentis), he recalled those present from their tears to fortitude, asking repeatedly where were the precepts of their wisdom (praecepta sapientiae)? Where, after contemplating it for so many years, was that reasoning (ratio) in the face of looming adversity?” (Tac. ann. 15.62.2).79 The diction and phrasing of the Latin original resonate with the language of Seneca’s philosophical treatises, works in which he cajoles and upbraids, reasons, and issues precepts, all with the goal of making the reader/listener securus, i.e., free from anxiety concerning the non-necessities of life and the irresistible dictates of fate.80 Through his writings Seneca makes of Stoicism a strategy for re-ordering oneself in relationship to fate, for living the realization that all are integrated with all through the dynamic, corporeal life force that permeates the universe.81 In his death throes Seneca refers, almost as an afterthought, to Jupiter the Liberator, a symbol of the liberation achieved by

78 Dio 62.25.1 says that Seneca wanted Paulina to die with him; Tac. ann. 15.63.1f. instead has Seneca acquiesce in Paulina’s own wish to join him. Dio 62.25.2 says that Seneca revised one book and deposited others with friends out of fear that Nero would destroy them, while Tac. ann. 15.63.3 refers explicitly to Seneca’s final dictations to his scribes and implicitly (passage cited in text) to his whole body of philosophical writing and teaching. 79 Simul lacrimas eorum, modo sermone, modo intentior, in modum coercentis, ad firmitudinem reuocat, rogitans ubi praecepta sapientiae, ubi tot per annos meditata ratio aduersum imminentia; translation in text is from Woodman 2004: 335. 80 On the dynamic relationship between reason and exhortation in Seneca’s writing, see Habinek 1989: 238–254 and Inwood 1995. On securitas in Seneca’s writings and Nero’s political program, see Lana 2001b. 81 The force known as pneuma in Greek, spiritus in Latin. For its corporeality, and the implications thereof, see Sambursky 1959, Reesor 1989: 2–21, and Rosenmeyer 1989.

imago suae vitae: seneca’s life and career

the man who truly does not care whether he lives or dies, as long as he behaves virtuously.82 Tacitus dramatizes the lessons of Seneca’s philosophy, while at the same time pointing the reader in the direction of his writings. Even Dio’s account intimates that Seneca regards his books as a way of extending his life beyond the bounds of death: why else his final care to preserve them? Seneca’s actions thus give insight into a widespread Roman practice that is based on a belief, not a metaphor: i.e., that the self really does extend beyond the limits of biological life, and that such extension may come in the form of reanimation through reading of an author’s works. No wonder terms like ‘Cicero’ and ‘Seneca’ refer interchangeably to the man and his writings. Gender As for Paulina, her participation in Seneca’s final act of suicide speaks to the growing, but conflicted, acceptance of women as independent actors under the early Principate. Liberated from their male relatives’ exclusive legal authority by a set of laws under the emperor Augustus,83 women begin to assume a more visible role in political, economic, and social life. The very fact that Paulina’s agency—or lack thereof—is a topic of controversy speaks to the new condition of women and a corresponding shift in the expectations of ancient readers. The issue, as Seneca sees it (at least in Tacitus’s account), is whether Paulina will live to carry on Seneca’s reputation, die and thus acquire fame for herself, or worst of all, live just long enough to suffer further injustice. The last point is of special concern to Seneca because of his particular love for her (sibi unice dilectam: Tac. ann. 15.63.2). While such a feeling is potentially present in any marital relationship, it assumes greater significance, once again, in an era when a weakening of the affective ties of clan (gens) leads to an intensification of the pair bond.84 Paulina’s presence in the suicide narrative also speaks to the role of women as bearers of unifying cultural traditions, inheritors of both the real and the symbolic capital of their spouses: Seneca’s suicide is itself, as suggested above, a strategy for making sure his property is transferred to his heirs; and as heir, in every respect, Paulina is a potential

82 On the double significance of Jupiter Liberator, as symbol of catastrophe forestalled and of Stoic reason, see Veyne 2003: 186. 83 Especially the much-misunderstood lex Iulia de adulteriis. For proper interpretation, see Daube 1972, Daube 1986, Cohen 1991, and Cantarella 1991. Daube’s observation (1986) that legal personhood carries a high price is startlingly relevant to Paulina’s situation. 84 Habinek 1997.

thomas habinek

continuer of both her husband’s and her father’s traditions (she is, after all, a Pompey). Finally, Paulina’s presence reminds us that women’s relationships to philosophy are a leitmotif of Seneca’s life: his father tells his mother not to study it (a fact for which Seneca consoles her after his father’s death); Agrippina tells him not to teach it to Nero; a fragmentary essay On Matrimony raises the possibility of a couple’s joint quest for virtue; and both Dio and Tacitus imply that Paulina had learned wisdom from her husband.85 Tacitus even has Seneca assign to Paulina the glory (gloria) and renown (claritudo) that he himself had sought for his philosophical achievements.86 Slavery We can press the Tacitean death scene yet further for the access it offers to other dimensions of the social self in the middle of the first century ad. As Tacitus reports, when the conspiracy against Nero was being investigated, Seneca was denounced by one Natalis for having said that his well-being depended on that of Piso. When the accusation was transmitted to Seneca, rather than panicking, he responded assertively, reminding Nero that (among other things) “he did not have a ready temperament for sycophancy—which was known to Nero, who had more often experienced free speaking (libertas) from Seneca than servitude (servitium).”87 The contrast between freedom and slavery may seem to be a rhetorical one,88 since Seneca was in no legal sense anyone’s slave, but it points us to features of the death scene too easily overlooked, especially the role of slaves within it. It is slaves and freedmen who bind Paulina’s wounds

85 On Helvia and philosophy see dial. 12 (= cons. Helv.).15.1, 17.3. On Agrippina’s opposition to Nero’s study of philosophy, see Suet. Nero 52. The fragments of the De matrimonio (Haase 1853: 3.428–434) refer to Terentia’s acquisition of wisdom from Cicero, to women’s capacity for virtue, and to the sage’s need to love his wife judiciously and not out of affect or desire. The preservation of the fragments seems somewhat skewed by Jerome’s desire to use them to support his own views on chastity as the best moral state for a Christian. 86 Tac. ann. 15.63.2; cf. Habinek 2000. As Andreas Heil reminds me, Martial 10.64 associates Lucan’s widow with his poetic glory long after his death. In addition to the issue of women’s status and capabilities, Seneca is a touchstone for changing concepts of masculinity, especially for a senatorial class that has much less opportunity to demonstrate military prowess than it had had under the republic: for ample discussion, see Habinek 1997: 137–150, Habinek 2000, and Roller 2001: 99–107. 87 Tac. ann. 15.61.1, trans. Woodman 2004: 334, with modifications. 88 See Roller 2001: 213–288 for an excellent discussion of the use of the language of slavery to negotiate the new relationship between princeps and elites during the Julio-Claudian period. Armisen-Marchetti 1989: 113–115 documents the pervasiveness of slave imagery throughout Seneca’s writings.

imago suae vitae: seneca’s life and career

at the order of the soldiers (Tac. ann. 15.64.1) and, apparently, slaves who help the enfeebled Seneca into the asphyxiating bath, for it is they who are accidentally splashed by him (ann.15.64.4). It may well be a slave, or at least a freedman, to whom Paulina refers when she demands the hand of the “striker” (percussor: ann. 15.63.1), and a slave (or slaves) who delivers the blow that severs simultaneously the veins of husband and wife.89 The scene illustrates the omnipresence of slaves in elite Roman life—even at the most intimate and emotionally charged moments—but also reminds us of how easily slaves can be overlooked by modern scholarly commentary. For all commentators on the scene, Seneca’s invocation of Jupiter Liberator refers either to Stoic philosophy (as above), civic religion, or Socratic precedent.90 But the invocation is prompted by his accidental splashing of nearby slaves with the water from his bath. Leaving aside how such slaves might have taken Seneca’s final sharp reference to “liberation” (was he prevented from manumitting them when denied access to his will by Nero’s thugs?), we might at least see the obsession with freedom on the part of Roman Stoics such as Seneca for what it is: a psychological and ideological response to the ubiquity of slavery and to the master class’s dependence on it.91 Elite anxiety about slavery finds ample expression in Seneca’s writings, especially in two of his most famous letters. In one, letter 47, he makes the case for the merciful treatment of slaves, without challenging the institution of slavery per se. The letter is a direct attempt to ameliorate the fear provoked by an expanding and diverse slave population in the city of Rome.92 In

89 The Latin says that “they opened their arms with the same blow” (eodem ictu brachia ferro exsoluunt, Tac. ann. 15.63.2). But how could they (Seneca and Paulina) have delivered such a blow together? It seems more likely that the Latin here, as frequently elsewhere, includes within an action attributed to a master the actual action of a slave or slaves: on this linguistic feature of classical Latin, see Reay 2003 and 2005. To Reay’s dossier we might add Sen. nat. 3.7.1 in which Seneca describes himself as “a diligent digger of vines”—almost certainly what he means is that he orders his slaves to be diligent diggers of vines. It is hard to imagine Seneca doing the dirty work of the vast vineyards he possessed. Also relevant is Sen. epist. 12.2, in which he describes himself as having planted the plane trees at his villa: ego illas posueram. Ego here would seem to encompass much more than the modern “I,” limited as it is to a single, distinct bodily organism. The expansive self of the Roman aristocrat could encompass family members as well: see Sen. benef. 5.19.8 f. 90 Veyne 2003: 187 notes that the civic Jupiter rescues a community from potential enslavement, but he fails to draw the connection to the slaves in Seneca’s household or to the Roman experience of slavery more generally. 91 For the emergence of freedom as an ideal in the context of slave societies, see the classic historical and sociological study of Patterson 1991. 92 For the master-class’s reaction to the expansion of slavery, see the remarks attributed to Cassius Longinus in ad61 (Tac. ann. 14.44.1–4). Seneca’s letter recommends easy coexistence

thomas habinek

another, letter 70, which ostensibly constitutes a defense of suicide, Seneca in fact exposes the cruelty that sustains the Roman slave economy, while also equating his position, at the mercy of Nero, to that of a slave, at the mercy of his master. “Law eternal has brought about nothing better than this,” he writes. “There is one entrance to life, but many exits” (Sen. epist. 70.14). The exits Seneca chooses to describe are those of Socrates and a noble Roman lady, but also, in a rising crescendo, the suicides of gladiators, slaves, and criminals who have been thrown to the beasts. Each finds a way to avoid the torture (tormentum: Sen. epist. 70.11, 70.15) his superiors have prepared for him. For Seneca, as for many Roman writers, the experience of dominating intensifies the fear of being dominated.93 This double bind—relying on slaves and therefore fearing the condition of enslavement—may have been particularly intense for Seneca and others of his background, since there is good reason to believe that the use of slave labor increased significantly in the late Republic and early Principate, especially in Seneca’s home province of Baetica,94 where there also seems to have been an expansion of the class of honesti—men who met the income qualification (by one calculation, 2.6 times subsistence) for service on local juries,95 and who constituted, in one historian’s view, the prime beneficiaries of economic growth.96 Seneca’s constant reversion to the topic of slavery, directly or indirectly, transfers to the literary register the social realities of his era. Whether Seneca himself treated his slaves cruelly or harshly—an unanswerable question—is less important than the testimony he provides for Roman awareness of the peculiarity of the institution of slavery and Roman understanding of the complex relationship between the elite self and the system of domination in which it takes form.97

with slaves in order to secure their loyalty and repeats the usual Stoic exhortation against “enslavement” to one’s vices and desires. Seneca also taps fear of slaves in De clementia 1.24 and awareness of their potential power over their masters in De beneficiis 3.23.1–4. 93 For an interesting parallel, see Patrick Henry’s famous speech before the Virginia House of Burgesses. Henry exhorts his fellow slave owners to resist the British (“give me liberty or give me death”) by summoning up images of the British doing to the Virginians what the Virginians have license to do to their slaves. On the frequency of slave suicide in Rome, see Bradley 1994: 112f. 94 Alföldy 1996 summarizes the evidence and notes that the expansion of slavery correˇ sponds to a broader process of social differentiation. See also Stearman et al. 1987. 95 Haley 2003. Haley’s fullest evidence is for a later period than the life of Seneca, i.e., 70–190ad, but the various trends he describes clearly have their start in earlier periods. 96 Haley 2003: 3–5. 97 For further discussion of the role of literature in maintaining structures of domination, especially slavery, see Fitzgerald 2000 and Habinek 2005c. McCarthy 2000 and Roller 2001

imago suae vitae: seneca’s life and career

Wealth and Economic Transformation From the power of writing and issues of gender and slavery, we move to consideration of Seneca in relationship to the imperial economy. As Evan Haley notes, a new scholarly consensus holds that during the Julio-Claudian period the Roman imperial economy experienced real growth, especially in the settled western provinces such as Baetica.98 The triumphant rise of the Annaei, including the entry of Seneca and his brother Gallio into the Roman Senate, was surely spurred in part by that growth. Indeed, the Senate persisted as an important institution under the Principate because its members represented the economic elite, both landed and commercial, without whom the Empire could not function. Turnover in the Senate’s membership tracked both economic and military success throughout the Empire.99 But the expansion of wealth, combined with the shift in identity of those who controlled it (i.e., from a handful of old Roman families to a broadly based Italian aristocracy to an evolving pan-Mediterranean elite), naturally generated hostility, concern, and misunderstanding among various parties. This aspect of political and social life manifests itself in the accusations directed against Seneca concerning his extraordinary wealth and in Seneca’s concern throughout his writings with the problem of the reflective individual’s proper relationship to ownership and consumption.100 The clash between the two phenomena (personal wealth versus philosophical belittlement of possession) led in antiquity and beyond to charges of hypocrisy, but once again the ethical dimension is less interesting and less easy to resolve than the social. During Seneca’s lifetime his prime attacker was P. Suillius Rufus, former quaestor of Germanicus (Tac. ann. 4.31.3), who had been relegated under Tiberius for taking money in connection with a trial, then restored and elevated under Claudius, during whose reign he was closely associated with

show how representations of slavery become a way for the slave-owning class to think through their own subordination within the hierarchies of Roman society. 98 Haley 2003: 3. 99 Hopkins 1983, Garnsey and Saller 1987: 123, Scheidel 1999, and Habinek 2000. 100 Attacks on Seneca are summarized in Tac. ann. 13.42.1–4. Seneca’s own treatise, De vita beata, parries such attacks and thus provides indirect evidence of their nature and extent. For the sources and extent of Seneca’s wealth, see Griffin 1976: 286–314. In denying that the treatise De vita beata responds to the attacks of Suillius, Griffin 1976: 19f. evinces needless skepticism. Seneca routinely used his writings to comment on contemporary affairs, e.g., De clementia following the murder of Britannicus, letter 47 on the execution of the familia of Pedanius Secundus, letter 73 on his own relationship to Nero.

thomas habinek

Messalina (Tac. ann. 11.1.1–11.5.1). He himself was suffect consul in ad 43 or 45, and his son became consul ordinarius in ad50.101 In a sense, Suillius’s career anticipated that of Seneca by about half a generation: movement toward the center of power facilitated in part by maternal connections; success through rhetorical achievement; extended relegation followed by triumphant return; enrichment through willingness to work closely with those holding greater power; senatorial rank joined to influence within the imperial household, the latter attained at least in part through close association with an emperor’s wife (Messalina for Suillius, Agrippina for Seneca); and success that carried over to other members of the family (e.g., Suillius’s son, Seneca’s brothers and nephew). Suillius’s hostility to Seneca might thus be attributed to psychological causes (e.g., narcissism), court politics (associates of Messalina undone by associates of Agrippina), or social transition: one generation of upwardly mobile political neophytes resisting the next. The last-mentioned motive has the broadest implications, for it allows us to see in the titanic struggle between Suillius and Seneca not just the supplanting of one generation of beneficiaries of the new order by another but also the crystallization of two conflicting tendencies in the Roman economy. Suillius, we are told, earned his fortune by accepting payment for representing others in legal cases.102 Seneca, on the other hand, grew wealthy at least in part through participation in the gift economy of the court.103 During the period in question, the Roman elite found it difficult to make up its collective mind about the appropriateness of the Suillian method:

101 For evidence pertaining to the birth and career of Suillius, see Syme 1970: 27f. The son is M. Suillius Nerullinus. 102 This is the gist of the situation described in Tac. ann. 11.5.1–3, where Suillius is said to have accepted 400,000 sesterces from a Roman knight named Samius, presumably to defend him, but then turns against him. The incident prompted the senators to ask for reinstatement of the Cincian Law of 204bc, which forbade pleading for money. The Senate apparently succeeded in its request, at least in 54ad, when, according to Tacitus, it was decreed that “no one should be bought by wage or gifts to plead a case” (Tac. ann. 13.5.1, trans. Woodman 2004: 247). In his final appearance in the Annals, Suillius as much as admits that he pleads for money: Tacitus calls him venalis (ann. 13.42.1), i.e., “for sale,” and he speaks of the modest fortune (modicam pecuniam) he has acquired through hard work (labore: Tac. ann. 13.42.4). 103 We cannot be sure whether the wealth of the Annaei, prior to Seneca’s elevation by Agrippina, derived from land, commerce, or both. Seneca describes himself and his brothers as locupletes (dial. 12 [= cons. Helv.].14.3), a term that is more likely to indicate that their wealth derived from landowning than other sources (see Levick 2003). But then wealth derived from land continued to hold a higher status under the Principate than wealth from commerce, so Seneca would be more likely to depict his family holding as being of the former sort.

imago suae vitae: seneca’s life and career

first it restored, then ignored, then restored, then ignored republican-era legislation against accepting pay for legal representation. 104 The issue lay at the heart of the social system and of elite self-definition. Was legal service a form of patronage (i.e., noblesse oblige) on the part of those who had influence and education, as it had been during the glory days of Cicero, Hortensius, and other late republican orators, or was it a useful skill to be compensated accordingly? Suillius, although older than Seneca, represents the newer, more “market-based” approach, while Seneca in effect reverts to a more traditional model. The paradox is that the Senecan approach to patronage and gift-giving, which he defends and discusses at length in his essay On Benefits,105 seems to falsify the changing economic reality, that is the move to a more market-based system of exchange that characterizes the early Empire and prompts the economic growth that stabilizes the political system.106 Suillius in effect represents the reality principle, Seneca the force of mystification, the former exposing, albeit unintentionally, the potential harshness of a system based upon payment for services, the latter patching over the contradictions between economic reality and cultural ideals. Tacitus, in particular, characterizes the struggle between Suillius and Seneca as a personal one, and modern historians have happily followed his lead. But here, as elsewhere in ancient historiography, the personal becomes a way of exploring social issues for which the ancients had no other conceptual language. Suillius’s accusations against Seneca, it should be noted, comprised not just the charge of hypocrisy (i.e., the contradiction between a philosophy that preached detachment from wealth and the accumulation of a huge fortune), but also the claim that Seneca had acquired his wealth through no useful or productive activity. In Suillius’s view, Seneca’s “inert endeavors”

104 The lex Cincia, banning payments to advocates, was passed in 204 bc and reaffirmed under Augustus in 17 bc. In ad47 it was de facto inoperative, since the senators found themselves pressed to insist on its enforcement (Tac. ann. 11.5.2), only to be reinstated or reinforced in ad54 (Tac. ann. 13.5.1). Yet in ad58, despite Suillius’s admission that he pleads for pay, Seneca and his allies rely on charges of provincial mismanagement and embezzlement of public funds in order to secure his relegation (Tac. ann. 13.43.1–5). 105 Griffin 2003 makes much of the difference between patronage and gift exchange, arguing that the latter is the true topic of the work De beneficiis, but for our purposes the distinction is immaterial: both patronage and gift exchange are expressions of an embedded economy (i.e., one in which exchange is “embedded” in social structure) in contrast to the disembedded nature of a market economy. More suggestive on De beneficiis is Andrew 2004, who views it through the lens of Diderot’s apologia for his own state of economic dependency on Catherine the Great. A classic instance of Seneca’s resistance to markets can be found inbenef. 3.15.1–4. 106 In Haley’s words (2003: 12), the early Roman Empire is best described as having a “subsidized market economy.”

thomas habinek

(studiis inertibus: Tac. ann. 13.42.3; a deliberate oxymoron) contrast with his own labor (labore: ann. 13.42.4), effort (adsequi: ann. 13.42.3), endurance (toleraturum: ann. 13.42.4), and risk (periculum: ann. 12.42.4). Cultural Authority During Seneca’s lifetime, the tension between old and new economic systems (or better, old and new aspects of one evolving system), and old and new methods of acquiring wealth, corresponds to a struggle between old and new ways of conceptualizing and organizing human endeavor more generally. For example, the question of pay for legal representation overlaps with that of the cultural authority of legal knowledge. As Aldo Schiavone has shown, perhaps the most famous legal case of the Neronian period—the debate over the collective punishment of a household of slaves for the murder of the master Pedanius Secundus—has implications not only for our understanding of Roman attitudes toward slavery (crucial as that is), but also for our conceptualization of the role of law and legal expertise in the Roman world.107 As Tacitus relates the episode (ann. 14.42.1–45.2), although laws demanding collective punishment were still on the books, masses of the people and many of the senators were reluctant to see them enforced in this instance. But the old-style rigor of Cassius Longinus, whom Tacitus elsewhere describes as the pre-eminent jurist of the era (ceteros praeeminebat peritia legum: ann. 12.12.1), carries the day. In Cassius’s view, “every exemplary punishment contains an element of unfairness (aliquid ex iniquo), which, being directed against individuals, is outweighed by the public good in general (utilitate publica rependitur).”108 Cassius is appealing not just to the mos maiorum but to the autonomy of the law and ultimately of those, such as himself, who specialize in it (studium meum: ann. 14.43.1). His argument is an intervention in the long-standing negotiation between jurists and emperors concerning the role of law and its relationship to imperial authority and in an equally long-standing dispute between adherence to the letter of the law and acceptance of a ius aequitatis, i.e., a tempering of the law through application of a basic sense of fair play.109 Cassius’s remarks, in particular his belittlement of aequitas, provide a context

107 Schiavone 2003. Bradley 1994: 113f. describes repeated Roman attempts to deal with the legal aftermath of assault or murder by slaves. 108 Tac. ann. 14.44.1, my translation. 109 Schiavone 2003, who traces the conflict in the writings of Cicero, Seneca, and the Roman jurists.

imago suae vitae: seneca’s life and career

for Seneca’s repeated appeals to aequitas, misericordia, clementia, and so on throughout his career. Seneca’s preference for the spirit over the letter, attractive as it may seem to humanistically inclined readers, nonetheless positions him in opposition to those who would preserve the autonomy and distinctive power of legal discourse. Seneca’s reluctance to privilege the professional claims of jurisprudence is but one part of a much broader program of asserting the force of generalized over specialized discourse.110 The preference for the general is especially apparent in Seneca’s approach to philosophy: his decision to write in Latin, as opposed to the Greek still preferred by “professionals”; his loyalty to the Roman sect of Fabianus and, before him, Sextius; his preference for Stoicism, the favorite philosophical viewpoint of the Roman elite, as opposed to more exclusivist movements;111 and his (paradoxically) rigorous defense of the preceptive or hortatory aspect of philosophy and mode of philosophical writing.112 While the contrast between the generalist and the specialist may appear to be merely a matter of intellectual or personal preference, in fact it has important social implications as well. Andrew Wallace-Hadrill has gone so far as to argue that the disintegration of patriarchal and noble authority prompted by the rise of specialists (in law, religion, temporality, etc.) is characteristic of the fall of the Roman Republic, just as the attempt to reintegrate specialists into the imperial project—and often, quite specifically, into the imperial household—is critical to the development of the Principate.113 This tension between the centrifugal forces of specialization and the centripetal efforts of the emperors is still manifest in the later years of the Julio-Claudian dynasty: Seneca’s recurrent attention to the issue is itself evidence of its continuing vitality. Once again, his actions and writings provide access to a different kind of truth than that sought by his biographers. Indeed, the central drama of his life, as transmitted through his writings and those of the other ancient sources, is the struggle between the pursuit of philosophy as an autonomous and self-transforming endeavor and the desire

110 For further discussion of this important topic, see Habinek 2000, esp. 284–292. The philosophical eclecticism with which Seneca is sometimes charged is but one manifestation of this much broader tendency. 111 On Stoicism as the unifying discourse of the Roman elite during the late Republic and early Principate, see the excellent but neglected article of Shaw 1985. 112 Especially Sen. epist. 94 and 95, as discussed by Habinek 1989: 238–254, Inwood 1995, and Habinek 2000: 289–292. 113 Wallace-Hadrill 1997. The social and political consequences of the rationalization of various activities previously controlled by nobiles is the topic of an important study by Moatti 1997.

thomas habinek

to participate in the broader shaping of society. As I have argued elsewhere, that struggle resonates with successive generations of Romans who regard his life as “good to think with” about the tensions and priorities that shaped imperial society.114 Succession, Imagination, and the Future A preference for generalizing discourse, whether philosophical or rhetorical, is part and parcel of a movement toward the center of power. Roman absolutism, like that of other times and places, sought to encompass specialized skills within its patronage and to orchestrate them to its benefit.115 Under Tiberius and Claudius, this effort took the form of reliance on freedmen, especially in the imperial household. Their specialized knowledge made them useful to the emperor; their antipathy toward the traditional aristocracy guaranteed their loyalty. The traditional aristocracy (or those who sought to inherit their privileges and positions) fought back by emphasizing aspects of cultural performance that were less accessible to those who had not been raised in the right circles. The contrast between the literary productivity and creativity of Seneca, his father, and his nephew, on the one hand, and the freedman Polybius’s “dumbing down” of high culture on the other, in the form of prose paraphrases of Homer and Vergil, can be understood as one front in a wide-ranging social war.116 Seneca’s preference for a gift economy and for aequitas as opposed to legal precision is part of the same struggle. The conflict between specialized and generalizing skills, discourses, and the like overlaps with and reinforces two other constitutive contrasts of the early Empire: birth versus achievement, and orientation to the past versus orientation to the future.117 Not surprisingly, Seneca’s life epitomizes these concerns as well. Birth is critical to all aspects of power in the Empire, as it had been under the aristocratic republic that preceded it. Seneca as much as anyone else assumes differences in outlook, worth, and so forth based on

114

Habinek 2000; cf. Ker 2006. E.g., France under the absolute monarchy: see Mousnier 1979, vol. 1, ch. 10, “The Society of Corporations.” 116 On Polybius’s literary activities, see Sen. dial. 11 (= cons. Pol.).8.2 and 11.5. Seneca also seems to suggest that Polybius was intending to write fables (dial. 11.8.3)—a characteristic genre of slaves and freedmen, who acquired voice only through the imitation of inferior creatures: see Bloomer 1997b, Marchesi 2005, and Kurke 2006. 117 I have examined at length the conflict between birth and achievement as a structuring principle of Senecan philosophy in Habinek 1998: 137–150. 115

imago suae vitae: seneca’s life and career

one’s social and geographical status at birth. Yet his own rise to power made him the defining exemplum of the importance of achievement as opposed to birth. When the poet Juvenal, writing some two generations after the death of Seneca wants to illustrate the principle that the low-born have contributed more to Rome’s success than the nobility, he includes Seneca among the supporting examples, along with the republican general and consul Marius, who warded off barbarian invaders, and the “new man” Cicero, who defended the state from the conspiratorial activities of the higher-born Catiline.118 In his writings Seneca managed this contradiction through the advancement of an aristocracy of virtue: readers and writers such as himself who earned the benefits of high birth by re-creating themselves in a Stoic mode. The exhortation “begin to live” had, for Seneca and others, a social and political, as well as an ethical significance. Through the process of self-transformation, Seneca in effect acquired the right to speak to his fellow Romans as father to son, nobleman to commoner, and pointed the way for others to do likewise (his popularity in the generation after his death is just one indicator of the social role he came to fill).119 Setting oneself up as an exemplum implies an orientation to the future, a hope or even expectation that others will follow one’s own path to glory. But the future is not the direction to which an aristocracy usually turns. The essence of a system that grants entitlement based on birth is that the past, not the future, authorizes one’s status. This turn to the past, a measuring of anything and everything against the real or purported ‘custom of the ancestors’ (mos maiorum) is one of the defining features of republican Roman politics. Its importance had not yet been abandoned during the early Principate when Augustus in particular, but also Tiberius and even to some extent Claudius, justified their rule, their responses to crises, even their political and cultural innovations, by reference to the authority of the past.120 So pervasive is this turn toward the past that one scholar (wrongly, in my view), has argued that all imperial politics are about preserving the status quo, rather than preparing for the future.121

118 Iuv. sat. 8, especially verses 211f., on the theme quid stemmata faciunt (“what difference does genealogy make?”). For discussion, see Habinek 2000. 119 The theses of this and the following paragraphs are defended at much greater length in Habinek 1998: 137–150 and Habinek 2000. 120 The Augustan slogan res publica restituta is the most obvious of such attempts to justify innovation through reference to the past. 121 Veyne 2003: 20.

thomas habinek

Past and future pivot on the issue of succession. Under the Republic, as long as the mos with respect to elections, character of offices, and so on was upheld, one could expect a smooth transition of power from one generation to the next. But the Principate foundered and threatened to collapse precisely over the issue of transition of power. How was one to follow precedent when the whole enterprise was unprecedented? Thus the preparation of successors, the creation, in effect, of a past for the future, was a crucial, perhaps the crucial feature of the politics of the early Principate. Most of the bloodletting that makes Tacitus’s account of the Julio-Claudian era such disturbing reading is symptomatic of chronic uncertainty over succession. Much of the positive effort of the more successful players of imperial politics also consists in the cultivation of reliable heirs: as Greg Rowe has shown, the grooming of princes is a key component of the new political culture of the early Principate.122 In this matter, as in so many others, Seneca’s life epitomizes the successes and failures of his era. Authorized—but only to a certain extent—by his Spanish equestrian origin, Seneca reaches the center of power never to look back at his provincial beginnings. Lack of “national feeling,” as it has been called, is a hallmark of Seneca’s life and writings.123 Recalled to Rome to educate the prince, Seneca flourishes precisely due to his mastery of the traditional skills of elite culture. 124 Arguing against economic innovation in favor of the traditional, aristocratic practice of gift exchange, he becomes an important investor—someone who looks not to the preservation of past goods but to the possibility of future growth and profit. Even at the moment of his death, Seneca turns simultaneously to the past and the future. Denied the opportunity to amend his will, he offers to his friends and family the “image of his life,” employing, at least in Tacitus’s vocabulary, the traditional word for the death mask of a noble ancestor: imago.125 Although a committed supporter of the imperial system, and, for a long time, of the particular regime of Nero, Seneca nonetheless invokes an age-old and essentially republican institution, one that entails the display and reanimation of the waxen images of generation upon generation of ancestors. To have created his own imago is an odd, albeit not unprecedented, achievement.126 To hand it over to his

122 See Rowe 2002, who discusses more generally the transition to dynastic politics under the Julio-Claudians. 123 Griffin 1976: 255 and Sørensen 1984. 124 For further discussion, see Habinek 2000. 125 Tac. ann. 15.62.1. On the social function of imagines, or death masks, in Rome, see Flower 1996, Habinek 2005a: 97 f., 122–132, 258 f., and Dufallo 2007. 126 See the excellent discussion of Cicero’s investment in his own imago: Dugan 2001. In

imago suae vitae: seneca’s life and career

successors is to acknowledge the new authority of the future at the expense of the past. Unlike the newly dead aristocrat of republican times, whose funeral entails a speech of defense on his behalf before an audience of his reanimated ancestors, Seneca looks to the future for support and validation. In so doing he instantiates the final movement from republican to imperial politics, from Rome as city to Rome as universal Empire, from philosophy as study of dogma to philosophy as “art of living,”127 from self as construct of the past to self as projection into the future. The ratio of such a life can only be grasped and articulated by a societas of present and future observers.

many respects, Cicero anticipates Seneca’s reorientation of Roman culture from past to future and militaristic to humane, as Seneca was well aware. 127 The phrase is that of Veyne 2003: viii.

THE WORKS OF SENECA THE YOUNGER AND THEIR DATES

C.W. Marshall The works of L. Annaeus Seneca cannot be dated with any great precision. This is frustrating, since the interpretation and understanding of his immense and wide-ranging output would benefit from a precise chronological sequence. The works themselves resist any such systematization, however: Seneca makes very few references to his personal circumstances, which is appropriate considering his philosophical emphasis on the inner life, and this reticence has led one scholar to ask ironically, “Est-il possible de ‘dater’ un traité de Sénèque?” 1 Nevertheless, some headway is possible, and Giancotti (1957) on the Dialogues, Abel (1967) and Griffin (1976) on all the prose works, and Fitch (1981) and Nisbet (1995: 293–311) on the tragedies have made significant advances in understanding the dates of Seneca’s literary writings. This chapter seeks to integrate the conclusions of these studies. When something can be said in relation to a landmark event in Seneca’s life, it is often limited to terminus ante or post quem: his exile to Corsica following the accession of Claudius in ad 41 (Dio 60.8.5); his recall to serve as personal tutor to Nero in ad49 (Tac. ann. 12.8.2); his rise to prominence on Nero’s accession in ad54 (ann. 13.2.1); his diminished influence following the murder of Agrippina in ad 59 (ann. 14.14.2); and his withdrawal from all influence with Nero in ad62 (ann. 14.52.1). Within the spans bounded by these points, a generally coherent picture emerges. There are of course many methodological issues associated with assigning dates (both relative and absolute) to literary works. Internal references, stylistic features, external testimonia, and other factors may be employed to argue for a date, and different types of argument will carry different weights with different readers. Crucially, circularity must be avoided, and interpretations of a work cannot presume a date for which evidence does not exist. It may be possible to perceive a development in thought from one work to the next, but that in itself cannot be used as an argument for the relative dates of the works in question. There is also a danger with this sort of analysis in assuming a tendency toward the limits: a given work that shows

Grimal 1949a, and see also Griffin 1976: 5 n. 2.

c.w. marshall

indications of being written before ad 54, for example, does not need to have been written close to ad 54; the limits identify boundaries, but in most cases do not establish more or less likely dates within the possible range. Indeed, the opposite is true: as one approaches the limit, there is a greater need for independent, unrelated points of reference. While any two arbitrary facts touching on an author’s life may be close in time to one another (e.g., when a given work was written and an event recorded by Tacitus), it is improbable that such clusters will occur repeatedly, given how few data points survive. Agnosticism often remains the most prudent course. These issues are of course further confused if works are re-worked or re-edited following their initial circulation.2 My hope here is not to overstate the case, but within each section to describe works in what may reasonably be thought to be chronological order, given the appropriate cautions offered below. Dialogues Ten treatises in twelve books, as found in the eleventh-century Ambrosian manuscript, are collectively known as Dialogues (Dialogi) and are numbered 1–12. The earliest of these, Ad Marciam de Consolatione (= dial. 6), probably dates to ad39 or 40, although a later date into the 40s is possible.3 Seneca writes with authority to console Marcia, daughter of the historian A. Cremutius Cordus, on the death of her son Metilius three years earlier, and Seneca may have written works before this that are no longer extant. Reference to the republication of Cordus’s works (cons. Marc. 1.3), which occurred under Gaius (Suet. Cal. 16.1), establishes a terminus post quem.4 Further, praise of Tiberius (as is found in cons. Marc. 3.2, 15.3) “would not have been prudent before 39” (Griffin 1976: 397, citing Dio 59.16.4 and Suet. Cal. 30.2). For the upper limit of the range, Abel argued that it must be before Seneca’s exile, based on in qua istud urbe, di boni, loquimur? (cons. Marc. 16.2), which suggests both speaker and addressee are in Rome.5 This is not convincing: loquimur could equally be an epistolary conceit, whereby the letter creates the air of intimate communication, regardless of where the sender is; indeed, this effect would 2

E.g., Schmidt 1961. For previous discussions, see Giancotti 1957: 45–73, Abel 1967: 159f., Griffin 1976: 397, and references there. 4 Bellemore’s argument (1992) for an earlier, Tiberian date for the work requires rejecting Suetonius’s evidence. 5 Abel 1967: 159 f., and references there. 3

the works of seneca the younger and their dates

be enhanced if Seneca were in exile. If this view is accepted, then Griffin’s calculation of Marcia’s age excludes only a date after ad 49 (1976: 397). While a Gaian date is perhaps reinforced by the absence of any mention of Gaius in the text, a date in the 40s remains possible, depending on the interpretation of cons. Marc.16.2. Two other consolationes, Ad Polybium de Consolatione and Ad Helviam Matrem de Consolatione (= dial. 11 and 12), certainly belong to the period of Seneca’s exile.6 The first, to Polybius, a freedman and secretary to Claudius a studiis (Suet. Claud. 28), offers consolation on the death of his brother. It was evidently written before Claudius’s conquest of Britain in ad43, which remains an anticipated event in cons. Pol. 13.2 Britanniam aperiat (“may [Claudius] open up Britain,” though the verb could conceivably be used for a short while thereafter). The letter is an unsuccessful effort to win Seneca’s recall. The date of his letter to his mother, offering her comfort on his own exile, cannot be circumscribed so precisely: Abel (1967: 163) took the ten months mentioned in cons. Helv. 16.1 literally; Griffin cautions against this, emphasizing instead the length of his absence, stressed in cons. Helv. 1.2 and 2.5. Neither of these need be determinative, however, and it is safer to accept a larger range, between ad 41 and 49. The three books of the treatise De Ira (= dial. 3–5) likely also belong to Seneca’s exile.7 They are addressed to Seneca’s older brother L. Annaeus Novatus, whose name was changed to L. Junius Gallio Annaeus (PIR2 I 757), by ad52 at the latest, perhaps due to a testamentary adoption.8 References to Gaius make it clear that he is dead (de ira 1.20.8, 2.33.3–4, 3.18.3–4), but the force of modo (“recently”: 3.18.3) to describe an action of Gaius cannot be pressed unduly, as Seneca himself noted in epist. 49.4. It is not necessary to follow Coccia and Abel in asserting that the work must be confined to the period following Gaius’s death and before Seneca’s exile; any time between 41 and, at the latest, early ad 52 is possible. Griffin argues forcefully for a date of ad 55 for De Brevitate Vitae (= dial. 10), although she insists on nothing more specific than a date between ad

6 For previous discussions, see Giancotti 1957: 74–92, Abel 1967: 163f., Griffin 1976: 397f., and references there. 7 For previous discussions, see Giancotti 1957: 93–150, Abel 1967: 159, Griffin 1976: 398, and references there. 8 Griffin 1976: 48, n. 2. The so-called “Gallio inscription” (see Plassart 1967, Oliver 1971, and Hemer 1980), found at Delphi, provides the lynchpin for dating the New Testament (cf. Acts 18.12–17), and was written in the first half of ad52, and it shows that he was at that time proconsul of Achaia and was using the name Gallio. Since the proconsulship would have begun in ad 51, it is likely that the adoption occurred before this.

c.w. marshall

mid-48 and mid-55.9 This was the period during which the work’s addressee, Pompeius Paulinus, Seneca’s father-in-law (see ann. 15.60.4), was praefectus annonae. Brev. 18f. urge Paulinus to retire from his administrative duties managing Rome’s grain supply to more important things (maiora: brev. 18.2). In this period at least some of Seneca’s tragedies are likely to have been composed (see below), a fact that bears on the interpretation of the life Seneca recommends. De Vita Beata (= dial. 7) is dedicated to Seneca’s brother, to whom De Ira had also been dedicated, after he had been adopted by Gallio and had taken his name.10 The Gallio inscription (see n. 8) attests his name in ad 52, so the adoption must have taken place before this. Vit. beat. was written after the adoption, but conceivably still written before ad 52. While Gallio did not die until ad66, the subject matter, in which Seneca justifies his great wealth and prosperity, strongly suggests a Neronian date before ad 62. It does not follow, however, that the work was composed as a direct response to the attacks in ad58 by P. Suillius Rufus (ann. 13.42.1–43.5), as Griffin argues (1976: 19 f. and 306–309). Two, perhaps three works were dedicated to Annaeus Serenus, who died as a prefect of the nightwatch (Plin. nat. 22.96) sometime before ad 64, when it is probable Seneca wrote epist. 63.14 (and perhaps before the beginning of ad62, when Tigellinus became praefectus vigilum).11 The first of these, De Constantia Sapientis (= dial. 2), seems to have been written after the death of Valerius Asiaticus in ad 47 (ann. 11.3.2), given the reference to him in const. 18.2. De Tranquillitate Animi (= dial. 9), in which Serenus is a Stoic (tranq. 1.10), is later than const., in which he is still an Epicurean (15.4; the sequence of Serenus’s philosophical development is secured by const. 3.2, and see Griffin 1976: 316). The fragmentary De Otio (= dial. 8) is also addressed to a Stoic, and there is reason to believe this too is Serenus, although this is conjectural. If so, it also postdates const., but the relative position between it and tranq. cannot be determined.12 Many have believed De Otio to be the last of the three, but Seneca’s “shifting positions […] may well be more experimental

9 For previous discussions, see Giancotti 1957: 363–445, Abel 1967: 162f., Griffin 1962 and 1976: 398 and 401–407, and references there. 10 For previous discussions, see Giancotti 1957: 310–362, Abel 1967: 160–162, Griffin 1976: 399, and references there. 11 Griffin 1976: 447f. and Giancotti 1957: 153–157. See also Giancotti 1957: 151–177, Abel 1967: 159 and 162, Griffin 1976: 316 f. and 399, and references there for the relationship between the three works. For const., see also Giancotti 1957: 178–192; for tranq., see also Giancotti 1957: 193–224; for de otio, see also Giancotti 1957: 225–243 and Williams 2003: 12–16. 12 Griffin 1976: 316 f.

the works of seneca the younger and their dates

than a ‘sincere’ reflection of his own beliefs on either or both occasions” (Williams 2003: 16). All three works were then composed between ad 47 and (probably) ad62, with const. being composed first. There is no solid basis for establishing the date of De Providentia (= dial. 1), a work dedicated to Lucilius, who is also the dedicatee of works composed in the 60s, except to say that it was written after the death of Gaius (see prov. 4.4).13 Abel placed it securely in ad 64 because the work presents Lucilius as an avowed Stoic, but the literary nature of this (or any) philosophical work makes such an inference untenable. Tying the work to the period of the Letters (Epistulae) however, is not unreasonable, even if it cannot be proved. Tragedy, Verse, and Satire Of the more than seventy epigrams that have been ascribed to Seneca, none can be dated securely. The Anthologia Latina ascribes only three to him explicitly (Anth. Lat. 232, 236, 237), the last two of which concern exile on Corsica. This, I suggest, undermines the attribution rather than reinforces it: Seneca was Corsica’s most famous refugee, and the subject is a likely one for someone appropriating Seneca’s voice. The reference to Seneca’s poetry as a model for the versiculi of Pliny (epist. 5.3.2–5) is not a statement about the genre Seneca employed, as other names in the list demonstrate.14 Nevertheless, if any epigrams are authentic, some may date to the period of his exile. It is to this period that one may also place the first group of Seneca’s tragedies, Agamemnon, Phaedra, and Oedipus. Eight authentic tragedies exist,15 and the echoes of Hercules Furens in Apocolocyntosis (see n. 20), if valid, set a terminus ante quem for that play of ad54. The earliest certain reference to Seneca’s tragedies is Quintilian, inst. 8.3.31, in which Quintilian recalls an exchange between Pomponius Secundus and Seneca about a tragedy in his youth (iuvenis admodum) that must date soon after the return

13 For previous discussions, see Giancotti 1957: 244–309, Abel 1967: 158, Griffin 1976: 400f., and references there. 14 Note also that Pliny is not certain that Seneca’s poetry was recited, at least by the author: recito tamen, quod illi an fecerint nescio (epist. 5.3.7, “But I recite [my verses], though I do not know whether they [my predecessors] did”). 15 Despite attempts of its revival (Kohn 2003), the idea that the plays are not by L. Annaeus Seneca has not won general approval. Quintilian, inst. 9.2.8, quotes Medea 453 and attributes it to Seneca, but that single reference is sufficient for the attribution. Two other plays in the manuscripts are spurious and post-Neronian. They are discussed at the end of this section.

c.w. marshall

of Pomponius to Rome at the end of ad 51: this suggests a public presentation of a play in some form in the early 50s, but says nothing about this being the time of composition.16 Stylistic features discussed by Fitch (1981), particularly the increased incidence of sense-pause mid-line in the iambic verses (which he demonstrates is also a useful diagnostic for Sophocles and Shakespeare, but not for Euripides) and the increased incidence of shortened final -o, identify three clear chronological groupings for the plays. Fitch reckons the percentage of sense-pauses that occur within the line as measured against the total number of sense-pauses within an iambic section: Agamemnon (32.4%), Phaedra (34.4%), and Oedipus (36.8%) form a coherent group, but contextual variation should create a margin of error that does not place the works within this group in a certain order. The play with the next smallest ratio, Medea (47.2%), clearly belongs to a different cluster. The plays in this group predate Hercules Furens (which belongs to the middle group) and therefore all were also composed before ad 54.17 The nature of the evidence does not allow any conclusion more precise than this: the three plays were composed (in whatever order) in a cluster, but no certain sequence can be determined. A Claudian date seems reasonably certain for these first three plays, but they could conceivably date to the exile, or before (into the reigns of Gaius or even Tiberius), or after.18 This metrical approach is to be preferred to those who seek covert antiNeronian messages in the tragedies, placing some or all in the 60s (e.g., Bishop 1985). Töchterle (1994: 44–48 and infra, p. 483) believes verbal parallels, particularly with nat., point to a composition date for Oedipus between ad 62 and 65. The implications are significant: in addition to vitiating Fitch’s conclusions, the date would point to a particular political purpose for the tragedies, whereby, in this case, Nero, Claudius, and Agrippina are to be mapped onto Oedipus, Laius, and Jocasta. Accounts of Jocasta’s death (hunc pete | uterum capacem: Oed. 1038f., hunc petite uentrem: Pho. 447) and Agrippina’s death as presented in post-Neronian sources (Oct. 368–372; uentrem feri: Tac. ann. 14.8.4) would therefore allude to Agrippina’s death in ad59. It is easier, if less scandalous, to believe that Tacitus and the author of Oct. “saw in Seneca’s presentation of Jocasta [in Oed.] a suitable model for their account of Nero’s mother, one which carried implications of incest and

See Tarrant 1985: 12 and Fitch 1987a: 50 f., and references there. This date suggests that the apparent allusions to Ag. 330–341 in Einsiedeln Eclogue 1.22–33 are real and not incidental; see Tarrant 1985: 11. 18 Coffey 1957: 150 adopts this agnostic position, followed by Tarrant 1976: 6f. 17

the works of seneca the younger and their dates

moral perversion at the heart of the late Julio-Claudian court” (Boyle 1997: 102). This process may have begun with Seneca himself, when writing Phoen. (see infra, p. 40). The second group of tragedies, Troades, Medea, and Hercules Furens, possibly centers around ad54, but again it need not, and all three could well predate this. As with the first group of plays, incidences of sense-pause within a line strongly suggests they were composed within a short time of one another: Medea (47.2%), Troades (47.6%), and Hercules Furens (49.0%). As with the first group, a reasonable margin of error does not allow for the order of works within this cluster to be established. Nisbet argues that Med., which likely alludes to Claudius’s invasion of Britain in ad43, is unlikely to have been composed after Claudius’s death, and suggests that ad51 to 52, when Gallio was proconsul in Corinth, “would be quite a good moment for Seneca’s Corinthian play” (1995: 295). This is appealing, but hardly certain. Verbal echoes have been detected between Herc. f. and apocol., and the gratuitous nature of the tragic Hercules in apocol. argues that the tragedy is the earlier work.19 Fitch believes “the tragedy was fresh in his mind and had either been written, or at any rate presented in recitatio, within a year or two of 54” (1987: 53), and this certainly could be the case. That would also mean that De Ira had probably been composed by the time Seneca began working on these plays, and this has some bearing on the interpretation of these works. Strictly speaking, however, though the relative position of Herc. f. and apocol. seems probable, there is nothing to require composition near this date, and it is conceivable that all the plays in these first two groups come from a period much earlier in Seneca’s career. The Apocolocyntosis, a prosimetric Menippean satire on the death of Claudius, almost certainly dates to November or December of ad 54: Eden accepts Furneaux’s hypothesis that it was produced for the Saturnalia, which began on December 17, at which Nero was rex (ann. 13.15.2).20 This would also establish some time between the more traditional encomium for Claudius’s funeral that Seneca wrote (ann. 13.3.1), to which it would have been compared in any case at the time. At least one play in this middle cluster (Herc. f.) predates apocol., and possibly all three do. A number of factors suggest instead that Seneca returned to the tragic form late in his career, and that the last two plays (the third group) are to be

See Mesk 1912, Weinreich 1923, and Fitch 1987a: 51–53. Eden 1984: 4f., esp. 5 n. 11. See also Griffin 1976: 129 n. 3. Tacitus surprisingly accepts that the position of rex fell to Nero by lot. 20

c.w. marshall

dated to the 60s, even though, strictly speaking, only their position relative to the second group is established. Thyestes exhibits another jump in its use of mid-line sense-pause (54.5%) and a significant increase in the use of a shortened final -o (Fitch 1981: 303–305). Tacitus, ann. 14.52.3, describes how, before ad 62, Seneca’s poetic output (carmina) had increased once Nero had taken a liking to it. While carmina may refer to epigrams and other poetic forms, it could equally refer to a return to tragedy (see Tarrant 1985: 12f.), and Tarrant uses this comment to give an approximate range for the play of ad 60– 62. Nisbet denies this—it is “a false clue” (1995: 296)—but nevertheless sees in Thyestes a series of historical allusions that put its composition at ad62 (1995: 300–309). Given that Seneca’s retirement may have been a gradual process (cf. ann. 14.53.1–57.2), distinguishing between these positions is difficult, and any date close to ad 62 remains possible. Perhaps the most distinctive result of Fitch’s metrical analyses is the confluence of two measures in determining a late date for Phoenissae. The play has the highest incidence of sense-pause within a line (57.2%), and, by some margin, the highest incidence of shortened final -o. Both of these strongly indicate Phoenissae was the last tragedy Seneca composed, a conclusion corroborated by its apparent incompleteness. If the argument for ascribing Thyestes to ca. ad 62 is accepted, then Phoenissae would date to the final years of Seneca’s life, as Seneca chose a mythical subject that attracted both Euripides and Sophocles in the final years of their lives (Nisbet 1995: 309). This has bearing on the date of Hercules Oetaeus, a play included among the Senecan tragedies alongside the certainly spurious Octavia. The play has been defended as authentic by Rozelaar (1985) and Nisbet (1995: 209–212), who treat it as a late play, composed “shortly before Seneca’s death in 65; that would explain the anomalies, the verbosity, the other signs of haste” (Nisbet 1995: 210). Several indications tell against this. The incidence of shortened final -o is very small, as Nisbet admits (1995: 310), and the verbose nature of Oetaeus, and its sheer length, suggest a fundamentally different approach to playwriting than that suggested by Phoenissae: both are unlikely to have developed from the same author composing at the same time in his life. Hercules Oetaeus is therefore not by Seneca,21 and may date as late as the early second century, as suggested by Zwierlein (1986b: 313–343). Three dates have been argued for the inauthentic Octavia: several scholars have argued for a date in ad68 and the reign of Galba;22 Ferri cites

21 22

See Leo 1878: vol. 1, 48–74, Friedrich 1954, and Axelson 1967. Kragelund 1982: 38–52 and 1988, Barnes 1982, and Wiseman 2001: 10 and 14.

the works of seneca the younger and their dates

parallels between the play and certain poems of Statius, which, if the directionality of the reference is correct, would argue for a date in the 90s (2003: 5–30). The early years of Vespasian’s reign (ad 69 to the mid-70s) is also possible (Junge 1999: 199 f., Smith 2003: 426–430, and Boyle 2008: xiii–xvi).23 Longer Prose Works There are four longer prose works that survive, all of which are Neronian and date to the last decade of Seneca’s life. The first, De Clementia, is dated to Nero’s nineteenth year (clem. 1.9.1–2), i.e., sometime between December 15, ad55 and December 14, ad 56, and is dedicated to the new princeps.24 Originally in three books, only the first and part of the second survive. Seneca had composed Nero’s speech at Claudius’s funeral (ann. 13.3.1–2), as well as other speeches critical to securing his authority (Dio 61.3.1), and early in ad55 he wrote several speeches for Nero that had the princeps clementiam suam obstrigens (ann. 13.11.2, “pledging himself to compassion”)—the less respectful apocol. having been composed at exactly this time. It was early in ad 55, of course, that Britannicus was murdered (ann. 13.15.1–17.3), which makes the historical situation ofclem. roughly one year later all the more interesting for an understanding of Seneca’s purpose.25 Clem. is an overtly political work, and probably represents a consolidation of the ideas adumbrated in these speeches, a coherent policy statement for the new regime. The seven books of De Beneficiis are dedicated to Aebutius Liberalis, who is the subject of epist. 91 (from book 14, about which see below). The work was written between ad56 and ad64 (Griffin 1976: 399). Seneca returned to the subject of favors and ingratitude in epist. 81 (from book 10): epist. 81.3 makes clear that benef. predates the letter. As with any multivolume work, composition over time must be considered. In this case, books 1–4 appear to form a cohesive unit, and benef. 5.1.1 begins with an acknowledgment that the remaining books are of a different character. While they have the same

23 Tanner 1985 offers a radically different division of the plays based primarily on his perceptions of the performance demands. He suggests that Thyestes, Medea, Agamemnon, Oedipus, and Phaedra (and Phoenissae if produced as it survives) were composed by Seneca during his exile from ad41–49, and that the remaining plays (Octavia, Troades, Hercules Furens, and Hercules Oetaeus) are not by Seneca and were composed between ad70 and 80. Dingel 2009 appeared as this volume was due to go to press, and could not be taken into account. 24 Griffin 1976: 407–411. 25 See Griffin 1976: 134–139 for a discussion of this tension.

c.w. marshall

addressee, there is no reason they could not have been composed after the space of several years, even following the composition of epist. 81. Indeed, several works may be ascribed to the period following Seneca’s retirement in ad62, which was a particularly prolific period: we have seen that De Otio and De Providentia might belong to this period, as might Thyestes and some or all of De Beneficiis. Almost certainly Phoenissae comes from these final years, as do Naturales Quaestiones and the Letters to Lucilius (Epistulae). Hine (2006: 71) provides a conservative back-of-the-envelope calculation, suggesting a rate of composition during this period of (at least) one book every forty days, even suggesting that this was “a very modest rate of composition compared to what Cicero achieved at the end of his life” (71, note 124), who achieved an average of about one book every twenty-four days. There are reasons to believe that Seneca was in fact composing faster than this, which demonstrates that much of this final period was dedicated to literary endeavors. The text of Naturales Quaestiones is corrupt, but the work may originally have had eight books addressing various natural phenomena: seven books survive, with a clear break evident in book 4.26 The date of the work is established by reference to specific natural phenomena within it. Mention is made of an earthquake in Greece that preceded an earthquake in Pompeii by one year (nat. 6.1.13). Tacitus (ann. 15.22.2) dates the Italian earthquake to ad62, which is close to the date of ad63 suggested by Seneca (nat. 6.1.2).27 This could provide a context for the mention of earthquakes in Greece following the appearance of a comet (nat. 7.28.3), which would therefore refer to a comet that was visible from August to December of ad 60 (Ramsey 2006: 140–146). Similarly, when Seneca says “two such [comets] have been seen in our lifetime” (quales duo aetate nostra visi sunt: nat. 7.6.1), it seems certain he is referring to this comet and Claudius’s comet, visible for a month in ad 54 (Ramsey 2006: 136–140). This passage, therefore, was not revised after the appearance of another comet, in May of ad 64.28 These details combine to suggest that nat. was completed sometime between ad 61 and early 64. Seneca

Hine 1996: xxiv argues that the original order was 3, 4a, 4b, 5, 6, 7, 1, 2. Hine 1996 assumes nat. 6.1.2 Regulo et Virginio consulibus to be a gloss from Tacitus. See also Wallace-Hadrill 2003: 182, who argues for a date of late ad 61 for the Greek earthquake, and Hine 2006: 68–72, who summarizes the issue and emphasizes that there is no reason to doubt the Tacitean text. 28 The language of the sources (sidus cometes: Tac. ann. 15.47.1; stella crinita: Suet. Nero 36.1) shows that this object was thought to be a comet, even though it may have been a nova. See Ramsey 2006: 146–148. 27

the works of seneca the younger and their dates

may have begun it earlier, and some relationship doubtless exists between passages here and several lost works, including De Situ et Sacris Aegyptiorum (T 19 Vottero), De Situ Indiae (T 20–21 Vottero), De Motu Terrarum (T 55 Vottero), and De Forma Mundi (T 56 Vottero). Nat. is dedicated to Lucilius Junior (PIR2 L 388), as was prov. Lucilius was a slightly younger contemporary of Seneca (he is in senectute in epist. 19.1, and see 96.3), who in the early 60s attained a procuratorship in Sicily. Lucilius is also the recipient of the collection of Stoic letters, Epistulae, which survive as 124 letters divided into twenty books, although Aulus Gellius 12.2.3 refers to a non-extant letter in book 22, which shows that the surviving collection is incomplete and that its precise original size cannot be determined. The book format is important for interpreting the letters, although modern editions typically obscure this aspect (Wilson 2001). Several letters mention real events, and while some historical details may be included to create a sense of dramatic moment, isolating the dramatic date for the letter from the actual time it was composed in the tradition of Athenian philosophical texts is problematic. Many details refer to comparatively personal issues in the lives of Seneca and Lucilius that cannot serve this function for a broader readership: a lawsuit, a new book by Lucilius, personal illness, retirement, and so on. These are of a different order than the reference to the fire at Lugdunum (Lyons) in epist. 91 (dated to July 64 by Tacitus, ann. 16.13.3), for example, which could more easily be used to establish a dramatic date, if that were Seneca’s intention. Nevertheless, the rate of composition suggested by the letters clearly points to an inherent artificiality in the nature of the correspondence between the two, indicating “not only that he failed to wait for a reply before writing (as he does in 118.1 as a concession), but that he sometimes sent letters, not individually, but in packets” (Griffin 1976: 418). Seneca knew he was writing for publication (epist. 21.5), and it is likely that there was some editorial work introduced either by him or by someone else soon after his death, which means that any apparent allusion to real events may serve multiple unrecoverable purposes. References to the passing of seasons do coincide with a relatively tight sequence for the letters between the autumn of 63 and Seneca’s death in April 65 (Griffin 1976: 347–353 and 400), with some books appearing for the public “perhaps in the latter part of 64” (Griffin 1976: 349). From this Griffin concludes that Lucilius’s spiritual development, described in the letters, must be fictional, although this cannot be taken as certain. There are, of course, other works of Seneca that no longer survive. Martial 7.45 implies that there was a collection of letters sent to Caesonius (or Caesennius) Maximus, a friend who had accompanied Seneca during his exile

c.w. marshall

(see also 7.44). There are no means by which to date the lost De Matrimonio (T 22–F 54 Vottero), De Superstitione (T 64–F 75 Vottero), or the life of his father (De Vita Patris, F 97 Voterro). In his account of Seneca’s suicide, Dio (62.25.2) has him revising a book and leaving it, and others, with his friends. It is not known which works these are. Tacitus, too, emphasizes that Seneca was composing to the very end, even though he had cut himself repeatedly to increase the flow of blood: et novissimo quoque momento suppeditante eloquentia advocatis scriptoribus pleraque tradidit—“and, even at the very last moment, his eloquence in full supply, he called his scribes and dictated a great many things” (ann. 15.63.3).

TRANSMISSION

Rolando Ferri Quintilian states, in his catalogue of Seneca’s works in inst. 10.1.125–131, that he treated all fields of study. He lists orationes, poemata, epistulae, and dialogi, probably omitting the scientific works as irrelevant for an appreciation of Seneca’s eloquentia.1 Virtually nothing survives of Seneca’s activity as an orator (Testimonia 2–13 Vottero), but the other three genres are well represented among his extant writings. Yet it is estimated that about half of Seneca’s literary output may have perished. After his lifetime, Seneca’s fortune seems to have undergone a period of bad press, mainly at the hands of the “archaizing” writers, who disapproved of his “modernist” style, the banality or plainness of his lexicon, and his censure of the early writers, as witnessed by Gellius (12.2.1).2 Seneca’s works, however, remained in some fashion, as shown by the fourth-century forged correspondence with St. Paul: even the comparatively illiterate Faustus, the Manichaean master, made a point of knowing his Seneca (Aug. conf. 5.6). Seneca never entered the canon of “core” school authors: lexicographers and grammarians hardly ever quote from the prose works3 and his name is absent from the list of school readings in Hermeneumata Celtis, ll. 38f. Dionisotti.4 Priscianus, alarmingly, even mistook the Elder for the Philosopher.5 However, quotes from the tragedies are more frequent in works by grammarians, commentators, and Christian authors—a fact that may suggest inclusion in some advanced school syllabi at least.6 1 In addition to nat., Pliny and Servius record a De situ et sacris Aegyptiorum (T 19 Vottero), and a De situ Indiae (T 20 f. Vottero). Cassiodorus owned a copy of De forma mundi (T 56 Vottero), and Seneca himself refers to his De motu terrarum in nat. 6.4.2. 2 Trillitzsch 1971: 69–75 Holford-Strevens 2003: 276 f. 3 Except Diomedes, who refers to the “dialogue” De superstitione, GLK 1.379.17, and to De officiis, GLK 1.366.11. 4 Dionisotti 1982: 100. 5 Vottero 1998: 15 f. 6 E.g., Diom. GLK 1.511.23 [Med. 301]; Ps.-Probus, De ultimis syll. GLK 4.224.20 [Hecuba = Tro. 861]; and Prisc. inst. GLK 2.253.7 [Phaedr. 710 and Ag. 365]. Christian authors mentioning or imitating the tragedies include Augustin, Jerome, and Ennodius (cf. Trillitzsch 1971: 379, 386). Ennodius, in particular, in Libellus pro synodo 38, presents a quote from Medea as a recollection from his juvenile readings.

rolando ferri

On the other hand, Christian writers concentrate, predictably, on the treatises that are more compatible with Christian doctrine and morals, above all on the protreptic Exhortationes (F 76–89 Vottero), on De superstitione (F 65–75 Vottero), and on De matrimonio (F 23–54 Vottero). Unfortunately, echoes from extant prose works are rare and do not enable us to identify strands of textual tradition in antiquity different from what we have in the medieval period.7 Three late-antique fragments containing Senecan works have come down to us. A palimpsest codex of Biblical content, the Biblioteca Apostolica Vaticana, Pal. Lat. 24, was put together perhaps in the seventh century from leaves taken from ancient books by different classical authors. The lower script of six of these, dated by E.A. Lowe to the third or fourth century,8 preserves barely legible fragments of the otherwise unknown Quomodo amicitia continenda sit and the initial words of De vita patris (F 58–60 and F 97 Vottero). Just before the second fragment, a tantalizing scribal subscription by one Nicianus, not otherwise known or identifiable, yields proof of some editorial activity on Seneca. A fragment of the tragedy Medea has recently been published: it belonged to a small (120 ×180mm, the size of an OCT), early vellum codex, perhaps from the fourth century (P. Mich. inv. n. 4969).9 Also late-antique and re-used is Ambrosianus G. 82 sup. (= R in modern editions), five leaves of which, in a fifth-century capitalis rustica, transmit in the lower script passages from Medea and Oedipus. Both late-antique MSS, however, are difficult to relate to the two known branches of the tradition of the tragedies, E and A (see infra).10 Among the last important instances of Seneca’s Nachleben at the end of antiquity are Martinus Bracarensis’s imitations and excerpts from Seneca in the sixth century (most notably his De ira, and the Formula vitae honestae, perhaps an abridgement of the lost De officiis).11 The last to quote the Ad Lucilium was Gregorius Magnus, in a letter written in ad591 (MGH Epistulae

7 Tertullian, De anima 20 has a verbal quote from benef. (4.6.6), with only minimal differences, which might be simply memory slips (omnium aetatum, omnium artium semina] o. artium et aetatum s. Tert.). The passage of De ira quoted by Lactantius, De ira Dei 17.13 (= Sen. dial. 3.2.3b Reynolds) comes from the initial lacuna of A (see infra). 8 Lowe 1964: 106. 9 Publication in Markus and Schwendner 1997. 10 Zwierlein 2004: I 263–266. 11 Trillitzsch 1971: 211–221, which also mentions other sententious collections of dubious authenticity.

transmission

1, p. 47).12 After that, for a space of two centuries (7th–8th cent.), no evidence survives indicating either knowledge or copying of Seneca’s works. The earliest surviving medieval MS of a Senecan text is also the archetype of the tradition of both De beneficiis and De clementia, Biblioteca Apostolica Vaticana, Pal. Lat. 1547, identified by the siglum N, for Nazarianus, with reference to the Monasterium Sancti Nazarii of Lorsch, where the MS was hosted until the fifteenth century before being moved to Heidelberg and then, after the Catholic capture of the city in 1622, to Rome. N has been the subject of extensive study and has been shown to originate from Northern Italy, perhaps Milan, where it was copied around the year 800. A copy of N, R (BAV, Reginensis Latinus 1529), appearing not much later than the parent MS, was soon moved north to France and was the origin of all subsequent tradition.13 Several layers of correction are identifiable in N, perhaps when the MS was being prepared for copying. Winterbottom 2001 has suggested that not all corrections of R on N may have been the result of conjecture. Not much later than Nazarianus are the earliest extant MSS of the Epistulae morales ad Lucilium, from the beginning of the ninth century. The ninth century is when indirect evidence of knowledge of the letters also reappears.14 The tradition, initially, is divided, and two different groups of MSS transmit letters 1–88 and 89–124.15 Reynolds has argued that the divide goes back to an ancient dispersal of the Letters over several codices, certainly two, but possibly three, because in some MSS letters 1–52 also seem to form a separate codicological unit. In addition, the fact that Gellius 12.2 quotes a passage from book 22 (our MSS reach only book 20) suggests that there was at least another ancient ‘tome’ of Seneca’s letters to Lucilius and that it did not survive antiquity. Reynolds has located the spheres of influence for the two groups in France (1–88) and Southern Germany (89–124). The earliest MS in which the two groups are joined is Q, Brescia, Biblioteca Pubblica Quiriniana B.II.6.

12 Ut tibi aliquid saecularis auctoris loquar: cum amicis omnia tractanda sunt, sed prius de ipsis (= Sen. epist. 3.2, tu vero omnia cum amico delibera, sed de ipso prius). 13 The fullest account of the tradition of De clementia is in Malaspina 2001a, who has also championed the thesis that one of the recentiores, Q, can claim direct descent from N. For De beneficiis the most recent edition is that of Préchac 1926. 14 Walahfridus Strabo transcribed letter 120 in a miscellaneous MS, Sangallensis 878, f. 348r, ca. 809, and Paschasius Radbertus of Corbie quotes epist. 10.2 in De fide spe et caritate, ca. 846, PL 120.1442. 15 According to Reynolds 1965a: 56–65, the latter group derives from a late-antique uncial MS from the fifth or sixth century.

rolando ferri

The Naturales Quaestiones survive in about 100 manuscripts, all of fairly recent date. At the end of antiquity, the work was known to Ammianus Marcellinus16 and a brief Greek paraphrase survives in the work De mensibus by Iohannes Laurentius Lydus, a Constantinopolitan historian and antiquarian. Lydus’s paraphrasis is particularly important because he was using a codex more complete than the archetype of our tradition. The earliest MSS ofnat. are not older than the twelfth century, with the single exception of a ninthcentury florilegium written in Brittany or in the Loire region (Y, Munich, Bayerische Staatsbibliothek, Clm 18961-II). The most recent editor of the work, H. Hine, followed by Parroni 2002, has divided the MSS in three families, YζΨ, but M.D. Reeve has queried the solidity of ζ, which is weakly defined by errors that were easy to correct and that may therefore have been corrected in the other branches independently.17 The order of the books differs greatly in the MSS: that of the archetype is reconstructed by Hine to have been III–X, suggesting a loss of two initial books. A great lacuna has also occurred at the archetype level in Book IV, De Nilo, and, following Haase, editors use the sigla 4a and 4b for the two remaining fragments.18 We owe to the library at Montecassino the survival of the twelve so-called Dialogi. The identification of the eleventh-century Ambrosianus C 90 inf. (A) with the “Seneca” made to copy in Montecassino by Abbot Desiderius in a notice going back to 1058–1087 (Chronica Monasterii Casinensis,MGH SS 34, III.63, p. 446 Hoffmann) has not been challenged and A appears to be the archetype of the majority of the MSS of the twelve dialogues.19 Thereafter, knowledge of the Dialogi spread slowly north and copies multiplied. A may have been transcribed from a very early MS: some indication that its exemplar was perhaps a late-antique book is the lay out of the list of contents on 3r. The recentiores have been divided by the most authoritative editor, L.D. Reynolds, into two classes. One, β, descends in toto from A and becomes useful only where A is illegible or has subsequently suffered textual loss after being copied. A second, less numerous, class of recentiores, γ, appears to descend not from A but from a twin, possibly also of Montecassino origin.20

Gercke 1895: 99–103. Reeve 2000: 202. 18 Lydus’s Greek paraphrase of a portion of the text from the medieval lacuna is reproduced by Hine on pp. 187–189 of his 1996 edition. 19 At about the same time, imitations from the Dialogi are recognizable in two saints’ lives composed by Guaiferius Salernitanus (PL 147.1293–1310), who was active in the same area and lived at Montecassino. 20 A further problem is posed by the beginning of De Ira, 1–2.3, where A left f. 14r blank, probably because the exemplar was illegible; a later, twelfth-century hand (a) supplied the 17

transmission

The tragedies are transmitted by two independent branches. The more authoritative, the ε-family, is represented in its pure form by the single eleventh-century codex Etruscus (Florence, Laurentianus 37.3), identified by Giuseppe Billanovich with the tragedies listed in a catalogue of classical books owned by Santa Maria della Pomposa, near Ferrara, dated from 1093. An excerpt from the tragedies in a ninth-century MS, perhaps written in Fleury, now Paris (Bibliothèque Nationale de France, lat. 8071) the Florilegium Thuaneum, was traditionally related to the same family.21 However, the εfamily produced no offspring until E was rediscovered by Lovato de’ Lovati and brought to the attention of the Paduan pre-humanist. However, E later fell into oblivion; the first modern editor to use it was J.F. Gronovius, who saw it in Florence in 1640 while preparing his edition of 1661. The second, so-called “interpolated,” family, A, encompasses by far the majority of the 400-odd known MSS. It is French in origin and must have come from a MS that resurfaced probably at the end of the twelfth century. The two earliest representatives of the further two branches in which A is subdivided, δ and β, are, respectively, P (Parisinus Latinus 8260) and C (Cambridge, Corpus Christi College 406). It is clear that the two families depend on two late-antique subarchetypes and that the entire tradition is ultimately derived from the same, corrupt archetype. The tradition of the satirical Apocolocyntosis rests on about forty MSS, of which only three are used by editors, that is S, Sankt-Gallen 569, V, Valenciennes 411, and London, British Library Additional 11983. The tradition is bipartite, with V and L forming the second family. Only S and L produced copies. Knowledge of the Apocolocyntosis began to spread at about the same time as the earliest MS, S: an unmistakable echo of apocol. 1.2f. is found in Paschasius Radbertus’s Vita Walae or Epitaphium Arsenii, written after 846 (PL 120.1563), even if, curiously, Seneca’s satiric element is completely missed.

text printed in modern editions, up to capitis damna. This means that in the twelfth century a twin or even the exemplar of A was still available. However, the a-text is different from both β and γ. This means that, for this section, the parent of β also drew on a MS more complete than A. 21 For a different view, namely that the Florilegium Thuaneum goes back to a non-bifurcated stage of the Seneca tradition, see Brugnoli 2000b.

PART ONE

LIFE AND LEGACY Seneca and Senecae: Images of Seneca from Antiquity to Present

SENECA THE PHILOSOPHER*

Matthias Laarmann

1. Anomalies of Fame; or, Seneca’s Journey through Time It is necessary to keep considerable prosopographical and biobibliographical peculiarities constantly in mind in order to be able to appraise the legacy of Seneca the Younger from antiquity to the present day in an adequate fashion.1 First, already in the Early Middle Ages, the distinction between Seneca rhetor (Seneca the Elder) and Seneca philosophus (Seneca the Younger) was lost. Only the lawyer Andrea Alciati (1492–1550) and ultimately Justus Lipsius (1547–1606) and the Jesuit Andreas Schott (1552–1629) were able to regain this knowledge (van der Poel 1984: 262–264). Second, early humanists—the first was probably Giovanni Boccaccio (1313–1375) around 1365—distinguished between a Seneca moralis, to whom the prose works were ascribed, and a Seneca tragicus, who was regarded as the author of the tragedies (D’Alessi 1978). The fact that Seneca tragicus was identical with Seneca philosophus was proved by the Jesuit Martin Antonio Delrio (1551–1608) and the Latin philologist Isaac Pontanus (1571–1639; see van der Poel 1984: 264–266). Third, the widespread fame Seneca enjoyed from the early medieval period until the Renaissance was based on texts, above all collections of sayings, that were not in fact written by Seneca but ascribed to him due to a closeness of content and style (among them was even a collection of aphorisms of Arabian-Oriental origin; see Blüher 1969: 54f.). Fourth, Seneca’s journey through the Middle Ages would be unthinkable without taking the history of the reception of the apocryphal exchange of letters between him and Paulus into account, a correspondence that gave a considerable boost to a positive perception of

Translated by Tobias Budke. Overviews can be found in Summers 1910: xcvi–cxiv; Faider 1921; Blüher 1969: 13–175; Spanneut 1973; Ross 1974; Spanneut 1980; Armisen-Marchetti 1989; Spanneut 1990; Dionigi 1999; Trovato 2005; Walter 2006; Carron 2007; Kraye 2007. Due to limited space, this chapter will focus mainly on the reception history in German-speaking areas. 1

matthias laarmann

Seneca-like philosophizing. However, this exchange and its reception history will only be touched upon in this chapter (for more detail on this, see Fürst, infra, pp. 213f.). 2. A Reevaluation of Evaluation; or, Ancient Christianity Bucking the Trend Already during his lifetime, Seneca became the darling of the public and a writer whose subject matter fascinated readers of all ages—who saw it as being of topical interest and vitally important—and whose brilliant style evoked enthusiastic reactions especially among young readers (Trillitzsch 1971). But, near the end of the 1st century ad, Seneca’s critics mustered and intensified their moral and above all stylistic objections to his works. This increasingly critical and hostile attitude regarding pagan culture bearers from antiquity could be discovered by Latin-speaking Christians of the first two centuries ad, for example, in the works of Quintilian (inst. 10.1.126), Fronto (Ad M. Antoninum de orationibus 2–3), or Aulus Gellius (12.2; see Kraye 2007: 826–829). Nevertheless, Seneca did not disappear completely from the canon of literature and education. This was indirectly attested by Augustine (conf. 5.6.11), who admitted that his opponent, the not very well educated Faustus of Riez, was at least acquainted with various speeches by Cicero and also with some writings by Seneca. “But at the very moment when the stylists rejected Seneca’s language, the Christians learned to appreciate his subject matter” (Summers 1910: xcviii; regarding the subjects that were adopted, see Spanneut 1957). Tertullian († ca. 220), already inclined to Stoic philosophy, cried out, full of enthusiasm: Seneca saepe noster (De anima 20.1. CCSL 2: 811, 3). Lactantius (†325), who had access to many of Seneca’s writings now lost (Exhortationes, De immatura morte, Libri morales, or Libri moralis philosophiae, resp.; Lausberg 1970) even conjectured that Seneca would have converted to Christianity, if only someone had conveyed to him the message of the Gospel (inst. 6.24.14; Spanneut 1990: 584 f.). The Christian declarations of sympathy reached their peak with Jerome (347–420), who included the Roman philosopher in his catalogue of Christian writers, De viris illustribus, and who ascribed to him a correspondence (actually apocryphal) with Paul the Apostle (see infra, pp. 213f.). In this, Jerome assumed a close affinity between the Stoics and Christianity: nostro dogmati in plerisque concordant (Comm. in Esaiam 4.11. PL 24: 147). His friend Augustine (354–430), although he displayed only a passing interest in Seneca’s sayings in his works, introduced him nonetheless between 413

seneca the philosopher

and 415 as a critic of pagan polytheism (civ. 6.10f.) and accepted Jerome’s thesis concerning the correspondence between Seneca and Paul (epist. 153.14. CSEL 44: 412). The only place where a principal rejection of Seneca based on the pagan character of his thoughts can be found is a poem of the Anthologia Latina that has been handed down anonymously and has sometimes been ascribed to one Honorius scholasticus from the 6th century (nr. 666. ed. Riese 18942; see Trillitzsch 1971: I 191–193; II 385f.; Fohlen 2002: 52f.). Generally speaking, though, it must be stated that, for all the sympathy he evoked from Christians, Seneca’s philosophy never became a determining factor in Christian theology (see Herrmann 1979; Krefeld 1992b; Fuhrmann 1997: 329–340). 3. A Merry-Go-Round Exchange; or, Compilation and Confusion between Antiquity and the Middle Ages The premodern concept of the “author” makes the well-nigh infinite willingness of the time to compile and quote without giving any sources understandable. Regarding the reception history of Seneca’s works, this was to become their biggest advantage. For more than a thousand years, Bishop Martin of Braga (ca. 515–580) became a relay station for Seneca’s reception (Blüher 1969: 24–28; Trillitzsch 1971: I 211–221). From Seneca’s De ira, Martin extracted a treatise bearing the same title. From Seneca’s De officiis, now lost, he compiled a writing that became known either asFormula vitae honestae, De quattuor virtutibus, or—complemented by sentences from the Epistulae morales—as De copia verborum (Trillitzsch 1971: II 393–399). As the more than 600 extant manuscripts show, it became immensely popular during the Middle Ages and was at times considered one of Seneca’s main works; until the 17th century it was seen as a basic presentation of Christian education. (Bickel 1905a; Fohlen 1980; Spanneut 1990: 586f.; Orselli 1999; Torre 2006). This confusion led to profusion. A large number of florilegia (Meerseman 1973; Munk Olsen 2000; Walter 2006: 135f.), helped by Seneca’s attractive aphoristic-sententious style, ensured a high presence of Seneca’s or Senecalike sentences, but without providing a clear grasp of the methodical significance of his philosophical concept as a whole. Among these were, for example, the Florilegium morale Oxoniense (ed. Talbot, C.H. Analecta Mediaevalia Namursiensia 1956) and the Auctoritates Aristotelis, Senecae, Boethii … (ed. Hamesse, J. 1974. Philosophes Medievaux 17: 273–286). Substantial excerpts were provided by Roger Bacon (1214–1292/94) in hisOpus magnum

matthias laarmann

(ed. Bridges, J.H. 1900: 299–365), and it was Bacon who, incidentally, in 1246 rediscovered Seneca’s Dialogi and ensured their circulation. Among the encyclopedists, Vincent de Beauvais’s (1184/94–ca. 1264) Speculum historiale VIII, chap. 102–135 (written ca. 1247–1260; ed. Douai 1624: IV 309–320; Ruhe 1969) occupied a central position due to its large number of quotations. In any case, the extremely positive resonance Seneca had with his Christian readers eliminated almost all of the negative judgments pagan antiquity had passed upon him because of his style (be it literary or life) until, not before the 15th century, Italian humanists rediscovered the writings of Tacitus and Cassius Dio, the most important authors to pass on the ancient tradition of adverse criticism. 4. Progressive Christianization; or, Incorporating Seneca into Spiritual-Monastic Theology and Christian Ethics of Government In the 12th century, the “Christianization” of Seneca was widely established (Nothdurft 1963; Spanneut 1964; Lapidge 1988; Töpfer 1996; Smiraglia 2000). For Peter Abelard (1079–1142), Seneca was maximus ille paupertatis et continentiae sectator et summus inter universos philosophos morum aedificator (epist. 8. PL 178: 297B). In the opinion of Godfrey of Saint-Victor (1125/30– ca. 1185), Seneca’s teachings came a close second to the Gospel itself (Fons philosophiae 410–412: Quid tibi de Senece documentis edam? / Seneca Lucilio commendavit quedam, / que vix evangelio postponendam credam), causing Walter of Saint-Victor (†1180) to reply to this high praise in a work called De blanda et ideo mortifera Senecae doctrina (ed. Mastandrea 1988: 80– 83). William of St. Thierry (1075/80–1154), Alanus ab Insulis (ca. 1120–1202), William of Conches (ca. 1180/90–after 1154) and especially John of Salisbury (ca. 1115/20–1180)—incidentally one of the very few medieval authors to defend Seneca’s style against his critics from antiquity (Kraye 2007: 830)— made similarly emphatic statements (Smiraglia 2000: 272–274). Our Roman philosopher was enjoying a huge success even in the sermons written during that era (Spanneut 1990: 588). The theological-spiritual reception reached a climax with the Epistola ad fratres de monte Dei (ed. Dechanet 19832; Middle High German translation ed. Honermann, V. MTU 61. 1978) by William of St. Thierry (Déchanet 1951). A later echo can be found in the works of Meister Eckhart (ca. 1260–1326). He adopted Seneca’s dictum Nulla sine deo mens bona est. Semina in corporibus humanis divina sunt (epist. 73.16) several times to his theory of unio mystica (Expos. libri Sapientiae nr. 52. LW II 379,4–12; Liber

seneca the philosopher

parabolarum Genesis, nr. 198. LW I/1 670,13–671,9; Von dem edeln Menschen. DW V 111,9–21; see also Bray 2008: 177–181). Seneca’s doxographic overviews of the Platonic teachings regarding ideas and the Aristotelian teachings regarding causes in his Epistulae morales 58 and 65 (cf. Theiler 1934: 1–10, 15–26, 34–39; Long and Sedley 1987: No. 27A with commentary; Krefeld 1991: 5–58) enjoyed a first intensive reception in the 12th century in the works of Achard of St. Victor (1100–1172), De unitate divinae essentiae et pluralitate creaturarum (ed. Martineau 1987), as well as in those of Thierry of Chartres (1085–1155) and John of Salisbury (Nothdurft 1962: 182–191), and gained an additional boost at the end of the 13th century, influenced by Robert Kilwardby (ca. 1215–1275) and Henry of Ghent (1217– 1293), Quodlibeta 7.2 and 9.2 (Hübener 1977; Laarmann 1991: 57f.). Regarding the phrase aliquid, quo nihil maius cogitari posit, of central importance for Anselm of Canterbury’s (1033–1109) ontological proof of God’s existence, it is possible to consider a link to phrases originating with Seneca (nat. I, praef. 13; epist. 58.17; Nothdurft 1962: 192–201; Vinti 1979; Laarmann 1991: 147 f.). That Seneca was easily accepted into the Franciscan school is well attested by Bonaventura (1221–1274; see Rivier de Ventosa 1965). Seneca, the tutor of princes, was also given a permanent place in the “Mirrors of princes” genre (Spanneut 1990; 588 f.). The conservative Franciscan theologian Guibert de Tournai, in his 1259 Eruditio regum et principum 3.3 (de Poorter 1914: 200–222; partial German trans. in Anton 2006: 288–447, esp. 432–435), for example, referred with many variations to Seneca’s concept of clementia, which he modified according to the perspective of the Christian theory of affectus pietatis. 5. Seneca Disputatus; or, the Oppressive Dominance of Aristotelianism in Academic Scholasticism The university was invented in the Christian Middle Ages. However: “Nothing is more distant from our university and school teaching, itself cut off from life, than ancient philosophy, which understood itself as ‘ars dicendi’ and ‘ars vivendi’ at the same time” according to the trenchant judgment of Pierre Hadot (1989c: 799). How was Seneca going to find his way around the structure of this new and institutional form of philosophy? The scholastic philosophy of the Middle Ages was ruled by Aristotelianism. Term, conclusion, judgment were the tools of logical thinking. With Thomas Aquinas (1224–1274), who quotes Seneca above all in his doctrine of the affections (Verbeke 1961, 1983; Spanneut 1990: 590f.), as well as with all

matthias laarmann

other scholastic writers, Seneca’s stimulating, psychagogically aimed rhetoric as an integrated component of a therapeutic directive to achieve the ascetic self-control of an ethically educated mind, unaffected by Fate and Death and given to leisure, fell by the wayside as far as interpretation was concerned. What was preferably sought after—in a process of hermeneutically questionable adaptation—were terms equivalent to the established terminology of Aristotelian philosophy, as can be seen in the Super X libros Aristotelis ad Nicomachum 10.4 (ed. 1489, fol. 209ra–211 vb) of John Buridan (ca. 1300–ca. 1358), a student of Ockham, who at length confronted an Aristotelian virtue ethics transformed by Christianity with the Stoic virtue ethics of Seneca (Walsh 1966; Korolec 1979). As a reader-friendly effect of the scholastic-formal treatment of problems, which, at the same time, revealed the disparity of Senecan and scholastic philosophizing (in the sense used by Hadot, P. 1989c: 799), one must regard the lemmatization of Seneca’s texts in register-like tabulae (Fohlen 2002: 19–22). A remarkable instance of such a systematic access to the texts was provided by the Dominican Lucas Mannelli (†1362), the Bishop of Osimo. To Pope Clement VI (1341–1352) he gave a work including a dedication, a prologue, and two series of lemmas (Abstinentia-Iuuenis et Labor-Ydea) (Kaeppeli 1948; Munk Olsen 2000: 174f.; Fohlen 2002: 67f.). This alphabetically arranged Seneca florilegium was later translated—to give only one example—into Spanish, probably at the instigation of King Martin I, by Alonso de Carragena (1385/86–1456), the Bishop of Burgos and a well-known representative of scholastic theology (Blüher 1969: 98f., 101f.). Typical of the medievalscholastic treatment of Seneca was the constant production of commentaries on the texts and florilegia, stemming from a theologically sophisticated knowledge of the differing Christian aspects, not indulging in unconsidered affirmation and undertaking an ongoing Christian rectification of Seneca’s ancient autosoteric views. Insofar as in this way any kind of unreserved and uncritical hero-worship of ancient and pagan writers—something that had been sometimes observable since early humanism—was opposed, the scholastic reception of Seneca finds itself unexpectedly close to the position held by modern historical and critical Seneca philology, which emphasizes the epochal alterity of the autosoterics of ancient-pagan, especially Seneca’s, thinking in contrast to the theosoteric fundamental position of Christianity (Fürst 2006: 85–107; Fuhrer 2006: 108–125).

seneca the philosopher

6. Praise and Reprimand; or, Bolstered Fame and the Seeds of Criticism in 13th–15th-Century Humanism Early humanism prepared the ground for Seneca’s true aurea aetas. Even if the paintings of Peter Paul Rubens may come to the mind of the modern observer far more easily and quickly, the unsurpassable visual culmination point of the history of Seneca’s reception and influence for Seneca philosophus was doubtless the illustration on the cover of this volume showing the Codex Glasgow, Hunterian Library MS 231, created by Roger of Waltham (†1336), a canon of St Paul’s, London, who served as King Edward II’s Keeper of the Wardrobe from 1322–1323: three people, anachronistically clad in garments worn by university professors—unacceptable to the welleducated Seneca, who held self-centered, sterile book knowledge in contempt (see, for example, epist. 98)—are holding books containing significant theological theses in their hands. They are the two giants of classic Greek philosophy, Plato on the right and Aristotle on the left. Seneca—and not, as one would expect, Socrates—occupies the center! Through the arrangement of quotations for Plato and Aristotle, the orientation toward an ideal of theory can be perceived, while Seneca is a witness for the prioritization of ethics according to the Hellenistic schools of philosophy. The parade of the early humanist laudatores Senecae was, in chronological terms, led by Dante (1265–1321), for whom Seneca was above all “Seneca morale” (Mezzadroli 1990; Pasquini 1999). Around 1365, Giovanni Boccaccio (1313–1375) developed his thesis, inspired by an epigram of Martial (1.61.7) and propagated through his commentary on Dante’s Divina commedia, of the necessity of distinguishing between a Seneca rhetoricus and a Seneca tragicus (ad Inf. 4.141, ed. Padoan 1965: 252), a viewpoint shared in 1371 by Coluccio Salutati (epist. 3.8, ed. Novati 1891: I 150–155), who referred to Sidonius Appolinaris (carm. 9: Ad Felicem, vv. 230–238)—an error that was to survive for more than two centuries (Martellotti 1972; D’Alessi 1987). Albertino Mussato (1261–1329) and Giovanni Colonna (†1348), one of Petrarch’s patrons, provide the first instances of the claim that Seneca had not only been a friend of St. Paul the Apostle but also a Christian himself (Momigliano 1950: 70f.; Sottili 2004: 676–678). Seneca’s works turn up several times in the list of favorite books (libri mei peculiares) drawn up by Petrarch (1304–1374; see Ullman 1923; Bobbio 1943), with De brevitate vitae occupying a special position insofar as Petrarch found there the main arguments in favor of the vita contemplativa in his discussion of otium-negotium (Enenkel 1999). The Italian early humanist and theoretician of epistolography Gasparino Barzizza (ca. 1360–ca. 1431) was starting to hold lectures on Seneca in Padua

matthias laarmann

The texts read: Plato (left): Genitorem et opificem universitatis tam est difficile invenire, quam inventum digne profari (from Chalcidius, Commentary on the Timaeus of Plato); Seneca (center): Si vis deo propinquare, bonus esto. Si vis habere honorem, dabo tibi magnum imperium imperare (see for this Briggs 2008: 32–34); Aristotle (right): Prima causa est nobilissima, quae non alteratur nec mutatur, set (= sed) manet in sempiternum completa et perfecta (from Aristoteles, De caelo, and Averroes thereon).

seneca the philosopher

around 1407 (Panizza 1977 and 1984; Gualdo Rosa 2009). The philological and interpretative zeal of the humanists produced—as far as we know—at least five commentaries on the first epistle of the Epistulae morales (Fohlen 2002: 23). Vitae of Seneca also proliferated.2 The Plutarch-inspired parallel biography Vitae Socratis et Senecae (ed. Baldassari and Bagemihl 2003: 164–287; the part covering Seneca: 234–287), written in 1440 by Gianozzo Manetti (1396–1459), defended Seneca in detail against the criticism leveled against him above all by Tacitus, Quintilian, Gellius, and Petrarch (no. 29). A rare example of Christian-theological frontal criticism brought forward in a polemical tone can be found in the 15th-century French manuscript version of Monitum contra Senecae epistulas, in which the insurmountable incompatibility of Seneca’s thought with the Catholic faith is claimed: Ammonendus est, lector, ne hoc uolumen epistolarum Senece indiscrete et absque magna cautela quasi alicuius catholici uiri opus percurrat. […] Ipse enim auctor, licet ingeniosus et disertus, catholice tamen ueritati ignarus in plerisque locis a ueritate fidei inuenitur plurimum discrepare.[…] Diligencia adhibenda est quatinus preciosum a uili separetur. (ed. Fohlen 2002: 24 n. 81; 69)

Seneca was admitted to the literary canon of the Devotio moderna, particularly because the founder of the Brethren of the Common Life, Geert Groote (1340–1385), quoted liberally from Seneca’s works in his own texts (Meerseman 1973; Walter 2006: 138f.). In Thomas à Kempis’s (1379/80–1471) Libri de imitatione Christi, written between 1414 and 1425 and probably next to the Vulgate Bible the most frequently printed Christian book in Latin, the author refers in several important passages to Seneca, although the latter is never named.3 2 Some examples are: Iohannes de Columna, De viris illustribus (extr.: De Seneca Lucilio, Cordubensi) (ed. Ross 1970: 555–559), Iohannes Gallensis, Compendium de uita illustrium philosophorum et dictis moralibus eorundem 4.17: De Seneca (ed. Venise 1496: ff. 211–211v), Sicco Polentonus (1377–1447), De illustribus scriptoribus linguae latinae (1437), chap. 17–18 (ed. Ullmann 1928: 463–499; cf. Fohlen 2002: 74f.), Paulus Pompilius (1453–1491), Vita Senecae (ed. Faider 1921: 281–323; cf. Fohlen 2002: 70), Petrus Paulus Vergerius, De uita Senecae (ed. Ziliotto and Vidossich 1906: 355 f.). 3 In Thomas à Kempis, De imitatione Christi 1.5 (De lectione sanctarum scripturarum; Op. omn. II, p. 12f., esp. 12, lin. 20–28), Seneca’s demand to apply special qualitative criteria when choosing suitable authors and texts for reading (epist. 2.4) is in the background. The sentence Quotiens inter homines fui, minor homo redii (1.20: De amore solitudinis et silentii; Op. omn. II, p. 36, lin. 26 f.) is a chiastically varying echo of Seneca’s phrase on the morally questionable worth of mass events (epist. 7.3: Avarior redeo, ambitiosior, luxuriosior? Immo vero crudelior et inhumanior, quia inter homines fui).

matthias laarmann

Since the 12th century, Seneca’s personal piety—not unlike Socrates’s (Laarmann 1995b)—had frequently been perceived in a positive way, from the 14th century onward almost invariably so. Following the lead of Boccaccio, who was probably the first to formulate the idea—which was to attain a wide circulation during early humanism—that Seneca must be saved from eternal damnation (see Monti 2007), Heinrich von Neustadt (Singer 1906: v. 4804) and Hugo von Langenstein, who even explicitly mentioned Seneca’s suicide (Keller 1856: 21c67), ranked him among those heroes worthy of salvation who were permitted to stay in limbo (Laarmann 1991). From the end of the 13th century onward, it is possible to trace the legend according to which Seneca was really a Christian, a legend probably invented by Albertino Mussato (1261–1329; see Sottili 2004: 676–678) and still alive during the 19th century (Momigliano 1950; Panizza 1974). The Carmelite prior John of Hildesheim (†1375) wrote the poetical Laus Pauli et Senece (epist. 33; ed. Schmidt 2005: 250f.). With Seneca regularly mentioned as an exponent of the School of Wisdom, several collections of his sayings, which contained both Seneca’s own and Seneca-like dicta, were circulating (Ochsenbein 2000). For example, the great council theologian Johannes Nider, O.P. (ca. 1380–1438), ascribed the following dictum to Seneca: “Vnd wer kein got, dennoch solt man tugent würken” (“And if there were no God, one nevertheless ought to act according to virtue”; quoted in Williams 1989: 405, and Henkel 1992: 1994).4 Thus, the late medieval Seneca receptus provided a link between the per impossibile hypotheses of the nonexistence of God found in high scholasticism and the motto Etsi deus non daretur propagated by Hugo Grotius, which marks the modern, rational-autonomous natural law (Laarmann 1995c). 7. The Struggle for Authorship and Hegemony of Interpretation; or, Seneca and His Work during Renaissance Humanism and 16th-Century Neo-Stoicism Seneca became the patron of a whole era. Erasmus of Rotterdam (1466/69– 1536)—especially during the first half of his life—displayed an enormous liking of Seneca and his style (Trillitzsch 1965 and 1971: I 221–250; Walter 2008: 130f.), as can be seen in his prologue to the 1515 edition of Seneca (epist. 325; ed. Allen et al. 1906–1958: II 51–54, Latin-German version: Fürst 2006: 74–79;

4 Hans Sachs, however, places Seneca’s way of knowing God in opposition to the Christian one (Brunner/Wachinger 1986 ff.: 2S/3248).

seneca the philosopher

German trans. Trillitzsch 1981b: 327–331; English trans.Collected Works 1989: 29, 111–222; on the interpretation, see Sottili 2004: 647–667) and even more clearly in his praefatio to the 1523 edition of Cicero’s Tusculanes (epist. 1390, ed. Allen et al. 1906–1958: V 103ff.; Trillitzsch 1971: I 223). Seneca’s presence in the discussions of humanist peace ethics found in the works of John Colet (1467–1519), Thomas More5 (1478–1535), and Juan Luis Vives (1492–1540) also made itself felt in Erasmus’s Querela pacis (1517), the first clarion call of rational pacifism (Adams 1962; Dust 1987). Vives himself recommended Seneca as a warner and a prophet for a Christianity in need of reform: Audi Senecam, hominem gentilem, Christianos edocentem, quae illum conveniebat potius a Christianis discere (De subventione pauperum, praef.; Blüher 1969: 200–218). The proponents of the Reformation saw Seneca in a positive light, although Martin Luther (1483–1546) denied, in his metaphysical analysis of human action, that there was a possibility of acting morally outside Grace and explicitly criticized a dictum by Seneca (Tischreden 2873, 2890, WA 56: 236, 32 f.; Dieter 2001: 105). But Huldrych Zwingli (1484–1531), in 1530, put Seneca on the same pedestal as Moses, Paul, and Plato and spoke respectfully of Seneca, animorum unicus ex gentibus agricola (De providentia 2.3. ed. SW 93, 95; Dilthey 1914: 155 ff.). The fact that John Calvin (1509–1564), while still a Catholic clergyman, had written a commentary on Seneca’s De clementia (ed. Battles and Hugo 1969), in which he conferred the title et philosophiae et eloquentiae columen on him—second only to Cicero—paved the way for the facilitated reception of Seneca already during the time of early Calvinism (Strohm 1999). Justus Lipsius of Brabant (1547–1606) must be considered the founder of European Neo-Stoicism (Kraye 1988: 367–374). Above all, between 1579 and 1591, as a professor of history and law in Leiden with the humanists Isaac Casaubon and Joseph Justus Scaliger, whose fame had already spread all over Europe, as his colleagues, he became a leading figure in late humanism. Already early on—between 1557 and 1559—he had developed his enthusiasm

5 For Thomas More, the active involvement with Seneca reached far beyond direct quotations; Seneca was for him a true companion in life, even in his darkest hours. More, a father of four children between the ages of one and six who had become a widower in 1511 when he was 33, married the widow Alice Middleton in the same year. During the following years of marriage, he put great emphasis, quite unusual at the time, on the best possible education and high-quality teaching of his daughters, especially in the case of Margaret, who spoke Latin and Greek fluently. In a family portrait, the original of which has unfortunately not survived and which was painted by Hans Holbein, a close friend of the family, in 1528/29, Margaret is reaching out to touch a book written by—Seneca! (Espiner-Scott 1960).

matthias laarmann

for Seneca and Tacitus from Marcus Antonius Muretus (1526–1585) in Rome in the circle surrounding the highly educated cardinal Antoine Perrenot de Granvelle (1517–1586), and he went on to produce philologically epochmaking new editions of both writers’ works.6 With these editions—the Manuductionis ad Stoicam philosophiam libri tres, published in 1605, were originally meant as the introduction to the edition of Seneca’s works—Lipsius vigorously took part in the debate about Ciceronianism as the best rhetorical style and by his determined rejection of the ancient criticism of Seneca’s style caused a wave of “Senecaism” (a term probably coined by T.S. Eliot; Kraye 2007: 826, 834f.) and “Tacitism” (van der Poel and Waszink 2009: 414) that was to last almost a full century. The dialogic work De constantia (ed. Neumann 1998), published by Lipsius in 1584 (and to the present day without a critical edition)—“the flagship of Neo-Stoicism in the sixteenth and seventeenth centuries and a bestseller of its time” (Hartung 1999: 926b)—marked the final abandonment of the Platonist and Epicurean preferences of early humanism and the consequent embracing of a neo-Stoicism (Oesterreich 1989) in which the freedomdenying concept of fatum conceived by the older Stoa is criticized and further developed toward a theory of fatum in which the ratio of man and his free iudicium enable him to assume a self-determined position relative to fatum. “Will, reason, discipline are from now on becoming dominant values of the time” (Oesterreich 1975: 183), with Seneca’s vivere militare est (epist. 96.5) as the dominant metaphor (Sommer 2008). With his Politicorum sive civilis doctrinae libri sex (ed. Weber 1998), an annotated collection of quotations from ancient literature (with 547 quotations from Tacitus alone!), first published in 1589 and, with 36 reprints, an extremely successful work, Lipsius created the counterpart to the Libri de constantia. Whereas the first work was an instruction manual for the self-preservation of the individual, the later work did the same for the self-preservation of the sociopolitical body. Also worth mentioning are the painting “The Death of Seneca” (created in 1612/13; see Brandt 2000: 226–239) by Peter Paul Rubens (1577–1640), a member of the inner circle around Lipsius, and an untitled painting that was later named “Justus Lipsius and His Friends” (ca. 1615) and shows Lipsius and Seneca together (see Brandt 2000: 240–245). Both paintings have become important paratexts of Seneca reception, especially “in an age

6 Lipsius, by the way, adorned his edition of Seneca (1605, 16152) with a dedicatory poem written by himself (ed. / trans. Schäfer 2005)—the addressee being none other than Seneca himself, of course!

seneca the philosopher

of technical reproduction” of works of art (Walter Benjamin). This Seneca euphoria, created by secular neo-Stoicism, did not stop at church doors. The Jesuit Johann Baptist Schellenberg (1586–1645) wrote the anthology Seneca christianus, id est Flores christiani ex L. Ann. Senecae epistolis collecti (Augsburg, 1637), a work that, together with its translations, was printed in an astonishing number of variant editions under different titles (Sommervogel 1890–1932: VII 741f.). 8. Translations, Adaptations, Variations; or, Seneca in the European Nations during the 17th and 18th Centuries In the same way the subject matter treated by Seneca had done before, this era saw his writings enter the vernacular languages. In Spain, there had been continuously increasing attempts at this from the 13th century onward (Blüher 1969: 94–117). In Germany, the process of translating Seneca, already well-established and documented as a widespread phenomenon during the second half of the 15th century, was intensified during the first half of the 16th century, with Dietrich von Pleningen and Michael Herr being the most important names in this context (Worstbrock 1976). As for the Englishspeaking world, one should refer to the overviews made by W.C. Summers (1910: civ–cxiv), J. Espiner-Scott (1960), and G. Monsarrat (1974). Regarding the reception of Seneca in France during the 16th and 17th centuries—a time when this reception had found an especially influential exponent in Michel Montaigne (1533–1592; see Friedrich 1967: 62–68; Albrecht 2004: 173–192; Kraye 2007: 836)—the groundbreaking studies written by Julien Eymard d’Angers (ed. Antoine 1976) provide an excellent insight (see also Tobin 1971; Roche 1974; Spanneut 1990: 591–596). Spain went through a heyday of “Senequismo” (Blüher 1969) represented above all by Francisco Quevedo (1580–1645) and Baltasar Grácian (1601–1658; see Blüher 1969: 326–370, 371– 447). Northern Europe did not lag far behind: even a number of misogynic statements to be found in Seneca’s works (Manning 1973) did not prevent a woman, Birgitte Thott (1610–1662), from breaking into the phalanx of vernacular translators and, in 1658, becoming the first Danish translator of the moral writings of Seneca (Jensen 1995: 44). The ways in which Seneca was received within the rationalist and empiricist philosophical systems of the modern age are seldom clearly recognizable, and a complete analysis will yet have to be undertaken by scholars. In his letters to Princess Elisabeth of Sweden, René Descartes (1578–1650) thoroughly discussed Seneca’s De vita beata (epist. 397 ff. Oeuvres, ed. Adam and

matthias laarmann

Thannery, IV 263ff.). The Stoicism of Seneca and Epictetus inspired him during his conception of a “morale provisoire” (Julien-Eymard d’Angers 1976: 453–480), although he himself and to an even larger degree his pupil Nicolas Malebranche (1638–1715; see Kraye 2007: 837) rejected Seneca’s rhetorically charged style. Among the representatives of German school philosophy, Christian Thomasius (1655–1728) was most prominent because of the intensive academic treatment of Seneca’s De ira and his doctrine of the affections he provided in the lectures he held (Wundt 1945: 37, 43, 51). The degree of influence ancient-stoic philosophemes had on Baruch Spinoza (1632–1677) is currently a topic of controversy (Klessinger 2008). It is known, however, that Seneca’s Epistulae formed a part of his private library (Klessinger 2008: 998f.). Furthermore, Seneca is quoted by name in an important passage of Spinoza’s Ethica (5.42s). David Hume (1711–1776) presented Seneca in his Dialogues concerning natural religion—written in 1751, but published not before 1779—as an icon of rational religion, with Hume summing up his concept of a natural religion with a quotation from Seneca (epist. 95.47): deum colit, qui novit; everything else he considered “absurd, superstitious, or even impious” (ed. Gaskin 1993: no. 140). At the beginning of the Baroque era in Germany, the Silesian Martin Opitz (1597–1639) picked up the impulses coming from his teacher in Leiden, Daniel Heinsius (1580–1655), and became a great inspirer of Seneca reception (Stemplinger 1903; Stalder 1976; Wollgast 1988: 806–826; Riedel 2000: 85– 91), a reception that has been investigated in the context of—among other things—the martyr plays (Grätz 2008) and the didactic drama (Riedel 2000: 72f.) but also the bucolic and laus ruris poetry of the time (Lohmeier 1981). In addition to Opitz, Andreas Gryphius (1616–1664; see Riedel 2000: 93–96; Grätz 2008), Daniel Casper von Lohenstein (1635–1683; see Riedel 2000: 97– 100), and Johann Christian Hallmann (ca. 1640–1704; see Riede 2000: 100f.) deserve to be mentioned. The German literature of the 18th century, however, saw the continuous growth of a fundamental criticism of Seneca (Merrifield 1967). At first, there were still eminent supporters of Seneca like the tireless Wolffian Johann Gottfried Gottsched (1700–1766), who saw himself fully obliged to adhere to Seneca’s stylistic ideas (Ausführliche Redekunst, XVI. Hauptstück. 1736; Merrifield 1967: 531, 533). During the heyday of the Enlightenment, Gottfried Ephraim Lessing’s essay “Von den lateinischen Trauerspielen, welche unter dem Namen des Seneca bekannt sind” had a rather explosive effect insofar as Lessing was proclaiming the rejection of the heroic depiction of virtue in favor of the bourgeois-sentimental tragedy (Merrifield 1967: 537f.; Barner 1973; Riedel 2000: 135–140).

seneca the philosopher

The turning point toward a downright hostile rejection of Seneca in German literature is associated with Christoph Martin Wieland (1733– 1813), who was highly influential because of his translations of Horace and Lucian (Merrifield 1967: 529, 535, 537) and who, in the tradition of Horace’s criticism of the Stoa (sat. 1.3; 2.3), found fault with Seneca’s doctrine of the affections, which he considered bloodless and emotionally cold. In addition, the Enlightenment philosopher Christian Fürchtegott (1715–1769) charged Seneca with self-salvation and self-apotheosis (Späth 1992: VI 37f.). A critical saturation bombing against Seneca as a philosopher, writer, politician, and private person was carried out by Johann Gottfried Herder (1744–1803) in his treatise “Vom Einfluss der Regierung auf die Wissenschaften, und der Wissenschaften auf die Regierung” (1779): “If the Stoic philosophy of pretty words, lofty sayings, and unworthy living, if the philosophical education of a regent and the selfsame regent’s government under the eyes of his very well-paid teacher can ever have a stigma, they have it here” (trans. T. Budke; Werke, ed. Suphan 1877–1887: IX 389; Merrifield 1967: 535, 538). The fact that Herder later rescinded this verdict under the influence of Diderot (Werke XVIII/2, 391–401) did not receive any attention. Neither the continuing Seneca reception by Jean-Jacques Rousseau (1721–1778; see Bosshard 1967) nor the passionate attempts to rehabilitate Seneca to be found in Denis Diderot’s (1713–1784) two-part “Essai sur la vie de Sénèque, sur ses écrits et sur les règnes de Claude et Néron” (1778/82; German edition 1782, trans. F.L. Epheu; Conroy 1975) exerted any measurable influence. 9. The Opaque Continuum; or, Seneca’s Dormant Presence in the Philosophical Movements of the 19th and 20th Centuries Exit Seneca—literally and metaphorically. First, his plays, which had fascinated Europe for centuries, were culled from theater repertoires. Second, the prevailing preference for Greek culture and the original pathos of humanism was eroding the reputation of our Roman author as a writer and thinker. The persistent ancient tradition of passing harsh judgments on his personal conduct accelerated this process. Thus, the enthusiasm for Seneca displayed by the two main exponents of German Classicism, Johann Wolfgang von Goethe (1749–1832) and Friedrich Schiller (1759–1805), did not survive once they had reached maturity (Merrifield 1967: 531, 539–547; Albrecht 1999: 278–285). For G.W.F. Hegel (1770–1831), who discusses the Stoa in his “Lectures on the History of Philosophy” (ed. R.F. Brown; trans. R.F. Brown and J.M. Stewart

matthias laarmann

2006: II 265–279) the basic flaw of Roman philosophy, namely the complete loss of speculative interest (ibid. 260) becomes obvious in the works of Seneca, who “is known as a certain, limited Stoic” (ibid. 325, trans. T. Budke; cf. Hadot, I. 1969: 2f.). “In Seneca’s works themselves there is more pomp and grandiloquence about moral reflection than true solidity” (ibid. 292, trans. T. Budke). “In Seneca we find much that edifies, stimulates, and strengthens the mind—clever antitheses, rhetoric, and dialectic; but with these moral discourses we at the same time experience a certain feeling of coldness, a certain tedium” (Brown 2006: 279). Theodor Mommsen’s (1817– 1903) statement, “Even in our age, the charming style of his writings still causes delight even though they are devoid of content” (2005: 191, trans. T. Budke), seems to be a distant echo of Hegel. The well-known discussion about Seneca’s relationship to Christianity was continued on all sides around the middle of the 19th century. Against the attempt to hijack Seneca for Christianity, Ludwig Feuerbach (1804– 1872), in the appendix [no. XVII] of his magnum opus Das Wesen des Christentums (1841, 18483), went into battle with a powerful antithesis of Christian and pagan ideas about the end of the world (ed. 1956: II 482–485). In contrast, the Protestant theologian Christian Ferdinand Baur (1792–1860), the archeget of the historical-critical history of dogma, explicated at length and hermeneutically a clear difference between Seneca’s thought and the Christian theology of St. Paul (Baur 1858; Fürst 2006: 103f.). The left Hegelian and critic of religion, Bruno Bauer (1809–1882), lapsed into the other extreme: in his work Christus und die Cäsaren: Der Ursprung des Christentums aus dem griechischen Römertum (1877), he proclaimed the figure of Jesus an invention by Mark the Evangelist and called Seneca the true founder of the original Christian religion. During the transition to the positions held by the Philosophy of life (“Lebensphilosophie”) Stoic rigor, of which Seneca was a representative, became controversial. While Arthur Schopenhauer (1788–1866), whose writings show his wide reading of Seneca’s works, in 1848 reached a partly positive, ambivalent judgment of the entire Stoic philosophy (Neymeyr 2008a) in his main work Die Welt als Wille und Vorstellung (1988: II 99–109; cf. III 163–175), one of his main recipients, Friedrich Nietzsche (1844–1900), dubbed the Stoics “complete morons” (“vollendete Hornochsen”, trans. T. Budke; 1984: Nachlaß W II 3, Nr. 11. KSA 13: 125) and Seneca in 1887 a “toreador of virtue” (1911: The Twilight of the Idols: 60). Nietzsche’s general distrust of Stoic ethics, which he considered hostile to life (Neymeyr 2008b), was explicitly directed against Seneca, leaving out neither the Roman’s rhetoric style nor his philosophical content (Fröhliche Wissenschaft. 1984: KSA 3: 360f.).

seneca the philosopher

Even though Karl Marx (1818–1883) quoted Seneca in his doctoral thesis “Über Differenzen der demokritischen und epikureischen Naturphilosophie” (1841: I 266 f.) only en passant, some later Marxist-Leninist thinkers worked hard to force Seneca into the period framework of historical materialism (Schmidt 1960; Ley 1966: 450–457; Schmidt 1973; Seidel 1984: 159–162; Wollgast 1988: 760f.). In the meantime, historians of philosophy had begun contributing to a rehabilitation of Seneca. To Wilhelm Dilthey (1833–1911) belongs the honor of having rediscovered neo-Stoicism and the important role Seneca played in this movement, as presented in his study Weltanschauung und Analyse des Menschen seit Renaissance und Reformation (Dilthey 1914: 274–276, 443–452). In 20th-century German philosophy, notable references to Seneca can be found in existential ontology and in positions critically reflecting it. Martin Heidegger (1889–1973) illustrated the existential-ontological analysis of care (cura) with a long quotation from Seneca (epist. 124.14), found in §42 of his magnum opus Sein und Zeit (1927: 199). Hans Blumenberg (1920–1996) saw in the criticism of theoretical curiosity (curiositas) the aporia of Stoic philosophy “possibly most clearly”: to communicate the quest for knowledge as a teleologically inherent disposition of man together with a skeptical resignation necessary to prevent one from being drawn into the virtuethreatening infinity of curiosity (Blumenberg 1966: 299–301). Seneca and aporia? Where to go from there? 10. Seneca in the Present—a New Icon of Ars Vivendi? The Seneca renaissance we are experiencing today had to take the long road via painstaking and detailed analyses of classical philology and the history of ancient religions at the end of the 19th century. Paul Rabbow (1867– 1956), a pupil of the classical philologist Franz Buecheler (1837–1908), and the historian of religion Hermann Usener (1834–1905), presented groundbreaking studies on the ancient practices of self-shaping, in which the “care for oneself” (epiméleia heauto u¯ ) performs methodically controlled activities in order to treat affects and disarm misfortunes such as death through anticipation and preparation (Rabbow 1914). In Rabbow’s work Seelenführung: Methodik der Exerzitien in der Antike (1954), Seneca is the chief witness for the existence of a sophisticated repertoire of methods and measures, provided by pagan antiquity within the framework of autosoteric teachings regarding salvation, which, after hetero- or theosoteric reshaping (Seckler 2000) have lived on to the present day in the spiritual exercises of Christianity (cf. esp. the Exercitia

matthias laarmann

spiritualia [1548] by Ignatius of Loyola). Rabbow’s analyses were later and with much vigor taken up by Ilsetraut Hadot (*1930) and Pierre Hadot (1922– 2010). Michel Foucault (1926–1984), who incorporated results gained by Hadot and Hadot in his main work Le souci de soi (1984; English trans. 1986: The Care of the Self ), found widespread resonance in the intensive current discussions about the applications and meaning the ancient art of living can have for the present (Veyne 1993b; Schmidt 1996; Horn 1998, 2000, 2007). As a real transformation of Seneca’s philosophizing in the sense of providing concrete help in everyday life not by recurring to divine revelation and grace but by making use of the insight and the freedom of the client (not: patient!) one must consider Philosophical Coaching (Achenbach 1984, 2010; Marquard 1989; Achenbach). Clinical psychology, especially RationalEmotive Behavior Therapy (REBT), is directly connected to the ancient psychagogic knowledge contained in Seneca’s works (Hoellen 1986, 1987; Wiener 2008, 2009). Seneca is back in (the) practice! And in the same way he was mined for mirrors of princes in days past, he is now undergoing the same treatment for the ethics of the elite leaders within a modern mercantile society, demonstrated by, for example, Georg Schoeck’s (*1924) bilingual florilegium Seneca für Manager, with 40,000 printed copies a book that has attained wide circulation (Schoeck 1970). Seneca’s intellectual power is currently radiating far beyond the realms of specialized philosophy. Several important national literatures of the 20th and the beginning of the 21st century are offering clear and intensive direct references to and transformations of Senecan Stoicism (see Pauly 2008a). Fernando Pessoa (1888–1935) invented the stoicising authors “Ricardo Reis” and the “Baron of Teive” as proclaimers of their worldviews (see Pauly 2008a: 1214–1222). Regarding German national literature, Th. Ziolkowski used the title Seneca—A New German Icon?, referring to the immense growth and spread of an interest in Seneca prevalent in German culture since 1965, the 2,000th anniversary of his death. Günter Grass (*1927) portrayed—in his novel “örtlich betäubt,” published in 1969—the Studienrat Eberhard Starusch as someone who is fed up with politics and deeply influenced by Stoicism (see Pauly 2008a: 1243–1250). Those authors who experienced the repressive political and cultural conditions of the former GDR were highly interested in Seneca. The year 1977 saw the premiere of Peter Hacks’s (1928–2003) play Senecas Tod (Riedel 2000: 346–352). Seneca’s dignity during a time of political disaster was also the subject of Heiner Müller’s (1927–1995) poem “Senecas Tod,” composed in 1993 (ed. Hörningh 1998: I 250 f.). For the Büchner award winner Durs Grünbein (*1962), Seneca—notwithstanding all qualifying skepticism—might be called the leading intellectual reference point; he

seneca the philosopher

dedicated his long poem “An Seneca: Postskriptum” (Grünbein 2004: 9– 15) to the man who is unquestionably his favorite writer, and translated Seneca’s tragedy Thyestes (Grünbein 2006). “The thoughts of Seneca can be encountered virtually everywhere in Grünbein’s poetry and prose” (Pauly 2008a: 1252). With his poem “Sand oder Kalk” (Seidensticker and Grünbein 2002: 170f.), Grünbein took part in the debate about Seneca’s style and credibility. The title of a collection of essays written by Grünbein marks at the same time a substantial feature of some new literary horizons of the modern age: Antike Dispositionen (Grünbein 2005). Looking back at the reception of Seneca’s oeuvre—a reception full of twists and turns—one can say with certainty that every age to come will create its own personal Seneca. If we can trust in Seneca’s words, he would surely have been happy about successful transformations and transfigurations (cf. epist. 6.1) that have the power to advance the development of one’s own personality: Hoc ipse quoque facio: ex pluribus, quae legi, aliquid adprehendo (epist. 2.5).

SENECA THE DRAMATIST*

Werner Schubert Although there is no evidence that Seneca’s tragedies were performed in antiquity, there are many traces that indicate that the texts as such were generally known, read, and cited from Seneca’s lifetime onwards. The reception of his work, however, has varied. The remarkable revival of Seneca between the fifteenth and the seventeenth century was followed by a decline in the eighteenth century; in the nineteenth and twentieth centuries the interest in Seneca’s tragedies began to increase again, at least among scholars (Seidensticker and Armstrong 1985). Only in the past few decades have Seneca’s tragedies made an appearance on the stage,1—not in Latin, except for performances at schools and universities—but translated into modern languages and adapted to modern theatrical traditions.2 Antiquity As far as we can judge, Lucan, Seneca’s nephew, had certain passages of his uncle’s tragedies in mind when he wrote his (uncompleted) epic Pharsalia. Valerius Flaccus’s characterization of tyrants as well as the shadowy portrait of Medea in his epic Argonautica is indebted to Seneca, too, while Statius’s Thebaid is influenced both by the epic of his friend Lucan and Seneca’s tragedies, especially by Phoenissae and Oedipus. Concerning the crucial Troiae Halosis and the Bellum civile, which are both poetic fragments “composed” by the poetaster Eumolpus in Petronius’s Satyrica, at least the former poem may be a parody of Senecan monologues.

* Special thanks to Bettina Furley and Benedict Beckeld who read and emended my paper very carefully. This chapter owes a lot to the comprehensive studies concerning Seneca’s influence on the European drama by P.L. Schmidt, H.J. Tschiedel, K.A. Blüher, Chr. Wanke, B. Asmuth, R. Borgmeier, G. Dahlberg, W.-L. Liebermann, and W. Busch in Lefèvre 1978b, as well as to Trillitzsch 1978, and to the contributions in Billerbeck and Schmidt 2004. Cf. also the particular chapters on Seneca’s tragedies in this book. 1 See the list of performances of Senecan tragedies from 1993 to 2007 by K. Kagerer and W. Stroh (http://www.lrz.de/~stroh/schriften/seneca_scaenicus.html). 2 Cf. also the section “The Present” in this article, infra, pp. 92f.

werner schubert

The most considerable signs of a more or less contemporary influence are to be seen in the Corpus Senecanum itself. As far as we know today, the tragedies Octavia and Hercules Oetaeus do not originate from Seneca himself, but are strictly shaped, by unknown poets, according to the structure of the genuine Senecan dramas. The language and style of these two tragedies, however, are identical to the language and style of the others; that is why the discussion of whether these two tragedies are genuine has not yet ceased. Hercules Oetaeus is by far the longest tragedy in the Corpus Senecanum and thus in ancient Latin literature. It contains many resemblances to Hercules furens; the unknown author of Hercules Oetaeus entered an intertextual dialogue between himself and Seneca concerning the identical subject of the two works. In the Octavia we note the same dramatic technique as in the other tragedies. Moreover, the obvious “prince’s mirror” parts are doubtlessly influenced by the earlier Senecan dramas. Seneca’s tragedies were, then, starting from Lucan and Petronius, both imitated and criticized, but nonetheless read and transmitted throughout the Roman world. Seneca’s choruses, being very innovative, influenced ancient authors from Caesius Bassus (†79) to Boethius († 524). Quintilian recalls in his Institutio oratoria that when still a student he heard a Senecan praefatio to one of his tragedies. In 9.2.9 he quotes Sen. Med. 453; in 10.1.125–131, however, he gives his judgment on Seneca’s prose writings rather than on his poems. Pushed back because of classicistic and archaic tendencies, Seneca was to be rediscovered only some centuries later, starting about ad 370. One of the strangest adaptations of a Senecan drama in late antiquity is Hosidius Geta’s Medea, an odd cento of Vergilian fragments. The play is no boring l’ art pour l’art exercise of style, but a very demanding discussion of different traditions through its deconstruction and reconstruction of both Vergil and Seneca (cf. Schmidt 1978: 37f.). At this time “Seneca tragicus” and “Seneca philosophus” were wrongly regarded as two different persons. The following authors often refer to Seneca’s tragedies: Claudianus (Zwierlein 1984: 7–12; 46–57), Prudentius, Orientius (and the anonymous comedy Querolus), Sidonius Apollinaris, Avitus, Dracontius, Ennodius, and Boethius (Trillitzsch 1978: 121f.). Whether Augustine ever read Senecan tragedies we do not know; his few citations may just as well have been taken from anthologies, grammars, or somewhere else. As to the poets among the listed authors, we can assume that they knew the one or the other Senecan tragedy very well. Hieronymus evidently made use of some verses taken from Seneca’s Troades when he wrote his Vita Malchi. The Senecan prologues or choral parts with their more general concern with philosophy or mythology inspired a special interest among the authors of

seneca the dramatist

the early Christian period. In his Cathemerinon 10, Prudentius seems to have had in mind Sen. Tro. 371 ff. (cf. Schmidt 1978: 50). Prudentius not only made use of isolated Senecan expressions, but integrated entire scenes into his poems. The same can be said of Claudianus or Ausonius, whose knowledge of Senecan tragedies reflects the literary education especially in Gaul, both of the poets and the reading public. At the end of antiquity, the metrical patterns as well as the philosophical views in the inserted poems in Boethius’s Consolatio philosophiae resemble Senecan (or Pseudo-Senecan) choral parts. Schmidt (1978: 54) has pointed out that the Senecan tragedies at this time and in the Middle Ages were regarded as mere literary dramas. They formed part of the rhetorical education from the fourth century onward. During the last period of ancient poetry, at least for Claudianus, Prudentius, and Boethius, the gap between Vergilian classicism and Senecan modernism no longer existed. Middle Ages Interest in Seneca’s tragedies seems to have decreased during late antiquity and the beginning of the Middle Ages, triggered perhaps by Boethius, who in his Consolatio substituted the Senecan choruses with his carmina as patterns to be imitated by later poets (cf. Schmidt 1978: 62). As to the textual tradition at this time, we can discern only one palimpsest (5th–7th century). To Isidore of Sevilla (560–636) Seneca seems to have been unknown; only Aldhelm of Malmesbury (640–709) quotes two verses from Seneca’s Agamemnon, naming the author explicitly. Perhaps Aldhelm had access to a complete codex of Seneca’s tragedies. During the Carolingian Renaissance there are reminiscences of Herc. O. in Theodulf of Orléans (750–821). At the beginning of the tenth century there is evidence of Eugenius Vulgaris’s knowledge of Seneca in southern Italy. Subsequently also in northern Italy some literary documents show traces of knowledge of Seneca; Liutprant of Cremona (ca. 920–972) in his Antapodosis had Sen. Phaedr. 749–752 in mind. In the eleventh century, in which the highly important Codex Etruscus was written, knowledge of Seneca’s tragedies seems to have increased steadily. In the Lexicon of Papias (cf. Trillitzsch 1978: 126), Hercules furens is called the “prima tragoedia”; cf. also Petrus Damiani (1007– 1072) in De sancta simplicitate. In the twelfth century, Seneca’s tragedies were clearly read in schools. The most important documents stem from France (Zwierlein 1987). In the middle of the thirteenth century, Richardus de Fournival (see Trillitzsch 1978: 127) lists the ten tragedies of Seneca, including

werner schubert

Octavia, in the same order as in the MSS of group A. His contemporary, Vincent of Beauvais (†1264), frequently cites Seneca in hisSpeculum maius. We do not know if these authors really knew Seneca’s tragedies, or if they used anthologies and other secondary sources. Trillitzsch (1978: 128f.) points out that “insgesamt die wirkliche Lektüre der Senecatragödien im Mittelalter selten [ist], woran auch die spätere gelegentliche Aufnahme unter die Schulund Sentenzautoren nicht viel ändert.” 3 It is important to note that there is no evidence that Seneca’s tragedies ever inspired the medieval drama. It is not until the Renaissance that a productive adaptation of Seneca’s tragedies is perceptible. As to the question of what Seneca’s tragedies in the Middle Ages were used for, Schmidt (1978: 72) suggests: “In den Helden der Handlung sieht das Hohe Mittelalter Vorbilder und abschreckende Beispiele, und noch direkter nehmen sich in der mittelalterlichen Rezeption […] etwa der ‘Thyest’ oder die ‘Octavia’ als politisch-philosophische Lehrstücke, als Fürstenspiegel aus.”4 Renaissance (13th/14th–16th Century)5 It is during the Renaissance that Seneca’s tragedies were discovered as poetry (cf. Pastore-Stocchi 1964). This is attested by the great number of manuscripts that were written in the thirteenth and fourteenth centuries. The commentaries of Nicolas Trevet(h) played a most important role in this process.6 And here for the first time Seneca’s tragedies served as patterns for new tragedies, first in Latin, later in other languages. This was the birth of the European tragedy, long before the Greek tragedy was rediscovered and reevaluated. The process was initiated in Italy by the so-called Paduan pre-humanists. Their compendia show that they disposed of the whole corpus of Seneca’s tragedies, which, in the beginning, were treated as the

3 “[…] all in all, actual reading of Seneca’s tragedies [is] rare in the Middle Ages, which is not changed much by the occasional inclusion in textbooks and collections of aphorisms in later times.” 4 “The High Middle Ages sees both role models and cautionary examples in the heroes of the drama, and even more directly in the medieval reception […] are plays like the ‘Thyestes’ or the ‘Octavia’ exemplified as politico-philosophical pieces, as mirrors of princes.” 5 For Seneca in the Renaissance and Baroque, see especially Boyle 1997: 141–207. 6 While Trevet’s (or Mussato’s) works have met with much scholarly interest in recent years (for Trevet cf., e.g., Junge 1999; Marchitelli 1999; Marchitelli 2000; for Mussato cf. MacGregor 1980), many other 14th/15th-century commentaries on Seneca are still widely unnoticed as K. Hafemann observes in her edition of the commentary on Seneca’s Hercules furens by Iohannes de Segarellis (before 1400) (Hafemann 2003).

seneca the dramatist

philosophical texts of a pagan poet who had supposedly corresponded with Paulus the Apostle and whose thinking was close to Christianity. Perhaps Dante Alighieri (1265–1321) knew Seneca’s tragedies only from anthologies, but Boccaccio (1313–1375) obviously did know the texts; some chapters from the first book of De casibus virorum illustrium render tragic plots based on or influenced by Seneca. Lovato de’ Lovati (1241–1309) discovered the hitherto unknown Codex Etruscus, which differs in many ways from the manuscripts of group A. On the one hand, it does not include Octavia; on the other hand, there is a kind of introduction, a “notamentum” containing a definition of what a tragic poet should be; the verses and meters of the tragedies are explained, too. Hence the texts were no longer regarded as philosophical treatises, but as tragic poems. And it was perhaps Lovato de’ Lovati’s influence that inspired his friend Albertino Mussato (1261–1329) to write the first (known) tragedy since antiquity: Ecerinis (Braden 1985: 99–114; Boyle 1997: 141). The title shows an affinity to those of ancient epics (like Vergil’s Aeneis or Statius’s Thebais) and to some of Seneca’s tragedies in the Atradition (Thebais or Troas).7 As a matter of fact, the plot, which covers many years, is more epic than dramatic; it is based on a rather cruel story which had taken place not in mythic times, but only a century before the Ecerinis was written; the influence of Seneca is to be seen mainly in the structure (five acts, separated by choruses), the wide range of different meters, and the stock of typical scenes we know from Seneca. The language and style are, of course, the same as Seneca’s; as to the scope, it is Senecan as well: it is, as Tschiedel (1978: 83) has pointed out, “die ständig wiederkehrende Vorstellung eines Tyrannen von exzessiver Willkür und Grausamkeit, dessen Untaten keinem anderen Zweck als der Befriedigung der eigenen Natur dienen.”8 According to medieval practice this tragedy was never performed, but only recited. Mussato’s opus remained unrivalled for a long time. Among his successors, only Gregorio Corraro (1411–1464) chose a plot that is very close to Seneca’s for his Progne, which allowed him to imitate this prototype in many ways, as he himself explains: “Imitatur in hac tragoedia Senecam in Thyeste; ut ibi Tantalus ab inferis veniens introducitur, ita hic Diomedes Thrax Tyrannus.”9 Corraro also adapted the typically Senecan domina-nutrix scenes, which thenceforth belonged to the most important elements of

I am grateful to Andreas Heil for pointing this out. “[…] the ever recurring idea of a tyrant of excessive despotism and cruelty whose misdeeds serve no other purpose but the satisfaction of his own nature.” 9 “In this tragedy, I imitate Seneca’s Thyestes; as Tantalus is introduced coming from the underworld there, so the tyrant Diomedes Thrax here” (Latin quoted in Tschiedel 1978: 85). 8

werner schubert

Italian tragedy. The cruel details of the plot were willingly imitated—and exaggerated—by other tragedians. Even these plays were only read or recited. It was not until the end of the fifteenth century that the so-called Roman Academy, under its leader Pomponius Laetus (1428–1497), endeavored to bring to the stage the comedies of Plautus and Terence—and the tragedies of Seneca. The time for the Greek tragedy had not yet come. It was the kinship of Latin and Italian that fostered the prevalence of the ancient Latin dramas and perhaps also the bloodthirstiness and shocking effects in the action, the lapidary shortness of sentences, and the diction. Hence the term “Senechismo”! At the beginning of the sixteenth century, the first attempts were made to write dramas in the vernacular, by which the Aristotelian aims of eleos and phobos could be better attained than by retaining the use of Latin. This change required the use of new meters and verses. The poet Giangiorgio Trissino (1478–1550) successfully introduced the “endecasillabi sciolti” for the dialogue scenes and the “canzoni” for the choruses. His Sofonisba is based on Senecan models; influences from Euripides and Sophocles are already perceptible, but they are nothing more than ornaments. The drama itself is “Roman.” And it is this Romanness that fits into Renaissance thinking with its accentuation of man’s free will and responsibility. The Sofonisba was never performed; it was Giovambattista Giraldi’s (1504–1573) tragedy Orbecche (1541) whose fame spread throughout Italy and Europe and thus proved to be the beginning of modern tragedy (Braden 1985: 115–124). 10 The play is based on one of Giraldi’s own novels and on Seneca’sThyestes (Boyle 1997: 150 f.). Tschiedel (1978: 104) has pointed out that the obvious influence of the Italian novel was caused by the close relationship between the novel and Senecan tragedy itself: compared with Greek tragedy, Seneca had no longer treated the myths as stories telling of gods and men, but as pointed actions full of unexpected turns caused by men’s passionate reactions. The special predilection for exotic places is also to be seen in Seneca’s tragedies, especially in the enumeration of unknown, sonorous names and places. In the same way as Seneca used myths with their distant persons and places to mirror his contemporary world, the authors of Italian tragedy could hide their political messages, if there were any, in an “exotic” disguise. The Octavia, whether Senecan or not, with its unequal pair consisting of a tyrant and a wise counselor, precipitated a flow of Renaissance dramas. Seneca’s

On the general influence of Seneca on Giraldi, see Dondoni 1964.

seneca the dramatist

dramatic technique became omnipresent, although the plots and motifs were increasingly taken from elsewhere, as for example from Ovid in Sperone Speroni’s (1500–1588) Canace, from Livy in Pietro Aretino’s (1492–1556) Orazia, or from Vergil in Ludovico Dolce’s (1508–1568) Didone. Nevertheless, Torquato Tasso’s (1544–1595) not very successful tragedy Il re Torrismondo owes a lot to Seneca’s Oedipus. While there are no traces of knowledge of the Senecan oeuvre in Spain during the Middle Ages, the reception of Seneca’s tragedies in the Renaissance started with editions and translations (into Castilian as well as Catalanian) at the beginning of the fifteenth century. The school theater and the tradition of spiritual plays were dominated by the activities of the humanists. In 1543 Francisco Satorres published his tragedy Delphinus, which was influenced by Seneca. In the so-called Jesuit dramas, which were partly written in the vernacular, we observe the well-known Senecan disposition of a tragedy in five acts separated by choruses. In the dramas of the second half of the sixteenth century, which are most important concerning the development of the Spanish theater, Seneca’s influence, mediated by Italian tragedy, also is demonstrated by the predilection for mythology and history. Cruel tyrants (see MacCurdy 1964) are an important tragic theme in Spain as elsewhere in the sixteenth and seventeenth centuries. Positive characters are rare and mostly embodied by women; ghostly apparitions, foreshadowing dreams, and evil omens belong to the stock scenes; choruses are mainly to be found in plays with a strong classicist attitude.11 Similar things can be said of French tragedies. The beginnings are to be seen in the liturgical and sacred dramas of the Middle Ages, which were written very early in the vernacular. In the middle of the sixteenth century, the revival of ancient tragedy in France began. George Buchanan (1506–1582) wrote Latin plays, influenced by Euripides and Seneca, on biblical themes. Marc-Antoine Muret (1526–1585) was the first who chose a non-biblical subject when he wrote his Julius Caesar (Braden 1985: 124–129) in Latin. This tragedy is Senecan in structure, plot, and diction with its “adhortationes,” “altercationes,” and stichomyths. But there is at least one deviation from the Senecan tradition that was praised by his contemporaries: instead of a “nuntius” who reports the horrible slaughter of Caesar, it is Brutus himself, one of the main characters, who exults in having executed the murder (see Wanke 1978: 183).

For Seneca in Spain, see especially Blüher 1969 and Blüher 1978.

werner schubert

The first French tragedy was Etienne Jodelle’s (1532–1573) Cléopâtre Captive (1552), based on Plutarch but modeled on Seneca. Among the first French tragedies in which Senecan plots were adapted was Jean de la Péruse’s (1529– 1554) Médée (1553), which proved to be the starting point for an overwhelming and constant flow of Medea dramas. There are, though, important differences between Seneca’s Medea and de la Péruse’s Médée in the disposition, the diction, and the length of the respective dramas. Robert Garnier (1545– 1590) wrote eight tragedies between 1568 and 1583; three of them are based on Senecan tragedies: Hippolyte, La Troade, and Antigone. Garnier often mingles scenes from different dramas into one. Seneca’s short sentences are extended by learned comparisons. In creating word play he emulates or even surpasses Seneca. It was Garnier whom the French poets of the seventeenth century had in mind when they began to shape dramas in the Senecan mold. In England, the Netherlands, Germany, and Northern Europe, Seneca’s Nachleben was quite different from that in the Romanic tradition. To begin with Scandinavia, there are but few traces of Senecan knowledge before the middle of the seventeenth century (see Dahlberg 1978). During the Reformation, scholars put all their efforts into the demands of religious developments. Thus, Renaissance thinking lost its influence in Scandinavia. Seneca’s tragedies seem to have been neither read in schools nor performed on stage. Scant traces of Senecan reception in Denmark are to be seen in the Latin play Susanna by Sixt Birck (1537), which contains allusions to Seneca’s Phaedra. The knowledge of Seneca’s tragedies, however, seems to have been mediated only by anthologies; there is generally a special inclination to proverbs and sentences in sixteenth-century school drama in Denmark. In Sweden Seneca played a more important role than in the other Scandinavian countries. The first Swedish tragedy was Urban Hjärne’s adaptation of Jacob van Zevecote’s Latin drama Rosimunda, written in the 1660s, in which the Dutch humanist and poet made use of his knowledge of Seneca’s tragedies. Remarkable are the two redactions of Seneca’s Phaedra and Troades (Hippolitus and Troas), which were performed, and possibly written, by members of the first permanent Swedish theater company, Dän Swänska Theatren, at the end of the seventeenth century. In the Netherlands12 Seneca was already known in the Middle Ages; in the Renaissance his influence was augmented by Rudolf Agricola (1444– 1485) and, still more, by Erasmus of Rotterdam (1466 [1469?]–1536), who,

See Asmuth 1978.

seneca the dramatist

incidentally, still thought Seneca the philosopher to be the father or the brother of the poet. And it was Seneca the philosopher who exerted his influence on thinking and writing at the beginning of neo-Stoicism. But Seneca’s tragedies, though read and well known, did not serve as examples for school dramas, while Terence’s comedies did. It was not until the 1580s that Seneca the tragedian was revived in the Netherlands, according to the “europäische Großwetterlage des aufkommenden Manierismus” 13 (Asmuth 1978: 240). In 1600 Jacob Duym (ca. 1547–ca. 1606) for the first time translated a complete Senecan tragedy (Troades) into Dutch: Den Spieghel des Hoochmoets Wesende Troiades. The neo-Latin dramas of the neo-Stoics, not only in the Netherlands but also in Germany, were based on biblical plots, but shaped on Senecan models. “Mit der Figur des positiven, stoische Apathie verkörpernden Helden verwirklichen sie auf der Bühne, was Seneca als Lehre aus seinen grauenvollen Theaterstücken und als Gegensatz zu ihnen im Leben verwirklicht wissen möchte” (Asmuth 1978: 247).14 In Germany,15 too, the revival of antique drama started with comedy. In the time of humanism the reception of Senecan tragedies was fostered not only by the efforts of Erasmus, but also of Konrad Celtis (1459–1508), the first German “poeta laureatus.” He was a pupil of the above-mentioned Pomponius Laetus, and he was the first to stage antique comedies in Vienna. Some years before, he had given lectures on Seneca’s tragedies in Leipzig; he had also planned an edition that, however, was only partially finished (Hercules furens and Thyestes). In his preface Celtis compares the ten tragedies with the Ten Commandments. On German stages the spiritual plays and the “Fastnachtsspiele” were still popular, but gradually other influences became apparent. The dramatists of that time, however, who were certainly acquainted with pagan and Christian Latin tragedies, did not yet follow Seneca. As Liebermann (1978: 384) notes: “Senecas Figuren sind grundsätzlich autonom, und nur in der schuldhaften, selbst zu verantwortenden Preisgabe dieser Autonomie lassen sie sich von fremden Mächten beherrschen—wovon sie in Selbstanalyse und Selbstaussprache höchst bewußt Rechenschaft geben. Die Heteronomie der Menschen des 16. Jahrhunderts dagegen ist konstitutionell und in diesem Verstande wertfrei.”16 Seneca’s tragedies served mainly as a

“the general apparition of mannerism in Europe”. “Through the character of the positive and stoically disinterested hero, they realize on stage the teaching that Seneca sought to realize in life through the horror of his plays.” 15 See Liebermann 1978. 16 “Seneca’s characters are in general autonomous, and only in case of the loss of this 14

werner schubert

repertory of pointed sentences; if ancient plays had an influence on dramatic writing at all, it was through Terence’s comedies. As for England,17 from 1559 to 1581 Seneca’s tragedies appeared one by one in English translations by several authors. They were finally edited by Thomas Newton as The Tenne Tragedies, which remained the sole complete English translation up to the twentieth century. Seneca was regarded as the representative of ancient tragedy; Greek drama was hardly known to anybody at this time. But this high reputation was a theoretical one; there are no traces of Senecan influence based on the texts of the tragedies as such; perhaps most Senecan reminiscences were mediated by anthologies or excerpts. The Tenne Tragedies reveal different translation techniques. Dark allusions to mythology in the original plays are replaced by contemporary images; enigmatic antonomasies are substituted by their “solutions”; the iambic trimeter is dispensed with in favor of the iambic septenar. Alliterations are very common and even more frequently used than in the Latin texts. The first regular English tragedy, Gorboduc or Ferrex and Porrex (1561/62) by Thomas Sackville and Thomas Norton, was already influenced by Senecan elements in plot and structure (Bacquet 1964). Thomas Kyd’s The Spanish Tragedy (first performance between 1582 and 1592) was very important for Elizabethan drama, because here for the first time tragedy entered the domain of folk play (Habicht 1964). The plot is as alien from Seneca as from Elizabethan spectators. But the stock of characters and of typical scenes is strongly influenced by Seneca. As for Slavic literature,18 in Poland there was a remarkable interest in Seneca from the fifteenth century onward. As early as the beginning of the sixteenth century, there were editions. From 1589 onward there were also translations of Senecan tragedies. The Societas Jesu had an important influence on Polish intellectual life. In other east and southeast European countries, the performance of a Senecan play can be documented here and there. In 1573 the Thyestes was performed in Olmütz/Olomouc (Moravia). In Ragusa, now Dubrovnik, Giacomo Bona composed hisHerculis labores et gesta in Christi figuram (1513), which shows a Senecan influence.

autonomy, due to their own fault, do they let themselves be controlled by other forces— for which they consciously account through self-analysis and monologue. The heteronomy of sixteenth-century people, in contrast, is fundamentally and in the thought of this age non-judgemental.” 17 See especially Borgmeier 1978; Braden 1985: 153–223; and Boyle 1997. 18 See Busch 1978.

seneca the dramatist

Baroque Period (17th Century) The seventeenth century is perhaps the most “Senecan” period in the history of European theater. In Italy the development of stage plays had already climaxed in the sixteenth century. Lyric elements increasingly pervaded dramatic poetry. Thus, the interest in Seneca’s tragedies shifted from the dialogues and monologues, where iambic meters usually prevail, to the choruses, which were written in lyric meters. No wonder that music began to play an increasingly large role on stage. This development culminated in the birth of a new genre: the opera. The plots of the first operas stemmed from antiquity. It was, however, not the ancient drama itself that was responsible for this development, but the bucolic world with singing nymphs and shepherds and with mythological figures, like Daphne or Orpheus, depicted therein. It is remarkable that the texts of these musical dramas, later called “libretti,” had much in common with Senecan drama in their (sometimes schematic) structure, the importance of choral parts, and the predilection for “monologues,” or better, monodies (Schubert 2004: 378–387). Senecan influence is also perceptible in the growing importance of the theme “fickleness of luck.” Not only classical mythology, but also ancient history served as thematic and textual sources. One of Claudio Monteverdi’s (1567–1643) three great operas, which have survived, is based on ancient history: L’incoronazione di Poppea. In this opera Seneca himself appears on stage; he is the Stoic philosopher and teacher of Nero, who, enslaved to his mistress, compels his teacher to commit suicide (Manuwald 2008: 134–138). The libretto is based, as Manuwald (2005a) demonstrates, and as had been assumed by other scholars, on passages from Tacitus’s Annals and from Octavia, the only existing ancient fabula praetexta, which at this time was still believed to have been written by Seneca himself.19 In France the Senecan tradition, which started in the sixteenth century, remains today. In the early seventeenth century, both Senecan (or pseudoSenecan) Hercules dramas were adapted by Prévost (1614), de Mainfray (1616), and Rotrou (1634); La Pinelière followed them with his version of Phaedra (1635). In these adaptations, the Senecan choral parts were modified and partly transformed into dialogues. The main representatives of French tragedy in the seventeenth century were Pierre Corneille (1606–1684)20

19 20

For other operas with Nero and Octavia as protagonists, see Manuwald 2005b. For Seneca’s influence on Corneille, see Wanke 1964; Braden 1985: 134–152.

werner schubert

and Jean Racine (1639–1699).21 Some of their dramas are, at least in part, more Senecan than their authors were willing to concede, as far as we can conclude from their commentaries, introductions, and letters concerning, say, Corneille’s Médée (1635, cf. Stegmann 1964) or Racine’s Phèdre et Hippolyte (1677). In Médée there are long passages not only influenced by but also taken (and, of course, translated) from the Senecan tragedy. The main plot itself is enriched by an underplot, which, as in some operas of that time, contains elements of comedy. Thus, the action is far more complicated than in Seneca’s tragedies, but the logic of the development leading to the final catastrophe is sometimes more stringent. Not only Médée but also Corneille’s other dramas reflect Seneca’s style and his idea of man. Like Seneca, Corneille made use of two kinds of speech: passionate and ingenious. The structure of Corneille’s tragedies follows Senecan patterns, although his monologues are not as numerous as in the ancient models. Instead, Corneille makes use of scenes containing conversations with confidant(e)s, especially in the dramatic expositions. The influence of Senecan thinking is to be seen in the fearlessness before death and in the importance of freedom of will. But Corneille differs from Seneca in his judging of human passions. While Seneca rejects them all, Corneille retains and justifies the “great passions,” such as ambition and revenge. Like Corneille, Racine downplayed Seneca’s influence on his writing22 when he commented on his own dramas, although it is evident that Racine had in view Seneca’s Troades and Phaedra when he wrote his La Thébaïde and his Phèdre et Hippolyte.23 It was obsolete to refer to Seneca at this time, as Wanke (1978: 207) has pointed out. There are also reminiscences of Seneca’s Medea in the vision of hell in the jealousy scene in Phèdre et Hippolyte. As in Corneille’s tragedies, there are, however, some differences from the Latin author. In Phèdre et Hippolyte the plot is enriched by Hippolyte’s love for Aricie. The bloody details after the death of Hippolyte are softened in the French tragedy, which, as a whole, had an enormous effect on contemporary writing. One hundred and thirty years later, no less a poet than Schiller translated the tragedy into German.

For Seneca’s influence on Racine cf. Wanke 1978: 207–220; Zwierlein 2006: 29–53. Lapp’s study (1964) carefully analyzes the features that both poets have in common and those in which they differ. 23 According to Boyle, “even after Racine the reworking of Seneca is apparent” (Boyle 1997: 152 f.); cf. Prosper Jolyot de Crébillon’s Atrée et Thyeste. 22

seneca the dramatist

It was in the Netherlands 24 that the whole range of works by “Seneca the tragedian” and “Seneca the philosopher” was first regarded as a unity with the rise of neo-Stoicism in the mid-sixteenth century. In consequence of that development, the dramatists of “de gouden eeuw” (1580–1680) were well aware of the fact that Seneca was one of the most prominent mediators between ancient and modern stoicism. Justus Lipsius (1547–1606), who, with his De constantia (1584), had prepared the ground for the influence of neoStoicism, fostered the renewed interest in Seneca as well as in Tacitus; Daniel Heinsius (1580–1655) and Hugo Grotius (1583–1645) followed him by writing Latin tragedies modeled on Senecan examples. Pieter Corneliszoon Hooft (1581–1647) was the first to write Dutch dramas. He integrated passages taken from Seneca’s Troades and Agamemnon into his first tragedy Achilles en Polyxena. In his later tragedies, Hooft, like other poets, made use of Seneca’s typical scenes, and it was Hooft who revalorized the magic scenes and ghost apparitions, reflecting the superstition of his time, when the prosecution of witches had reached its climax. It is, however, interesting that Seneca’s influence is not so great in Hooft’s tragedies based on ancient plays and plots as in his historical tragedies; the recourse to ancient models generally served to justify modern dramatic production. Contrary to the Senecan models are Hooft’s apotheotic finali, in which the bellicose actions lead up to visions of calm and peace that can be realized hic et nunc. Asmuth (1978: 260) recapitulates the importance of Hooft’s adaptation of Seneca as follows: “Aus Senecas antiker, im MythischKultischen wurzelnder und von stoischen Maximen gerahmter Tragödie hat Hooft eine vaterländische Staatstragödie geformt.”25 Some authors in the Netherlands also catered to the spectators’ lust for cruel and bloody actions to which, above all, Seneca’s Thyestes served as a model, for example, Samuel Coster’s (1579–1665) Ithys. Jan Vos’s (ca. 1620– 1667) Aran en Titus, of Wraak en Weerwrak seems to be a combination of Shakespeare’s Titus Andronicus and Senecan tragedy. Guilliam van Nieuwelandt’s (1585–1635) Nero, in which the author portrayed Nero’s most gruesome crimes, is a still more atrocious example. One of the most famous Dutch poets is the prolific dramatist Joost van den Vondel (1587–1679). When Vondel began to write, perform, and publish his tragedies, the Senecan influence may have been mediated by others. After he

Cf. Stachel 1907: 137–179; Asmuth 1978. “Out of Seneca’s ancient tragedy, rooted in myth and ritual and framed by stoic maxims, Hooft created a patriotic national tragedy.” Cf. also Rombauts 1964 and Smit 1964. 25

werner schubert

had learned Latin, Vondel translated Seneca’s tragedies from 1625 onwards; henceforth his plays are full of Senecan echoes. In the 1640s we can see a “change of paradigm” in Vondel’s oeuvre after he had translated Sophocles’ Elektra (1639). This new acquaintance with Greek tragedy was the beginning of the end of Seneca’s influence on Dutch tragedy. In Germany26 the Senecan influence gradually grew toward the end of the sixteenth century, by way of contemporary French, Dutch, and English tragedy. During the Counter-Reformation, Jesuit drama, which aimed to appeal to its audience by every means of rhetorical persuasion and stagecraft, became very popular. The Protestant school drama also flourished in these times and nourished the beginnings of baroque drama. Seneca’s revival in Germany took place at this time;27 his tragedies were regarded as representative paradigms of human existence with a predilection for the tension between Stoic patience and imperturbability on the one hand, and tyrannical incalculability and cruelty on the other. In terms of the contemporary Weltanschauung, Seneca served as a Stoic paradigm in his way of thinking, living, and dying. The poet Martin Opitz (1597–1639), 28 after he had finished his “Buch von der deutschen Poeterey” (1624), translated Seneca’s Troades (Plard 1964a). This translation, however, is more extensive than the original text with meticulous periphrases and explanations for the readers and spectators who were not acquainted with ancient mythology, geography, or history. Opitz, incidentally, also translated Sophocles’s Antigone (1636). There are many difficulties in determining the boundaries between the immediate reception of Seneca’s tragedies in Germany and those that were mediated through secondary sources. The outstanding dramatists of the seventeenth century were Andreas Gryphius (1616–1664; see Stachel 1907: 204–274; Plard 1964b) and Daniel Casper von Lohenstein (1635–1683; see Stachel 1907: 274–324; Lefebvre 1964). It was Schings who first demonstrated that Gryphius was influenced by patristic and Stoic traditions not only in a vague sense, but also in his intimate familiarity with the essential texts, especially with Seneca’s writings (Schings 1966). It is evident from this time onward that the influence of Seneca the philosopher is inseparable from that of Seneca the tragedian. Incidentally, “Seneca” was used as a common

See Stachel 1907: 30–136 and 180–350. “In den strukturbestimmenden Erscheinungen ist Seneca für die Barockdichter der kongeniale Partner” (Liebermann 1978: 391). 28 See von Albrecht 2004: 197–202. 27

seneca the dramatist

epitheton for both Gryphius and Lohenstein by their contemporaries. As in Seneca’s tragedies, Gryphius’ dramas represent the condition of the world we live in, and they try to offer an interpretation of it. In Seneca’s tragedies, the cosmos is disturbed and distorted by man; the divine principles are thus invalidated. In Gryphius’s plays, on the other hand, the divine order reigns above all; man has to carry his earthly burden, such as social inferiority or prosecution by enemies, with stoic patience. Gryphius mastered the wide range of Senecan stylistic elements, which was, after all, characteristic of his contemporary dramatists. With Lohenstein the history of the political tragedy in Germany began. The ancient historian Tacitus became especially important. Lohenstein’s rich and pointed style is influenced by Seneca as well as by Tacitus. The Age of Enlightenment had already begun to dawn, for in Lohenstein’s tragedies reason gradually becomes the guiding principle, and this implies a new function of the learned allusions and metaphors taken from ancient models: “Diese ‘Frostigkeit’ diente ihm nicht anders als dem Archegeten des barocken Dramas Seneca dazu, das Überbordende und Maßlose reflektierend zu bändigen, in den Rang des vorab bereits gedeuteten Exemplarischen zu erheben und damit dem Leser und Zuschauer die weltbewältigende Distanz zu ermöglichen”29 (Liebermann 1978: 424). In Spain30 Seneca’s influence ceased in the seventeenth century when the new concept of the “comedia” spread throughout the country, fostered by the success of Lope de Vega’s (1562–1635) plays, which followed neither Aristotelian rules nor Seneca’s dramatic technique. But this is also the time when a Senecan tragedy was first translated into Spanish (Troades, by Gonzáles de Salas 1633). Some authors tried to adapt Seneca’s dramas to the “comedia,” like Francisco López de Zárate in his Tragedia de Hércules Furente y Oeta 1651. While Seneca the tragedian disappeared, Seneca the philosopher was rediscovered by his fellow “countrymen”. In England at the turn of the sixteenth and seventeenth century, Seneca’s influence continued to be felt among dramatists who adopted certain elements of his structure and style while taking their plots from elsewhere.

29 “This ‘frostiness’ served him in the same way it had served the founding father of baroque drama, Seneca, to control the excessiveness and extremeness through reflection, to raise these to the status of a previously interpreted example, and in doing so to provide the reader and the audience with the detachment necessary to come to terms with the world.” 30 See Blüher 1969 and Blüher 1978.

werner schubert

William Shakespeare’s31 early plays show an affinity for antique plots, mainly taken from history; these and other historical tragedies are often focused on tyrants and villains; Seneca’s influence is to be seen or heard mainly in the diction and in the use of typical scenes, whereas an emancipation from Seneca can be seen in the lack of mythological subjects and the new functions of old dramatic elements. So the—at first sight typically Senecan— ghost apparitions are not put at the beginning of a tragedy as a kind of hors d’oeuvre, but are integrated into the dramatic action. Seneca’s influence is more important in other regards: Shakespeare’s later tragedies as well as those of his contemporaries reflect Seneca’s stoicism. George Chapman (1559[?]–1634) combined the Senecan dramatic technique with the ideals of Stoic philosophy, thus being more Senecan than Seneca himself. Still more indebted to Seneca was John Marston (ca. 1575–1634; see Goldberg 2000: 218– 221), who in his Antonio and Mellida and Antonio’s Revenge not only followed Seneca in dramatic technique but also embedded quotations from Seneca’s tragedies.32 At the beginning of the seventeenth century, Seneca’s tragedies were regarded as classical models for modern tragedy, which, however, did not really influence writing for the stage, but fostered the writing of literary dramas. For the rest of the seventeenth century, Seneca’s influence in England was by far weaker than on the continent. As for Scandinavia, Seneca’s influence was mainly felt in Sweden. In 1648 Hercules furens was possibly staged in the presence of Queen Christine. Neither in Denmark, Norway, nor in Iceland are there any remarkable traces of a creative adaptation or philological preoccupation with Seneca. Regarding Slavic literature,33 in Poland Seneca was popular in the so-called school drama (Gregor Cnapius, ca. 1564–1638), but there is no influence on genuine Polish drama. Jan Alan Bardzinski (1657–1708) translated Seneca’s tragedies into Polish. As for Czech and Slovak literature we can only suppose that the Jesuits, at least, Karel Kolcava (1656–1717), for example, were influenced by Seneca. In Russia Seneca’s influence was not yet apparent in the seventeenth century.

31 “And then there is Shakespeare, whose plays rewrite Senecan scenes and speeches constantly” (Boyle 1997: 147); cf. also Braden 1985: 153–223, Miola 1992, Goldberg 2000: 214–218; and Bullough 1964. 32 See Boyle 1997: 144f. The Spanish Tragedy by Th. Kyd already contained several Latin quotations from Seneca (Boyle 1997: 143 f.). 33 See Busch 1978.

seneca the dramatist

The Eighteenth Century In the eighteenth century Seneca’s decline, which had already begun in the seventeenth century, became evident. This fits into a general and well-known development of which the causes need not be explained here: Latin literature was abandoned in favor of Greek texts, which were considered the only original literary heritage from antiquity. Italian poets, though, still adhered to the Latin tradition. In Pietro Metastasio’s dramas and libretti, for instance, which are still regarded as the best dramatic poetry of eighteenth-century Italy, we find several important elements of Senecan tragedy: the shortness and pointedness of the diction; the high estimation of ethical, especially Stoic, values; and the specific features of the so-called “prince’s mirrors.” In Italy, as in the whole of Europe, there was an intensified interchange between the Latin tradition and the new appraisal of Greek literature. The Italian dramatists, especially Vittorio Alfieri (1749–1803), turned away from French influences, but Seneca, whose tragedies were translated anew, was still imitated in contemporary tragedies. The “tyrant” was a common figure on stage, not only in Italy, but also in France, if we think of Crébillon’s Atrée et Thyeste (1707). Even more than before, Seneca’s Hercules dramas served as models or inspiration for tragedies. This was certainly influenced by the broad interest in philosophical thinking connected with the development of the Age of Enlightenment. In Germany, eighteenth-century rationalism, among other things, caused reflections on speech. Baroque bombast was anathematized. But in the field of dramatic theory, Seneca’s influence was still strong, especially on the development of the “Trauerspiel.” In Germany34 there was also a tendency to abandon the French examples of “antique” tragedies in favor of the Latin and Greek originals, as did Johann Elias Schlegel (1718–1749) in his Hecuba (1736), which was revised under the title Die Trojanerinnen in 1742. Schlegel made use of Euripides’s Troades and Hecabe and of Seneca’s Troades, creating his own version, not merely a patchwork. Some important exponents of that time show a shifting attitude toward Seneca: Gotthold Ephraim Lessing (see Barner 1973), for instance, had defended Seneca against his modern adversaries in his early years, but later turned away from him in favor of Greek tragedy. Seneca’s influence on Lessing’s writing is immediately perceptible in his Miss Sara Sampson in which the cruel Lady Marwood characterizes herself as a new Medea. Christian Felix Weiße’s (1726–1804) Atreus und Thyest (1766) is clearly influenced by Seneca’s Thyestes (Dammann

See Liebermann 1978.

werner schubert

2006). In the period of Sturm und Drang the dramatic pathos is comparable to Seneca’s. It is, above all, the Senecan Medea that inspired certain poets, such as Friedrich Maximilian Klinger (1752–1831) and, very important for the Senecan reception in music, Friedrich Wilhelm Gotter (1746–1797) in his Medea, which is almost entirely a single monologue, spoken by Medea, full of Senecan echoes; the composer Georg Anton Benda (1722–1795) adapted it in his melodrama Medea. There are also Senecan elements in Schiller’s Die Braut von Messina. In Spain, eighteenth-century classicism did not seek inspiration in Seneca, but in French tragedy of the seventeenth century. In the Netherlands, tragedy followed the same development. Sometimes the figure of Seneca himself was presented on the Dutch stage. There are at least two tragedies entitled De dood van Seneca (by C.G. Moering 1743 and J.A. Backer 1796); the German prose tragedy Seneca by Ewald von Kleist (1758) was also translated twice into Dutch under the titles De dood van Seneca and Seneka. In England, Seneca’s philosophical writing was still well known and read, but his influence on English drama was of no great importance. Greek tragedy gained more and more ground after John Dryden (1631–1700) made use of both Seneca and Euripides in his Oedipus. As for Poland, there are but few traces of a creative adaptation of Seneca’s tragedies; in southeast European literature we find some plays with ancient mythological plots, but it is not confirmed that these were influenced by Seneca. In Russia, Seneca’s Troades were translated by Mikhail V. Lomonosov (1711–1765), the eponym of Moscow University, whose drama Tamira i Selim (ca. 1750) shows Senecan influence. The Nineteenth Century In the nineteenth century, dramatists from almost everywhere were remarkably resistant to drawing from antiquity. On the other hand, original Greek and Latin plays were put on stage to a comparably large extent. But the modern drama as a whole had produced its own “classics,” which in the nineteenth century increasingly transcended the borders of national literatures. In Italy, for example, Goethe and Schiller were popular, German poets were fond of Shakespeare, and so on. Senecan influence is, if at all, to be observed only in those plays that deliberately turned to ancient subjects, as in Franz Grillparzer’s Das goldene Vließ. In this trilogy, however, the reception of Seneca is only part of a wider range of adaptation that was to become typical from that time up to the present; Grillparzer consulted Seneca’s Medea only as one source among others, including Euripides’s Medea and Apollonius Rhodius’s

seneca the dramatist

and Valerius Flaccus’s Argonaut epics. And it is significant that it was not primarily these texts that inspired Grillparzer, but a mythological lexicon he happened to come across. The German and Austrian “Schicksalsdrama” was, if at all, influenced by Greek tragedy, not by Seneca. In Spain there was no interest in Seneca’s tragedies as inventionis fons; there were, however, from the middle of the nineteenth century onward new translations there as elsewhere, including into Czech and Russian. In England, Seneca had no remarkable influence on the development of nineteenth-century tragedy. In the Netherlands, too, there are only single and discontinuous traces of reception as in S. Izn. Wiselius’s Polydorus (1813), which is partly based on passages from Seneca’s Troades. The interest in Senecan tragedy was almost wholly limited to his philological work, not only in the nineteenth but also in the twentieth century. The Twentieth Century The increasing accessibility of Seneca’s tragedies by way of translations was perhaps the reason why twentieth-century dramatists sometimes utilized Seneca again when they referred to ancient tragedy. In Italy, Gabriele d’Annunzio (Citti and Neri 2001: 93–96; Zwierlein 2006: 36–45), with his Fedra, showed a new interest in Seneca. The documents of (possible) receptions of Senecan tragedy are solitary, and often the Senecan influence is combined with that of others and depends on the choice of the subjects and their adaptations in the course of literary history. In France, Jean Anouilh (1910– 1987) took into consideration not only the Attic tragedians, but also Seneca, in his dramas based on ancient plots. The same must be said of Antonin Artaud (1896–1948), to whom Seneca’s “théâtre de la cruauté” naturally appealed. To Camille Claudel (1864–1943), Seneca was the greatest dramatist of all times. In Germany, Seneca was not as popular as in Italy or France. Although expressionistic mannerisms in German drama can be strikingly similar to Seneca’s, they do not result from any interaction with his texts. When ancient tragedy was reused for examples, it was the Greek literature that held sway. Franz Werfel (1890–1945), for instance, in his Troades recurred to Euripides’ Hekabe, not to Seneca’s Troades. In Spain, Miguel de Unamuno y Jugo (1864– 1936) in his Fedra followed Euripides and Racine, not Seneca (Citti and Neri 2001: 96f.), but he translated and performed Seneca’s Medea in 1933. In 1957 José María Pemán produced a version of the Thyestes theme, based on Seneca. In English literature, Thomas Stearns Eliot (1888–1965) wrote appreciative

werner schubert

essays on Seneca the tragedian (see Eliot 1934a and 1934b), but Seneca hardly had an effect on his own dramatic writing. In the Netherlands the picture is similar. On the whole, affinities to Senecan drama in twentieth-century writing are incidental; but there were some efforts to adapt Seneca’s tragedies in new versions for the stage, as did Hugo Claus (1929–2008) with Thyestes and Oedipus in Belgium.35 The Present At present Seneca’s tragedies have for some decades inspired new interest at least among philologists. To check the whole range of Senecan reception today is an easier task than it used to be. Information about performances of Seneca’s tragedies, not only on professional stages but also at schools or universities, that is readily available on the Internet or in audio-visual documents seems to reveal that today there is more interest in Seneca than in the last two centuries. But it is also possible that there were significant but forgotten or hidden activities at schools and universities that cannot be reconstructed or verified due to the lack of documents. New stagings36 of (or adapations of) Seneca’s dramas are rare when compared to the stagings of ancient Greek tragedies, but when there are any, they are noted in the newspapers as well as in philological reviews, as happened with Durs Grünbein’s (b. 1962) version of Seneca’s Thyestes (2001),37 which was performed at the Nationaltheater Mannheim, with the staging of Seneca’s Troades in Latin by members of the Institut für Klassische Philologie at the University of Munich in November 1993 (Stroh 1994; Volk 2000), or with the Latin version of Seneca’s Medea at Basel, Switzerland, in June 2000 (Lenz 2001), organized and performed by members of the Seminar für Klassische Philologie at the University of Basel.38 This discussion will close with a brief review of a special kind of reception of ancient drama in modern times: the musical adaptation.39 The (pseudo-)Senecan Octavia belongs to the infancy of the new musical genre, “opera”. Traces of Seneca are to be found in Charpentier’s Médée (Sinn 2008)

35 For further information on Seneca on the modern stage see the single contributions in Harrison (ed.) 2000a. See also Citti and Neri 2001: 81–148 (including the chapter “I tiranni senecani in O’Neill, Camus e Ayrton” 90–93); Coccia 2002; and Lenz 2001. 36 Cf. the chapter “Rappresentazioni teatrali” in Citti and Neri 2001: 82–87. 37 Cf. Reitz 2002 and Seidensticker 2002. 38 Cf. also Fantham 2000. 39 Draheim 1981 is an indispensable guide in this area.

seneca the dramatist

and Cherubini’s Médée (Trentin 2001) as well as in Mozart’s La clemenza di Tito (Seidel 1987). The structure of Seneca’s tragedy is mirrored by the structure and conciseness of many opera libretti (Schubert 2004: 371–392). In the twentieth century, there are composers who made use of Senecan texts taken from his tragedies without creating a stage play, as did Jan Novák, a Czech composer (1921–1984) who was a devotee of Latin literature. His impressive Planctus Troadum, based on the text of the first scene of Seneca’s Troades, was written in 1969, immediately after he had had to leave his home country in 1968 (see Schubert 2005: 182–188). Another very intense piece of music is Medea by Yannis Xenakis (1922–2001), in which the composer also does not make use of the drama as a whole but combines certain passages from the tragedy, mainly chorus parts. Perhaps it is this kind of adaptation that points to the future, at least in the field of musical works whose underlying texts very often are fragments intentionally broken off from an entire work or composed of fragments taken from different and sometimes remote sources and arranged into a new context (see Schubert 2004: 408–410). From antiquity until now, the popularity of Seneca’s dramatic oeuvre has often varied. So has the popularity of the single tragedies. But over the centuries five of them have proven to be of constant interest: Thyestes, Troades, Medea, Phaedra and Hercules furens.40 Although the other tragedies, Agamemnon, Oedipus, Phoenissae, and Hercules Oetaeus (and Octavia) have met with a considerably minor interest, Seneca’s Agamemnon and Oedipus served in the second half of the last century in the field of musical adaptations as starting points for the libretti of two operas composed by Josep Soler (Agamemnon 1960 and Edipo y Iocasta 1975). It is difficult to predict the reception of Seneca’s tragedies on stage, in concert halls, in libraries, or in lecture halls. To judge by the influence of Seneca’s tragedies up to the present there is hope that creativity will overcome sterility well into the future.

40 In the 20th century Phaedra and Medea are dominating within the adaptations of Seneca’s tragedies; cf. the chapter “Adattamenti, rifacimenti, riprese di singoli drammi” in Citti and Neri 2001: 93–148.

PART TWO

PHILOSOPHY

CONTEXT: SENECA’S PHILOSOPHICAL PREDECESSORS AND CONTEMPORARIES

John Sellars

1. Intellectual Background Lucius Annaeus Seneca lived, thought, and wrote during a complex and comparatively neglected period in the history of philosophy.1 The philosophical scene of the first century ad was quite different from the much better known philosophical culture of the Hellenistic period that came to an end in the previous century. During the third and second centuries bc, the majority of philosophical activity in the ancient world took place in Athens, just as it had during the days of Plato and Aristotle, and aspiring philosophers from all over the Eastern Mediterranean traveled to Athens where they could join in with the intellectual activity taking place at the Academy, Lyceum, Garden, and Painted Stoa. By the time of Seneca’s birth at the end of the first century bc, Athens was no longer the predominant center of philosophical activity in the ancient world; philosophy had undergone a process of dispersal and decentralization.2 Philosophical schools sprang up locally—in Rome, Alexandria, Rhodes, and no doubt elsewhere—and in Italy people had already started writing philosophy in Latin.3 Seneca first studied philosophy within this decentralized and bilingual philosophical climate, drawing upon both the earlier Greek tradition and this new contemporary situation.

1 For general studies of philosophy in Rome, see Griffin and Barnes 1989, Morford 2002, Trapp 2007b, and Sorabji and Sharples 2007. For studies of Stoicism in Rome, see the above plus Arnold 1911, Chevallier 1960, Haase 1989, Gill 2003, and Reydams-Schils 2005. For Seneca’s place within Roman Stoicism, see Grimal 1989. 2 On the decentralization of philosophy, see Frede 1999, Sedley 2003b, Ferrary 2007, and, with specific reference to the Stoa, Sedley 2003a: 24–32. 3 In the period before Seneca’s birth, the most famous examples of Latin philosophical texts were those of Cicero and Lucretius, but predating both of these were earlier Italian Epicureans who wrote in Latin, notably Amafinius, on whom see Cic. Tusc. 4.6f. with Ducos 1989 and Sedley 2009: 39 f.

john sellars

The tales of the rise of philosophy in Rome and the decline of philosophy in Athens are inevitably intertwined. The introduction of Greek philosophy into the Roman world is traditionally connected with the famous embassy of three Athenian philosophers who visited Rome in 155 bc, ostensibly there to ask for a fine imposed on Athens to be lifted, but since remembered for their fine oratory (and their beards).4 The earliest generations of Romans attracted to philosophy traveled to its natural home, Athens, in order to learn more. A little later, Cicero followed in their footsteps and went on to send his son to do the same, although by then the situation had already changed. A key moment in the transformation occurred in 88bc when Athens sided with King Mithridates against Rome and the city was subsequently put under siege by Sulla.5 Both the Academy and the Epicurean Garden were probably damaged, if not destroyed.6 Leading intellectuals fled the city, including the head of the Academy, Philo of Larissa.7 Some, like Philo, went to Rome, while others found a variety of new locations: Alexandria, Rhodes, and the Bay of Naples, to name the best known.8 Cicero observed first hand many of these upheavals. He welcomed exiled Athenian philosophers into his own home, and visited Athens in the aftermath, as well as new centers of philosophical activity that sprang up, such as Rhodes.9 Cicero also played a key role in the further decentralization of philosophy by writing popular accounts in Latin of the principal doctrines of the main Hellenistic schools, in the process laying the foundations for a comprehensive Latin philosophical vocabulary.10

4 On the embassy see, e.g., Gell. 6.14.8–10 and Plut. Cato Maior 22.1–3 with Griffin 1989: 2–5 and Ferrary 2007. On beards see Sellars 2003: 15–19. 5 On Athens and Mithridates see Posidonius apud Athen. 5.211d–215b (= frg. 253 Edelstein and Kidd 1972). On the siege of Sulla, see Plut. Sulla 12.1–13.4 and App. Mithr. 30–45. 6 Clay (2009: 27) suggests that both the Academy and Garden were destroyed. The evidence he cites (Plut. Sulla 12.3 and App. Mithr. 30) makes reference to the Academy, and Plutarch also mentions the Lyceum, but there is no explicit mention of the Garden. Nevertheless, the general descriptions of the siege certainly imply that the Garden, just outside the city walls and close to the Academy, would have suffered severely. See also Frede 1999: 790–793. 7 See Cic. Brut. 306. 8 Antiochus (the Academic), Posidonius (the Stoic), and Philodemus (the Epicurean) all studied in Athens around this time but left for Alexandria, Rhodes, and Herculaneum, respectively. 9 On Cicero’s travels to Athens and Rhodes, see Cic. Brut. 315f. and Plut. Cicero 4.1–4. For his reflections on the state of Athens, see fin. 5.1–5. 10 Previous philosophical work in Latin, by Amafinius and Lucretius, was limited to Epicureanism. In addition to his discussions of Stoic, Epicurean, and Academic doctrines, Cicero also produced a Latin version of Plato’s Timaeus. On Cicero and philosophy in Latin, see Levy 1992a.

seneca’s philosophical predecessors and contemporaries

Seneca’s own philosophical education took place in the aftermath of these dramatic changes. He wrote in Latin, but unlike previous Latin philosophical authors, such as Cicero and Lucretius, who simply made available to a Latin audience ideas derived from Greek philosophers, Seneca tried to do philosophy in Latin and for this reason it has been suggested that his works are the earliest properly Latin philosophical works that have come down to us.11 Indeed, they are the only properly Latin philosophical works to survive from pagan antiquity, for the majority of Seneca’s Roman contemporaries (Cornutus, Musonius Rufus) and successors (Marcus Aurelius) reverted to Greek for their philosophical writing. We have to wait until Augustine to find the next significant body of philosophical work in Latin. This turn to Latin no doubt reflects in part the fact that Seneca’s own philosophical education took place in Rome, as well as the fact that he came from the monolingual Western Mediterranean rather than the bilingual East, but it also reflects the much wider changes in philosophical culture outlined above.12 The decentralization of philosophy led to a number of flourishing philosophical communities in Italy, such as the circle of Epicureans around Philodemus in Herculaneum, while the works of Cicero and Lucretius opened up the Greek philosophical tradition to a new audience. Seneca’s philosophy was formed in a new specifically Roman intellectual context that would prove to be relatively short lived. 2. Seneca’s Teachers Seneca names three teachers with whom he studied philosophy: Papirius Fabianus, Sotion of Alexandria, and Attalus the Stoic.13 Taken together, these teachers reflect the transformed character of ancient philosophical culture. Attalus was from Pergamum in the East, while Fabianus and Sotion were

11 See Inwood (2005a: 13), “Seneca stands out for his striking choice to do what I would call primary philosophy (rather than exegetical or missionary work) in Latin”; also ibid. 20, “Seneca, much more than Cicero, is thinking creatively and philosophically in Latin.” However, Seneca also complained of Latin’s limitations (epist. 58.1), echoing the earlier complaint of Lucretius (1.136–139). On philosophy in Latin, see Grimal 1992a. 12 Seneca clearly knew Greek (he translates some lines from Cleanthes at epist. 107.10f.), but it seems reasonable to presume that it was learned in the classroom and as such would not have been his natural medium of thought. 13 Seneca is himself our principal source of evidence for his teachers. For discussion of his three teachers and the school of the Sextii, see Zeller 1880: III.1, 675–682 (trans. in Zeller 1883: 180–188), Grimal 1978a: 247–262, Fillion-Lahille 1984: 256–272, Lana 1992, and Inwood 2005a: 1–22.

100

john sellars

both products of a Roman school of philosophy founded by Quintus Sextius. Via both his pupils and his writings, Sextius also proved to be an important influence on Seneca, so it may be appropriate to begin with him.14 Sextius was the founder of his own philosophical school in Rome and it was probably handed down to his son, Sextius Niger, although Seneca reports that the school did not last for very long.15 Sextius wrote his philosophy in Greek, but combined this with a Roman sensibility.16 He also drew upon both Stoic and Pythagorean doctrines, and Seneca characterizes him as a Stoic, while noting that Sextius himself rejected the label.17 It is tempting to imagine someone in the mold of Cato the Younger, embodying the implicit Stoic values of traditional Roman morality, and such a comparison is given some credence by the fact that Sextius refused to accept an offer of public office from Julius Caesar.18 The Pythagorean influence on Sextius manifested itself in vegetarianism and the practice of daily self-examination, both of which Seneca adopted.19 Indeed, Seneca appears to have admired and emulated Sextius greatly. From what we know, Sextius combined a focus on practical ethical concerns with continual self-examination, ascetic training, and a broadly Stoic worldview, without accepting the limitations of doctrinal conformity. All of this is highly reminiscent of Seneca himself, and it also prefigures many of the features of subsequent Imperial Stoicism, such as we find in the Meditations of Marcus Aurelius. Although only a shadowy figure to us, it may be that Sextius influenced the subsequent development of Roman Stoicism far more profoundly than has hitherto been noted.20

14 Seneca mentions Sextius at epist. 59.7, 64.2–5, 73.12, 73.15, 98.13, 108.17–19, dial. 4 (= de ira 2).36.1, 5 (= de ira 3).36.1, nat. 7.32.2. 15 On the school of Sextius, see Zeller 1880: III.1, 675–682 (trans. in Zeller 1883: 180–188), Griffin 1976: 37–42, Fillion-Lahille 1984: 256f., Lana 1992, and Morford 2002: 133f. On its specifically Roman character, the plurality of Sextii, and its short life, see nat. 7.32.2. On Quintus Sextius as pater, see epist. 64.2, 98.13. 16 See epist. 59.7. 17 See epist. 64.2. On the Pythagorean influence on Sextius, see epist. 108.17f. with Kahn 2001: 92 f. 18 See epist. 98.13; see also Plut. mor. 77e. 19 See epist. 108.22 and dial. 5 (= de ira 3).36.1, respectively, with Kahn 2001: 92f. for further discussion. Seneca later dropped the vegetarianism (epist. 108.22). 20 Having said that, Inwood (2007c: 139) refers to the “widespread but misleading impression that later Stoicism is concerned excessively with ethics” and he notes late Stoic texts concerned with physics such as Seneca’s Naturales quaestiones, Cornutus’s Theologiae graecae compendium, and Cleomedes’s Caelestia. While Inwood is right to remind us of these texts, I still think there is room for a strong Sextian influence on later Stoicism. First, we might note that despite a focus on ethical matters within the school, the Sextian Fabianus also had strong interests in physics and may well have been a key influence behind theNaturales

seneca’s philosophical predecessors and contemporaries

101

Two of Sextius’s pupils contributed to Seneca’s own education. The first of these, Papirius Fabianus, noted as a rhetorician as well as a philosopher, is said to have written more works of philosophy than Cicero, although all of these are now lost.21 He took from Sextius a skepticism about obscure theoretical studies (or at least about their value for practical matters), but retained a healthy interest in physics, writing a book on natural causes.22 Seneca also reports a book on politics.23 This suggests a range of interests not too dissimilar from Seneca’s own. The second pupil, Sotion of Alexandria, also followed Sextius’s Pythagorean habits of vegetarianism and self-examination.24 We have limited information about Sotion, but a fragment from his work Περὶ ὀργῆς, preserved by Stobaeus, parallels material in Seneca’s De ira, suggesting an influence.25 Indeed, it is worth noting that Seneca also cites Fabianus on how to cure emotions, and so this topic may well have been a wider preoccupation of the Sextian school.26 It is striking how closely the interests of the Sextian philosophers correlate with Seneca’s own concerns, to the point that it is tempting to think of Seneca as simply an ex officio member of the school. However, by way of caution, it should also be remembered that almost all of our information about the Sextians comes from Seneca himself and so is no doubt to some extent colored by his own interests.

quaestiones (see below). Second, Sextius’s adoption of Pythagorean ascetic practices prefigures the focus on mental training (or “spiritual exercises” or “techniques of the self”) that we find in Seneca, Musonius Rufus, Epictetus, and Marcus Aurelius and that some have argued sits rather uneasily alongside the Stoic monistic psychology inherited from Socrates. One person who has hinted at the influence of Sextius is Frede (1999: 787). 21 Seneca mentions Fabianus at epist. 11.4, 40.12, 52.11, 58.6, 100.1–12, dial. 6 (= cons. Marc.).23.5, dial. 10 (= brev.).10.1, 13.9, nat. 3.27.3. He is discussed by Seneca’s father, in contr. 2. pr. 1–4, who also purports to record extracts of his declamations throughout Book 2 of the Controversiae. On the quantity of his writing, see epist. 100.9. See further Fillion-Lahille 1984: 258 f., Lana 1992: 117–122, and Ducos 2000. 22 For his skepticism, see dial. 10 (= brev.).10.1, 13.9; for his work Libri causarum naturalium, see Charis. 135,19–23 (note also 134,13; 190,8; 186,6 Barwick), which Seneca may well be citing at nat. 3.27.3. 23 See epist. 100.1. 24 On Sotion see epist. 49.2 and 108.17–22. 25 Compare Stob. 3.550.7–17 (Wachsmuth-Hense) with Sen. dial. 4 (= de ira 2).10.5 and see the discussions in Fillion-Lahille 1984: 261–272 and 1989: 1632–1636, who suggests that Sotion is the third key influence on De ira after Chrysippus and Posidonius. 26 The two passages where Seneca cites Fabianus’s skepticism toward technicality and sophistry (dial. 10 [= brev.].10.1, 13.9) are both concerned with the emotions. Fabianus’s point, as Seneca reports it, is that overcoming powerful emotions requires more than mere theoretical subtlety.

102

john sellars

Seneca’s third teacher, Attalus, differed from the Sextians insofar as he was a self-proclaimed Stoic.27 He came to Rome from Pergamum, a noted center of Stoic activity in the new, decentralized philosophical world.28 Seneca records that he was a regular attendee at the lectures of Attalus, “the first to arrive and the last to leave,” and that he used to talk with him outside the classroom as well.29 It was from Attalus that Seneca first learned his Stoicism and, notwithstanding the impact of the Sextians, Seneca always described himself as a Stoic, never a Sextian.30 Having said that, Attalus and the Sextians had much in common, including a preference for simplicity in life and the practice of regular self-examination, both themes that would mark Seneca’s own work. 3. Stoicism It was from Attalus, then, that Seneca probably gained his introduction to Stoicism. Whatever Stoic texts may have been available in Rome at the time, Attalus would presumably have had access to even more during his time in Pergamum, so it seems reasonable to assume that Seneca would have had access to a wide range of Stoic material, even if some of his knowledge was only secondhand.31 One of the consequences of the decentralization of philosophy that took place in the previous century was an increased focus on the foundational texts of each philosophical school by its members.32 It seems likely that Seneca would have spent much time reading the canonical texts of

27 On Attalus as a Stoic, see epist. 67.15 and Seneca the Elder suas. 2.12. Seneca mentions Attalus at epist. 9.7, 63.5, 67.15, 72.8, 81.22, 108.2f., 108,13–16, 108,23, 110.14–20, nat. 2.48.2, 2.1.1–3. For further discussion, see Fillion-Lahille 1984: 260 f., and Follet 1989. 28 On Stoics in Pergamum, see Pfeiffer 1968: 234–251. The first Stoic associated with Pergamum was Crates of Mallus, who moved there at the invitation of King Eumenes II. The most famous Stoic associated with the place was Athenodorus Cordylion of Tarsus, librarian and expurgator of Zeno’s Republic (Diog. Laert. 7.34), who was visited by Cato and traveled with him to Rome (Plut. Cato Minor 10.1, 16.1; mor. 777a). This is an example of the decentralization of philosophy commencing well before the siege of Athens. 29 See epist. 108.3. 30 At nat. 7.32.2 Seneca refers to the Sextians alongside the Academy and the school of Pythagoras as if it were a distinct philosophical school to which one might claim to belong. 31 We know that quite technical Stoic texts did make it to Italy by this time, thanks to the fragments of Chrysippus’s Λογικὰ ζητήµατα found at Herculaneum (PHerc 307). Note also Cicero’s reference to Lucullus’s library of Stoic texts (fin. 3.7) and the report that Seneca’s younger contemporary Persius owned a substantial collection of works by Chrysippus (Suet. Vit. Pers.). 32 On this, see Sedley 2003b: 36f.

seneca’s philosophical predecessors and contemporaries 103 the early Stoics.33 There is some evidence of such reading in his philosophical works. The founder of the Athenian Stoa was, of course, Zeno of Citium, and Seneca mentions him throughout his works.34 He also quotes from “our Zeno” (Zenon noster) a number of times, although in a number of cases only to mock his syllogisms, and it is difficult to detect any specific influence.35 Seneca also had access to texts by Zeno’s pupil and successor as Scholarch, Cleanthes of Assos, and he translates into Latin some lines from Cleanthes for the benefit of Lucilius.36 The most important of the early Stoics, however, was Cleanthes’s successor, Chrysippus of Soli. Seneca mentions him frequently,37 usually in glowing terms, although he is willing to criticize Chrysippus where necessary.38 Given the importance of Chrysippus to subsequent Stoics, a number of scholars have tried to point to Chrysippean sources behind some of Seneca’s works, especially his longer essays such as De beneficiis and De ira.39 However, in general there is little explicit debt to or sustained engagement with the Stoa’s canonical early texts. Occasional quotations and allusions suggest familiarity but there is no effort on Seneca’s part to join the slowly developing commentary tradition within ancient philosophy. Instead, like his Stoic compatriot Epictetus a little later, Seneca warns against becoming a philologist at the expense of philosophy, which, again like Epictetus, is for

33 A little later we find a number of passages in Arrian’s reports of Epictetus’s lectures that indicate that much time was spent reading through works by Chrysippus, although Epictetus warns against forsaking philosophy for philology; see, e.g., Epictetus diss. 1.4.14, 1.17.13–18, 2.23.44. 34 Seneca mentions Zeno at epist. 6.6, 22.11, 33.4, 33.7, 33.9, 64.10, 82.9, 83.9–11, 83.17, 104.21, 108.38, dial. 3 (= de ira 1).16.7, dial. 7 (= vit. beat.).18.1, dial. 8 (= de otio).1.4, 3.1, 3.2, 6.4, 6.5, dial. 9 (= tranq.).1.10, 14.3, dial. 10 (= brev.).14.5, dial. 12 (= cons. Helv.).12.4, benef. 4.39.1f., 7.8.2, nat. 7.19.1. 35 See, e.g., epist. 82.9 (= SVF 1.196), 83.9 (= SVF 1.229); note also dial. 3 (= de ira 1).16.7 (= SVF 1.215), dial. 8 (= de otio).3.2 (= SVF 1.271), dial. 9 (= tranq.).14.3 (= SVF 1.277). 36 See epist. 107.10f. (= SVF 1.527). The same lines are preserved in Greek in Epictetus Ench. 53 (= SVF ibid.). See also epist. 108.10 (= SVF 1.487) and benef. 5.14.1 (= SVF 1.580). Beyond these passages, Seneca also mentions Cleanthes at epist. 6.6, 33.4, 33,7f., 44.3, 64.10, 94.4f., 113.23, dial. 8 (= de otio).6.5, dial. 9 (= tranq.).1.10, benef. 6.11.1 f., 6.12.2. 37 Seneca mentions Chrysippus at epist. 9.14, 22.11, 33.4, 56.3, 104.22, 108.38, 113.23, dial. 2 (= const.).17.1, dial. 8 (= de otio).3.1, 6.4f., 8.1, dial. 9 (= tranq.).1.10, benef. 1.3.8f., 1.4.1, 2.17.3, 2.25.3, 3.22.1, 7.8.2. 38 See benef. 1.3.8–4.4 (= SVF 2.1082). 39 For a Chrysippean source behind De ira see Fillion-Lahille 1984: 51–118 and 1989: 1619– 1626, with critical discussion in Inwood (2005a: 27ff.). Braund (2009: 22) suggests a Chrysippean influence behind De beneficiis, citing Chaumartin (1985), although in fact the latter suggests that the main source standing behind De beneficiis is the Περὶ Καρίτων of Hecaton. Seneca mentions both Chrysippus and Hecaton at benef. 1.3.8 f.

104

john sellars

him above all a way of life. 40 Seneca’s debt to the early Stoa is, then, a broad philosophical debt of the sort shared by any admirer of the Stoic philosophy, but it is nevertheless a genuine debt to the orthodox Stoicism exemplified by Chrysippus.41 Just like any Stoic, Seneca owes a debt to the early Stoa. But what about the so-called “Middle Stoa”? The division between an Early and a Middle Stoa has recently been called into question.42 On the traditional view, members of the Middle Stoa watered down the high ideals of the Early Stoa, shifting focus from the moral perfectionism embodied in the idealized sage to the everyday moral concerns of real individuals. This shift in concern is most evident in the views of Panaetius, and Seneca is one of our key sources here: I think that Panaetius gave a very neat answer to a certain youth who asked him whether the wise man should become a lover: “As to the wise man, we shall see later; but you and I, who are as yet far removed from wisdom, should not trust ourselves to fall into a state that is disordered, uncontrolled, enslaved to another, contemptible to itself.”43

Seneca cites this with approval. Cicero suggests that this shift in focus evident with Panaetius, and perhaps initiated by Diogenes of Babylon, was the product of the influence of Plato and Aristotle on these two Stoics.44 He adds that the shift in emphasis in ethics was mirrored by a similar shift in political philosophy as well.45 These two shifts are evident throughout Seneca’s work: a practical concern with moral improvement for the imperfect and a pragmatic desire to engage in the messy world of real politics. Do these two features of Seneca’s work indicate the influence of Panaetius?

40 For Seneca’s famous warning against philology, see epist. 108.23 (within the context of reminiscing about the examples set by Sotion and Attalus). For parallel sentiments in Epictetus, see n. 33 supra. For how the Stoics conceived philosophy, see Sellars 2003. 41 Inwood (2005a: 47f.) argues for Seneca’s orthodoxy and agreement with Chrysippus on issues relating to psychology and the emotions, against charges of innovation; compare dial. 3 (= de ira 1).7.4 with Chrysippus apud Galen Plac. Hipp. et Plat. 4.2.8–18 (= SVF 3.462). 42 The idea of a distinctive “Middle Stoa” inaugurated by Panaetius was first proposed in Schmekel 1892; see Dyck 1996: 17. For a recent questioning of the notion, see Sedley 2003a: 23f. 43 Epist. 116.5 (= Panaetius frg. 114 van Straaten 1952): Eleganter mihi videtur Panaetius respondisse adulescentulo cuidam quaerenti an sapiens amaturus esset. “De sapiente” inquit “videbimus: mihi et tibi, qui adhuc a sapiente longe absumus, non est committendum ut incidamus in rem commotam, inpotentem, alteri emancupatam, vilem sibi.” 44 See Cic. fin. 4.79 (= frg. 55 van Straaten 1952), Tusc. 1.79 (= frg. 56/83 van Straaten 1952). On the influence of Plato and Aristotle, see Frede 1999: 782–785. 45 See Cic. leg. 3.13f. (= frg. 48/61 van Straaten 1952). On the political shift, see Sellars 2007: 20–24.

seneca’s philosophical predecessors and contemporaries 105 While some have argued that Seneca is following Panaetius here,46 others have suggested that these apparent shifts away from the orthodox Stoa merely illustrate Seneca’s own “epistemic humility”: Seneca prefers to focus on those things to which he has ready access via his everyday experience.47 One thing is clear, however, namely that Seneca rarely mentions Panaetius in his works compared with his frequent references to the early Stoic triumvirate of Zeno, Cleanthes, and Chrysippus. Beyond the quotation cited above, Seneca mentions Panaetius just twice and neither instance is significant.48 There is, then, little explicit evidence for a strong direct Panaetian influence on Seneca. In marked contrast to this lack of engagement with Panetius, Seneca often mentions and quotes from Posidonius, especially in the Epistulae morales and the Naturales quaestiones.49 Posidonius’s interests in physical phenomena are well attested, so it should come as no surprise to find Seneca drawing on his work in the Naturales quaestiones, and within the same context Seneca also draws on the work of Posidonius’s pupil Asclepiodotus.50 Beyond these explicit references, a number of scholars have argued that Posidonius forms an important implicit source for Seneca’s discussion of anger in the second book of De ira.51 The reason for positing a Posidonian influence in this work seems to have been to explain the apparently dualistic turn that Seneca takes here in his psychology. However, it has recently been argued that no such turn exists and Seneca’s position is more orthodox than some have supposed.52 Consequently, there is no need to posit a hidden Posidonian source behind Seneca’s text. Nevertheless, we can see that in general Seneca makes good

See, e.g., Cooper and Procopé 1995: xvii. See Inwood 2005a: 3. 48 See epist. 33.4 (= frg. 53 van Straaten 1952) and nat. 7.30.2 (= frg. 75 van Straaten 1952). In the former, his name appears in a list of Stoics; in the latter he is cited for his view on comets. Neither offers evidence for an influence of the sort under discussion. 49 Seneca mentions Posidonius at epist. 33.4, 83.10, 88.21–28, 90.5, 90.7–13, 90.20–25, 90.30– 32, 95.65 f., 104.22, 108.38, 121.1 and nat. 2.26.4, 2.26.6, 2.54.1–3, 4.3.2, 6.17.3, 6.21.2, 6.24.6, 7.20.1, 7.20.4, 7.21.1, and he quotes Posidonius at epist. 78.28, 90.7, 90.22f., 90.25, 90.31f., 94.38, 113.28 and nat. 1.5.10, 1.5.12. 50 Asclepiodotus is cited by Seneca at nat. 2.26.2, 2.26.6, 2.30.1, 5.15.1, 6.17.3, 6.22.2. On this Asclepiodotus, one of a number of ancient philosophers with that name, see Goulet 1989. He is thought to be the author of a short extant treatise on military tactics. 51 For example, Holler 1934: 16–24 (with Inwood 2005a: 41f.) and Fillion-Lahille 1984: 121–199 and 1989: 1626–1632 (with Inwood 2005a: 28, 33). 52 See Inwood 2005a: 23–64. There is also the question of just how heterodox Posidonius’s psychology really was, and whether our principal source for his views, Galen, is entirely reliable, on which see Gill 2006: 266–290. 47

106

john sellars

use of material from Posidonius, citing him as an authoritative Stoic standing alongside Zeno, Cleanthes, and Chrysippus.53 4. Other Greek Philosophical Influences Alongside Seneca’s obvious debt to a number of earlier Stoics, his philosophical works also include references to a wide array of other ancient philosophers.54 It will not be possible to discuss all of these here, many of which are of limited significance. Of those that are of greater significance, the one that has attracted the most attention is Epicurus.55 Many readers have noticed that Seneca frequently quotes from Epicurus in his correspondence with Lucilius, especially in the early letters.56 Indeed, Epicurus is mentioned more often than any other philosopher in Seneca’s prose works, and we might also note that Lucretius is the most cited poet.57 How are we to explain this taste for Epicureanism in the works of a supposedly committed Stoic? A number of explanations have been offered. One is to call into question Seneca’s commitment to the Stoa and brand him a philosophically muddled eclectic.58 Another is to suggest a complex pedagogic strategy within the correspondence to Lucilius, in which Seneca gently tries to draw in his Epicurean addressee at the opening of the exchange by offering him familiar tidbits.59 A

53 Or sometimes not citing him, but simply listing him alongside other eminent Stoics when making the point that philosophy ought not to rely upon quotations from authorities; see, e.g., epist. 33.3f. (= T54 Edelstein and Kidd 1972), 108.36–38 (= T55 Edelstein and Kidd 1972). 54 For a complete annotated list of Seneca’s references to other philosophers, see Motto 1970: 143–160. 55 Seneca mentions Epicurus too often to list them all here; for a complete list, see Motto 1970: 150 f. 56 Any attempt to discuss the motivations at work behind the Epistulae morales will open up questions about the status of these texts, such as whether the correspondence is genuine. It is not possible to address this issue here, on which there is a considerable literature. For a helpful overview of the status quaestionis and further references, see Inwood 2007a: xii–xv, with further recent discussion in Wilson 2001 and Inwood 2007c. Inwood follows Griffin (1976: 416–419) in claiming that the correspondence is “essentially fictitious” (Inwood 2007c: 134). The matter is complicated further by the fact that the collection of letters that has come down to us appears to be incomplete (on which see Reynolds 1965a: 17). 57 For a list of Seneca’s references to Lucretius see Motto 1970: 26. Braund (2009: 28–30) suggests a further potential Epicurean influence on Seneca, in the form of Philodemus in the De clementia. 58 See Rist 1989 for a discussion of Seneca’s status as a Stoic, although Rist doesn’t explicitly address the Epicurean element within the letters to Lucilius. 59 See, e.g., Hadot 1995: 210.

seneca’s philosophical predecessors and contemporaries 107 third suggests a subtle literary nod to Epicurus in order to indicate to readers that the correspondence with Lucilius is consciously modeled on the form (though not the content) of Epicurus’s own philosophical letters.60 It may well be that the reason is far more prosaic; Seneca simply happened to be reading Epicurus at the time that he started the correspondence and wanted to share some of what he found. We needn’t take this as a sign of burgeoning eclecticism either, for, as Seneca himself often says, Epicurus’s wisdom is not the sole property of his disciples but rather belongs to all humankind.61 The fact that Seneca feels the need to make these apologies to Lucilius for quoting Epicurus also counts against the suggestion that Lucilius was an Epicurean waiting to be converted.62 The Epicurean apophthegms that Seneca does share function more as examples of generic philosophical wisdom than samples of specifically Epicurean doctrine. If they do have a pedagogic function then it is more likely as part of an exhortation to the philosophical life as such. It is also worth noting that beyond the correspondence with Lucilius Seneca can often be quite hostile toward Epicurus.63 Looking back further, beyond the Hellenistic schools, we see fairly limited interest in or engagement with the great Athenian philosophers that came before: Socrates, Plato, and Aristotle. Seneca mentions Socrates a number of times, usually as an idealized model of a philosopher, but there is little by way of explicit philosophical influence beyond the wider Socratic flavor of the Stoicism that permeates his work as a whole.64 This is in marked contrast to Epictetus a few decades later, for whom Socrates figures far more prominently.65 Seneca also has little to say explicitly about Plato, beyond a series of generally praiseworthy remarks,66 but he does engage with the Platonism of his day in one of his few forays into metaphysics.67 In Letter 58 Seneca discusses the nature of being (τὸ ὄν) and recounts a Platonic hierarchy of

See Inwood 2007a: xiv and 2007c: 142–146. See, e.g., epist. 8.8, 12.11, 14.17, 16.7, 21.9, 33.2. These remarks appear in the opening part of the correspondence where Seneca quotes Epicurus most often. It’s also worth noting that Seneca does not restrict himself to Epicurus and often mentions other leading Epicureans such as Hermarchus and Metrodorus (see, e.g., epist. 6.6, 33.4). 62 For the claim that Lucilius was aphilosophical, see Motto 1970: xvii n. 25. 63 See, e.g., benef. 4.19.1–4. 64 For a full list of Seneca’s references to Socrates, see Motto 1970: 156–158. 65 For Epictetus’s debt to Socrates, see Long 2002. 66 See the list in Motto 1970: 154 f. 67 The key texts here are epist. 58 and 65, on which see Rist 1989: 2010f., Sedley 2005a: 122–138, Inwood 2007b (with references to earlier literature in Sedley 2005a: 122 n. 13). 61

108

john sellars

six senses of being, in marked opposition to the orthodox Stoic account in which being is limited to bodies and (along with certain incorporeals denied being) subsumed under a higher genus of “something” (τί). However, rather than merely oppose this Platonic account to the Stoic position that we might expect him to hold, it has been suggested that Seneca’s account is the syncretic product of a dialogue between the two schools: elements of Stoic ontology are now incorporated within a Platonic schema.68 However, Seneca’s stated aim in the letter is simply to present to Lucilius Plato’s account of being, and he doesn’t explicitly commit himself to holding the account he presents.69 A little later in the correspondence, in Letter 65, Seneca returns to Platonic metaphysics, and is critical of both Platonic and Aristotelian accounts of cause, although the position he outlines in response is not that of an orthodox Stoic.70 There is little general evidence for the claim that Seneca was drawn particularly to Platonism.71 As we have just seen, Seneca also engages with Aristotle in his works.72 He straightforwardly rejects the central Peripatetic doctrines on the emotions and the significance of externals, in just the way that one would expect an orthodox Stoic to do.73 But Seneca does make good use of Aristotle’s meteorological research in the Naturales quaestiones,74 and in the same place he also draws on the physical researches of Aristotle’s pupil Theophrastus.75 It is in the Naturales quaestiones that we also see Seneca draw on material from the Presocratics, mainly for their physical theories, and he cites Thales, Anaxagoras, Anaximander, and Anaximenes, among others.76

See, e.g., Sedley 2005a: 125. See, e.g., epist. 58.16. 70 At epist. 65.11, for instance, Seneca suggests that time and place must be counted among causes. 71 Rist (1989: 2010) claims that “Seneca’s ‘unorthodoxies’ tend towards Platonism.” It has also been suggested that Seneca Platonizes when he discusses the soul and body, in, e.g., epist. 92.1 f., although this seems mistaken; see Inwood 2005a: 38–41. 72 For a list of Seneca’s references to Aristotle, see Motto 1970: 145. 73 On anger, see, e.g., dial. 3 (= de ira 1).9.2, 17.1, dial. 5 (= de ira 3).3.1; on externals, see benef. 5.13.1. 74 See nat. 1.1.2, 1.3.7, 1.8.6, 2.12.4–6, 6.13.1 f., 6.14.1, 7.5.4, 7.30.1. 75 See nat. 3.11.2–5, 3.16.5, 3.25.4, 3.25.7, 3.26.2, 4.2.16, 6.13.1f., 7.28.3. 76 See, e.g., nat. 2.12.3, 2.18.1, 2.19.1, 3.13.1, 3.14.1–3, 4.2.17, 4.2.22, 4.3.6, 6.6.1–4, 6.9.1f., 6.10.1f., 7.5.3. We have already noted a Pythagorean influence on Seneca, via Sextius; see n. 17 supra. 69

seneca’s philosophical predecessors and contemporaries 109 5. Orthodoxy The presence of this wide variety of philosophical influences upon Seneca has led some to ask whether Seneca is in fact an orthodox Stoic.77 It seems fairly clear that Seneca is a Stoic, although one open to outside influences.78 This openness to other philosophical influences has traditionally been held to be a characteristic of the so-called Middle Stoa exemplified by Panaetius and Posidonius.79 However, recent scholarship has argued that this openness was a feature of Stoicism from the very beginning, and that the innovation supposedly introduced by Panaetius has been overstated.80 Indeed, Seneca himself notes disagreements between Cleanthes and Chrysippus in order to justify his own independence of thought while remaining a committed member of the Stoic tradition.81 So, in this respect it seems that Seneca does not deviate from many of his Stoic predecessors. As he himself writes, “we [Stoics] are not subject to a despot.”82 He is a committed Stoic, without being an unthinking disciple of Chrysippus.83 6. Contemporaries Having considered the most important of Seneca’s predecessors, let us now turn briefly to consider some of his contemporaries, beginning with those whom he knew very well. Seneca’s immediate circle included a number of philosophers and poets who shared his Stoic outlook. The most significant of these was probably Lucius Annaeus Cornutus, born in Libya and possibly a slave in Seneca’s household.84 Cornutus wrote philosophical, grammatical, and rhetorical works, in both Greek and Latin, of which his Theologiae graecae compendium survives, offering allegorical interpretations of traditional Greek mythology, following a tradition already well established within the Stoa.85 He

See, e.g., Rist 1989. See Inwood 2005a: 2. 79 See n. 42 supra. 80 See esp. Sedley 2003a: 23 f. 81 See epist. 113.23 (= SVF 1.525; 2.836). 82 Epist. 33.4: non sumus sub rege. See also dial. 8 (= de otio).3.1. 83 On Stoicism and the question of orthodoxy, see Sellars 2006: 10 f. For a concise defence of Seneca as a Stoic and a philosopher, see Inwood 2007a: xix. 84 For biographical information, see Fuentes González 1994: 462–466. 85 On the Compendium, see, e.g., Most 1989 and Boys-Stones 2007, who both supply references to further literature. For fragments of his other works, see Reppe 1906. 78

110

john sellars

is also known to have dabbled in metaphysics. 86 Like Seneca, Cornutus was exiled from Rome by Nero, but before his exile Cornutus taught in the city and his pupils included two famous Stoics within Seneca’s circle: Lucan and Persius. Lucan, author of Pharsalia, was the son of Seneca’s younger brother, Lucius Annaeus Mela. His epic poem draws on a number of Stoic themes, but perhaps the most striking Stoic element in the Pharsalia is the portrait of Cato the Younger, by now canonized as the archetypal example of a Roman Stoic sage.87 In his admiration for Cato, Lucan followed his uncle. Within the same circle around Seneca we also find the satirist Persius who, after losing his father, became a charge of Cornutus and a friend of Lucan.88 Persius dedicated his fifth satire to his teacher Cornutus and, after dying young, left to Cornutus both his library and the task of posthumously editing his works. In his philosophical outlook Persius was a committed Stoic, embracing a rigorous version of Stoicism that he may have contrasted with Seneca’s supposedly more moderate Stoicism. Beyond this immediate circle of Stoics, we also know of other contemporary philosophers who contributed to Seneca’s intellectual world. The most important was probably Demetrius the Cynic, who has been described as one of Seneca’s “living heroes,” and whose influence may well have contributed to Seneca’s taste for practical moral exhortation over formal philosophical argument.89 Seneca admired Demetrius’s poverty and simplicity of life, as well as his commitment to conceiving of philosophy as a practical guide to living. This admiration for an austere Cynic contrasts with the common image of Seneca as a moderate Stoic, some distance from the more rigorous end of the Stoa.90

86 See Sedley (2005a: 117), who notes the survival of a book title attributed to Cornutus in POxy 3649, Περὶ ἑκτῶν β (on which, see Cockle 1984: 12 f.). As well as this hint at a concern with metaphysics, Cornutus is also reported to have written a work entitled Against Athenodorus and Aristotle, responding to Athenodorus’s work Against Aristotle’s Categories. See Porphyry in Cat. 86,23 f. and Simplicius in Cat. 62,24 with Hijmans 1975: 106–109. 87 On Stoic themes in Lucan, see Colish 1990: 252–275, with references to further literature at 252 f. 88 On Persius and his Stoicism, see Colish 1990: 194–203. 89 Seneca mentions Demetrius at epist. 20.9, 62.3, 67.14, 91.19, dial. 1 (= prov.).3.3, 5.5 f., dial. 7 (= vit. beat.).18.3, nat. 4 prol. 7f., benef. 7.1.3–7, 7.2.1, 7.8.2, 7.11.1f. On Demetrius, see Billerbeck 1979 and, within the wider context of Roman Cynicism, Goulet-Cazé 1990: 2768–2773, Griffin 1996, and Trapp 2007a. The phrase “living heroes” comes from Griffin 2007: 89. On exhortation over argument, see Griffin 1996: 200. 90 While Griffin stresses Seneca’s admiration for Demetrius, Inwood (2005a: 16) suggests that his influence on Seneca was probably minimal, citing Demetrius’s “argument against the study of physical problems” reported by Seneca himself at benef. 7.1.5 and contrasting with

seneca’s philosophical predecessors and contemporaries

111

Alongside these figures, whom Seneca knew personally, a number of other contemporaries deserve a brief mention.91 The first of these is the Stoic Chaeremon who, like Seneca, is reported to have taught the Emperor Nero.92 If this is so then Seneca and Chaeremon may have met.93 Like Cornutus, Chaeremon followed the Stoic practice of offering allegorical interpretations of myths, in this case drawn from Egyptian religion, reflecting his position as a scholar in Alexandria. Although Seneca and Chaeremon may have met and may have appreciated one another as fellow Stoics, they inhabited quite different intellectual worlds. A second noteworthy Stoic of the first century is Musonius Rufus, with whom Seneca may have had more in common.94 Musonius came from an Etruscan family of the equestrian order. Like Seneca, Persius, Demetrius, and many others, Musonius found himself a victim of Nero, banished to the island of Gyara. He later brought charges against Publius Egnatius Celer for the latter’s involvement in the deaths of the Stoics Thrasea Paetus and Barea Soranus.95 Celer was defended by Demetrius, creating the odd spectacle of a Stoic and Cynic fighting in opposing corners of the court. Musonius’s philosophy was resolutely practical and exercised an important influence on Epictetus. In particular it displayed a concern with self-examination and practical training of the sort exemplified by the Sextians that would prove to be such an important influence on Seneca. However, there is no evidence of any direct contact between Seneca and Musonius, or of any indirect influence. The diatribes that have come down to us under Musonius’s name are written in Greek, marking a return to Greek as the natural language of philosophy. 96 The century or so of Latin philosophy

Seneca’s own interest in such problems in the Naturales quaestiones. However, Demetrius doesn’t argue against the study of nature, he simply notes that some details may be passed over without great loss: non multum tibi nocebit transisse, quae nec licet scire nec prodest. 91 It seems likely that Seneca knew Demetrius personally, although we cannot be sure; see Griffin 1976: 311 f. 92 See Suda s.v. ᾽Αλέξανδρος Αἰγαῖος = Chaeremon Test. 3 (in Horst 1987: 2). On Chaeremon, see Horst 1987 (containing fragments with facing translation) and Frede 1989. 93 While some have suggested that Chaeremon preceded Seneca in the role of Nero’s tutor, others have suggested they held roles concurrently; see Horst 1987: ix and 81. 94 On Musonius, see Lutz 1947, containing text, facing translation, and an extensive introduction. Note also Laurenti 1989. 95 See Tac. ann. 16.21–35. 96 Musonius’s diatribes are generally thought to be reports made by a pupil, Lucius; see Lutz 1947: 7. We have already noted that of the works of the bilingual Cornutus, the only surviving work is in Greek. Musonius’ pupil Epictetus wrote nothing but he lectured in Greek, and his lectures were recorded by Arrian. A little later the Emperor Marcus Aurelius chose to write his Meditations in Greek. We might also note the ethical treatise of Hierocles, in Greek, also dating from the Imperial period.

112

john sellars

exemplified by Cicero, Lucretius, and Seneca came to an end soon after Seneca’s death. Seneca stands as the only Stoic to write in Latin and, indeed, one of the few philosophers of any school in pagan antiquity who tried to do philosophy in Latin.

PART TWO

PHILOSOPHY Works

DE PROVIDENTIA1

R. Scott Smith

Date As is often the case with Seneca’s Dialogues, there is no conclusive evidence in De providentia by which we may securely date it. The notice of Tiberius at 4.4 (murmillonem sub Ti. Caesare […] audivi querentem) apparently provides a terminus post quem of ad 37, but that is all; supposed discoveries of allusions to later historical events are entirely unpersuasive. Seneca’s reference to his presence at one of Demetrius the Cynic’s lectures (3.3; cf. 5.5) makes it unlikely that he wrote the treatise in the later years of his exile on Corsica (41–48). Further attempts to narrow down the range or to champion a specific date have been overoptimistic and speculative (cautious remarks at Giancotti 1957: 244–309, Griffin 1976: 396, 400f.). In particular, critics have attempted to date the work by demonstrating a connection between the content of the essay and events in Seneca’s life. If, as has been commonly argued (recently Grilli 2000, adding stylistic arguments), Seneca wrote the De providentia as a consolation for his own troubles, it would belong to either the period of his exile or that of his retirement from Nero’s court (ad 62–65). Yet, the tone of the work, which is intensely protreptic rather than (self-)consolatory, does not support this thesis. Several indications, however, support a late date. In addition to De providentia, Seneca dedicates to Lucilius two other works, the Naturales quaestiones and the Epistulae morales, both of which date to the last years of Seneca’s life and likely to his retirement (ad62–65). Seneca’s remark at nat. 2.46 (at quare Iupiter aut ferienda transit aut innoxia ferit? In maiorem me quaestionem vocas, cui suus dies, suus locus dandus est) may be hinting at his designs to write a separate treatise on divine justice (contra, Grimal 1978a: 298–300). As Albertini noted (1923: 41), there is also a correspondence

1 Commentaries: Viansino 1968, Lanzarone 2008. Studies: Grimal 1950, Abel 1967: 97–123, Andreoni Fontecedro 1972, Dionigi 1994, Grilli 2000. For an overview of Stoic theology see Algra 2003; for Seneca and the problem of theodicy see Fischer 2008: 11–56.

116

r. scott smith

in form and content between the De providentia and three late letters (epist. 106, 108, 109), in which Seneca similarly deals with specific ethical problems as he prepares to write his all-encompassing magnum opus, the Moralis philosophiae libri. Content We refer to this work as De providentia, but this is only a modern convention. Lactantius’s copy (inst. 5.23.11) bore the title Quare bonis viris multa mala accidant, cum sit providentia. The index and subscription found in the Codex Ambrosianus (11th century) confirm this title. Despite being more cumbersome, the ancient title is more accurate, as Seneca’s primary concern is not to provide a general account of Providence but to treat the specific problem of theodicy: why, if the world is governed by a beneficent deity, do bad things happen to (good) people? Seneca himself acknowledges that a discussion of theodicy would best fit within the framework of a broader work on Providence itself, but because Lucilius does not so much doubt as complain about Providence (1.4), Seneca confines his discussion to this narrow topic. A general overview of the treatise runs as follows: Prooemium (ch. 1): definition of the topic (1.1); the orderliness of the universe is not accidental (1.2–4); gods are interested in the well-being of good men (1.5f.). Narratio (?) (ch. 2): good men are not harmed by adversa but consider them exercises in virtue (2.1–3), just as athletes seek adversity to make themselves stronger (2.3f.) and fathers toughen their sons with trials (2.5f.); the gods enjoy the spectacle of great men being tried by adversity, for instance Cato (2.7–12). Propositio and divisio (3.1): apparent evils are not actually evils, specifically: a) adversity benefits the individual; b) adversity benefits the collective; c) the virtuous cannot be miseri. Confirmationes (3.2–6.9): divided into three parts, corresponding to the divisio at 3.1:

de providentia

117

a) 3.2–4.16 adversity benefits the individual; b) ch. 5 adversity benefits the collective and is part of the divine plan; c) ch. 6 the virtuous cannot be miseri. The structure as given above essentially follows the interpretation in Grimal 1950, but not all critics are in agreement, nor do all share the view that Seneca’s composition is so well balanced and carefully composed (see Dionigi 1994, Abel 1967: 97–123, Albertini 1923: 103ff.). The disparity of opinion is owed, at least in part, to the great number of amplifications, analogies,exempla, and rhetorical devices that often obscure Seneca’s progression of thought. Topics Seneca’s essay treats a fundamental problem in Stoic theology, namely how to account for the existence of harmful forces that threaten the well-being of humans in a world governed by a supposedly beneficent deity. If the world is not only governed by, but even composed of a benevolent force, one concerned not only with the whole but also its parts, why is the world constructed in such a way that people, especially virtuous ones, suffer? After all, in everyday experience people encounter many things that one would consider harmful—illness, injury, hunger, and so on. One response to the problem might be to call on the concept of Stoic indifferentia (see entry of De constantia sapientis), whereby events and actions that are out of our control (or “not up to us”) have no impact on our virtuous state. But it is one thing to argue that illness, loss of a limb, or bereavement do not affect one’s virtue, quite another to explain why these dreadful forces exist in the first place if a benevolent Providence is at the helm. The Stoics acknowledged the existence of forces that were potentially harmful to humans. In an important fragment preserved for us by Aulus Gellius (7.1.1–13 = SVF 2.1169f., Long and Sedley 1987: 54Q), Chrysippus argued that if goods (bona) exist, evils (mala) too must exist, since nothing can exist without contraries. Later in the fragment Chrysippus specifies how mala come into being: they were created not directly by Nature, but as necessary “aftereffects” of the creation of goods (‹non› per naturam, sed per sequellas quasdam necessarias facta dicit, quod ipse appellat κατὰ παρακολούθησιν). This argument relieves the divine from purposeful generation of evils while accounting for their existence (on Stoic evil in general, see Long 1968; on moral evil in Seneca, see Hine 1995). Seneca too implies that adversa (he is keen to avoid the use of mala) are part of the natural world order (5.9, 6.6; cf. Cleanthes’s Hymn to Zeus).

118

r. scott smith

Seneca approaches the problem by arguing that hardships are not cosmic evils, but rather exercises that test, harden, and prove virtue in the virtuous man (omnia adversa exercitationes putat: 2.2). He goes even further: virtue requires an arena and an adversary in and by which to prove its strength (marcet sine adversario virtus: 2.4). In this view Seneca is joined by Epictetus, who (diatr. 1.6.32–36) argues that there would never have been a Hercules without such challenges; his valor would never have been known unless there were “such circumstances […] to make trial of (διέσεισαν) and to exercise (ἐγύµνασαν) him.” According to Epictetus, Zeus (as god/nature) provides us with the means (παρασκευή) and resources (ἀφορµαί) to deal with adversity. This echoes the words of the Stoic deus at prov. 6.6: Quia non poteram vos istis subducere, animos vestros adversus omnia armavi (“Because I could not save you from them [scil. adversis], I have armed your minds against all of them”). The good man’s worth is made evident only by hardship; just as fire authenticates gold, so too does adversity authenticate brave men (5.9). Consistently throughout the work Seneca seeks to emphasize the masculine aspect of the gods’ work, weaving it into the traditional Roman notion of virtus. The gods take a “fatherly” disposition toward good men (1.5 f., 2.5f.), hardening them with heavy labor and sweat; trainers test athletes with hardship (2.3f., 4.2–4); teachers challenge their students (4.11); gladiators (2.8, 3.4) and soldiers (4.4, 4.7, 4.8, 5.3), whose worth is only proven under adversity, face similar tests. For Stoicism as a masculine philosophy, see also the beginning of De constantia sapientis. Language and Style De providentia is among the most impassioned and rhetorically vibrant of Seneca’s dialogues. The treatise is framed as a defense of the gods (causam deorum agam: 1.1), a conceit emphasized at the beginning by language borrowed from the courtroom (particula, contradictio, lis), repeated later in Seneca’s explicit use of oratio (3.1), and culminating in the prosopopoeia of the Stoic god himself, which brings the work to a close (6.3–9). Fortuna, too, is personified twice and gives a speech in defense of the Stoic god (3.3f.); on two other occasions we encounter major historical figures speaking (Cato 2.10; Rutilius 3.7). In a work aimed at justifying the works of the Stoic god one is not surprised at the aggressive tone: note the vigorous use of sudden plural imperatives at 4.6 (nolite, obsecro vos, expavescere) and 4.9 (fugite delicias, fugite […]), as well as the incredulity expressed at questioning the god’s good intentions: quid miraris/quid mirum (2.7, 3.2, 4.12, 4.16, 6.2); non

de providentia

119

vides (2.5); numquid […] credis (4.11); quidni libenter (2.12). Seneca deploys rhetorical devices with even greater frequency than usual: polyptoton and puns: 1.5 (dis adversus optimos optimis and ut umquam bona bonis noceant); 2.10 (alter alterius manu caesi); others at 3.10, 4.3, 4.11, 4.13, 6.5; anaphora at 2.5 (numquam 3×), 2.9 (ecce […] dignum 2 ×), 2.11 (dum 3×), 3.5–7 (infelix est […] felicior esset/felix est 3×), 3.9 (documentum 2 ×), 4.5 (unde […] (si) 3 ×), 4.9 (fugite 2 ×), 5.5 (vultis 3 ×), 6.2 (quidni 3 ×), 6.6 (contemnite 4 ×), and 6.9 (sive 4×). Transmission Has the end of De providentia been lost in the course of transmission? Critics who view the treatise as incomplete bring to bear a number of arguments: 1) it is the shortest of Seneca’s Dialogi (save the mutilated De otio); 2) it ends abruptly with the end of the deus’s long speech and lacks a peroratio; 3) not all of the topics announced in the divisio (3.1) are treated at length; and 4) the material drawn from De providentia in Lactantius (reference supra) is not found in the transmitted text. All of these can be accounted for: 1) the treatise is indeed the shortest, but only slightly shorter thanDe constantia sapientis; 2) in an essay so rhetorically charged and with features of diatribe, ending the treatise with the speech of the deus is rhetorically forceful—the last word, so to speak; 3) in Grimal’s reanalysis of the structure the apparent lack of development has disappeared (nor is consistency always to be expected from Seneca); and 4) the material from Lactantius is not a lost fragment but a summary of the contents, perhaps from memory. Sources Seneca cites Demetrius the Cynic twice (3.3, 5.5). The latter runs several lines long; perhaps at 3.3 more than one line (thus Reynolds 1977) should be attributed to Demetrius (note enim in both instances). He also employs a long passage from Ovid’s Metamorphoses on the myth of Phaethon (5.10f.). Perhaps deriving from rhetorical handbooks/schools are the laudatio Catonis (2.9–12), the list of suicides corresponding to the elements (6.9), and the long list of historical exempla of great men facing adversity (3.4–14: Mucius, Fabricius, Rutilius, Regulus, Socrates, Cato). The same list is found at epist. 98.12f., truncated lists at dial. 6 (= cons. Marc.).22.3; dial. 9 (= tranq.).16.1; and epist. 67.7, 71.17 (see Mayer 1991: 153 f.). As for philosophical sources, it is impossible to determine which, if any, Seneca employed in composing

120

r. scott smith

De providentia. Two Stoics, Chrysippus and Panaetius, each wrote a Περὶ προνοίας, but of the former we have only fragments and of the latter only the title. Nothing in the surviving fragments of Chrysippus’s work suggests that Seneca consulted it directly, but the most extensive fragment (cited supra) demonstrates that Chrysippus was concerned with the problem of theodicy. Numerous writers treated the topic of Providence: see Epikt. diatr. 1.6, 3.17; Aelian, De providentia (very fragmentary); Philo, De providentia I– II (fragmentary); the six speeches Περὶ εἱµαρµένης τε καὶ προνοίας by Dio Chrysostom; and the ten speeches Περὶ προνοίας by Theodoretus; cf. the collection of fragments at SVF 2.1168–1186.

DE CONSTANTIA SAPIENTIS1

R. Scott Smith

Date The date of De constantia sapientis can only be established within a wide range. The mention of Caligula’s death (18.1) places it certainly after ad 41; it is also unlikely that Seneca would relate the embarrassing story of Valerius Asiaticus before his death in ad 47 (18.2). The treatise must have been written before the death of the addressee, Annaeus Serenus, reported in epist. 63.14 (likely dated to ad63–64). Beyond this, however, little can be said. De constantia sapientis likely predates De tranquillitate animi since Serenus is portrayed as not yet a Stoic in the former (3.2) but committed to Stoic principles in the latter (1.10). But since that treatise also eludes dating (perhaps ad60?), only a relative chronology is possible. If De constantia sapientis reflects Serenus’s (or Seneca’s) concerns as part of Nero’s court, a date of ad54–59 is perhaps called for (Minissale 1977: 9–13, Viansino 1968: 10f.; Albertini 1923: 31), but such a biographical reading of the text is impossible to corroborate without clear historical allusions, of which there are none. Content Although we refer to this essay as De constantia sapientis (“On the Steadfastness of the Wise Man”) there is no evidence that it was called this in antiquity. The index in the Codex Ambrosianus, the incipit, and the explicit all transmit the title Nec iniuriam nec contumeliam accipere sapientem (“That a Wise Man Receives neither Injury nor Insult”), which is itself drawn from the thesis advanced at 2.1. The term constantia does not occur in the essay, so here, as with De providentia, we likely have an attempt to shorten a cumbersome title (see Klei 1950: 2–5).

1 Commentaries: Klei 1950 (Dutch), Viansino 1968, Minissale 1977. Studies: Grimal 1949b, Ganss 1952 (rare): 31–37, Giancotti 1957: 178–192, Abel 1967: 124–146 (cf. 159), André 1989: 1739–1741, 1756–1764.

122

r. scott smith

An overview of the treatise: ch. 1 (exordium2): Seneca argues that Stoicism, though an aggressive, masculine philosophy, is not as harsh as it appears (captatio benevolentiae). ch. 2 (narratio): Seneca recalls an earlier conversation with Serenus about the mistreatment of Cato the Younger. Since Cato was a Wise Man, he could suffer neither injury nor insult. ch. 3f. (propositio/quaestio): Serenus objects to this Stoic paradox (3.1 f.); Seneca establishes the nature of the question, emphasizing the impervious nature of the sapiens through a series of analogies. ch. 5.1f. (divisio) separates contumelia from iniuria, establishing the general framework for the rest of the treatise. ch. 5.3–ch. 9 (argumentatio I): exposition of proofs that the sapiens is impervious to iniuria; exemplum of the Megarian Stilpo (5.6–6.8) ch. 10–18 (argumentatio II): discussion of contumelia, though both iniuria and contumelia are treated together at 12.3, 13.5, and ch. 15f.; exempla drawn from the senate and imperial court (ch. 17f.). ch. 19 (peroratio) Seneca extols the sapiens while offering advice for the rest of us imperfecti. Topics At the center of the work stands the Stoic paradox “the Wise Man is not subject to harm,” one of many paradoxes involving the Stoic sapiens. The figure of the sapiens, that rare—critics would say imaginary—human, perfect in every way, differentiated from god only in his mortality, and impervious to the blows of Fortune, was open to criticism for obvious reasons. How, for example, could a man of flesh and blood be impervious to injury, pain, bereavement, even death? Such is Serenus’s objection, but his criticism of

2 Rhetorical divisions following Grimal 1949b; for composition see also Albertini 1923: 75 f., 265; André 1989: 1756 f.

de constantia sapientis

123

the seemingly imaginary figure of the Stoic sapiens no doubt represents a widespread objection voiced by non-Stoics (cf. 3.3, 4.1, 4.3, 7.1, 7.3, 12.3, 14.2f., 15.1). Seneca’s treatise attempts to vindicate this paradox while making the Wise Man both relevant and meaningful to his audience. Seneca divides his treatment into two parts, separating contumelia (“slight”) from the more serious iniuria (“injury”), a division he may have found in a source (see below). The distinction, more of degree than of kind, is somewhat artificial, and Seneca stretches to establish a fixed line between the two. His quasi-legal distinction—iniuria is subject to redress under the law, whereas contumelia is not (10.1)—is found nowhere else in Roman legal texts (see Viansino 1968: 12f., Minissale 1977: 15f.), and his philosophical definition, that the sapiens feels, yet overcomes iniuria, yet does not even feel contumelia, is contrived. Yet his insistence on dividing the two topics relates directly to his two rhetorical aims: first, to prove the truth of the paradox; second, to attack Roman softness while giving us the means to keep our sanity in an increasingly hostile world. The discussion of iniuria involves several proofs in syllogistic form establishing the logical incompatibility of the sapiens and iniuria, that is of good and evil, of greatness and its opposite. Crucial here is the Cynic-Stoic concept of αὐτάρκεια (“self-sufficiency”). The sapiens, free from his reliance on externals (19.2), is concerned solely with what he can control, virtue, which in turn is sufficient for happiness. All else—health, status, wealth, even life itself—is relegated to second-order importance as indifferentia, or “ethically neutral conditions,” which may or may not be preferable (e.g., good or bad health), but which do not affect one’s ethical state precisely because they depend on Fortune. Herein lies the key to the paradox: the right-thinking sapiens understands that what is subject to Fortune does not in fact belong to him, and thus does not value it as highly as do others. Since iniuria affects externals, not virtue, they are merely inconveniences (incommoda: 16.2), ones which the Wise Man transforms into tests of his own virtue (9.3; cf. De providentia). So far iniuria, but what of contumelia? One might object that once Seneca has proven that the sapiens cannot suffer iniuria, it is unnecessary from a purely logical point of view to prove that he cannot suffer the less serious contumelia. Indeed, as noted above, whereas the sapiens feels the former (10.4), he does not feel the latter. Contumeliae are perceived slights due to faulty reasoning (vitio interpretantis: 10.3), physical contractions of weak-minded men’s souls (10.2f.), which the right-thinking sapiens cannot experience. Even if “greater” men, those supposedly endowed with more power and status, insult the Wise Man, he remains unaffected because of his magnitudo animi (= µεγαλοψυχία), “the most glorious of all virtues” (11.1;

124

r. scott smith

cf. 15.2). Certain of his own greatness, his own superiority (Stoicism turns the conventional notions of greatness and power on their heads), the Wise Man treats his would-be insulters as a father would his misbehaving children (11.2) or a doctor his fevered patients (ch. 13). Seneca does not so much prove that the Wise Man cannot suffer contumelia as illustrate the mental disposition that allows him to live unruffled by the everyday irritations of life caused by the diseased humanity in Rome. Seneca has thus subtly broadened the scope of the paradox and changed the lens of his rhetoric, focusing less on the sapiens than on the pettiness of a certain segment of Roman culture: the delicati (5.1, 10.2), those soft, womanly (muliebria: 10.3, 19.2; cf. 1.1) types who in the absence of real trials take offence at the most trivial slight. At the same time Seneca has abandoned his esoteric exposition of the Wise Man in favor of demonstrating how one can remain sane in a society of insane people—and how Stoicism can help. Magnitudo animi is possible not just for the Wise, but for all philosophically minded people (16.3). Even if we cannot achieve perfection—unlikely given that only one sapiens appears every several generations (7.1)—we can train ourselves through philosophy to care nothing about the petty cares of our peers and, unlike the delicati, to be strong and tough-minded.3 Language and Style De constantia sapientis, an outgrowth of an earlier conversation between Seneca and Serenus (1.3–2.3, 7.1), is meant to be read as a dramatized snapshot of a philosophical education in progress. Seneca’s close relationship with his addressee is evidenced here, as in De tranquillitate animi, by a vivid portrayal in the opening chapters: Serenus is painted as a skeptical, passionate young man wrestling with the principles of Stoic philosophy. As Abel (1967: 124–128) has amply demonstrated, Serenus and this earlier conversation are central to the conception and composition of the work. The work is organized along clear lines, with well marked transitions, but one observes a substantial difference in both language and style in the expositions of iniuria and contumelia. The treatment of iniuria, among the most systematic expositions in Seneca, is composed mainly of esoteric syllogisms (5.3f., 7.2, 8.1f., 9.3; cf. 11.2), for which Seneca elsewhere shows

3 Coping with contumeliae/iniuriae is a vital part of the therapy of anger found in De ira III, which has multiple points of contact with our treatise (see esp. dial. 5.25–28, 37).

de constantia sapientis

125

disdain (epist. 85.1, 108.12). Unsurprisingly, then, he soon turns to a lengthy personification of Stilpo as an example of the self-sufficient sapiens (5.6– 6.8), betraying his reluctance to rely solely on logical proofs to establish his point. By contrast, the section devoted to contumelia exhibits more rhetorical vigor, featuring elements of diatribe and drawing from everyday Roman life: children/parents (11.2), sarcastic slaves (11.3), the meaningless games children and adults play (ch. 12), doctors/patients (13.1f.), sordid slave sellers (13.4), attacks on women (ch. 14.1), and difficult household slaves (14.1 f.). When meditating on the sapiens Seneca employs striking images; he is beyond the reach of his attackers,4 unshakeable and invulnerable,5 and all but divine (3.3, 4.2, 6.8, 8.2 f., 14.4). Seneca echoes Lucretius’s description of the detached happiness of the Wise Man (14.1 = Lucr. 2.1–4), and also employs a quote from Epicurus to demonstrate the essential agreement between the two philosophies. Yet, the conceit that Stoic philosophy is a tough, masculine philosophy (as opposed to Epicureanism) is sustained throughout (virilem […] viam: 1.1, duritiam […] Stoicam: 15.4) and is emphasized by images from the military (3.5, 4.2f., 5.6–6.8, 19.3), athletic and gladiatorial contests (9.5, 16.2), and nature itself (3.4f.). For more on language and style, see Minissale 1977: 18–23. Sources The paradox of the Wise Man’s imperviousness was commonly asserted (SVF 1.216, 3.567–581, Plut. mor. 1044B), but no full treatment other than Seneca’s has survived. It is not among the six paradoxes treated in Cicero’s Paradoxa Stoicorum; Hecato’s “On Paradoxes” is lost. An important but short passage preserved in Stobaeus (Ecl. 2.7 p. 110 W. = SVF 3.578) seems to differentiate ὕβρις (“insult”) from a general category of ἀδικία (“maltreatment”) and may prefigure Seneca’s separation of contumelia from iniuria. Although in the Stobaeus passage ὕβρις does not conform to contumelia in all respects (Grimal 1949b: 250f.), there are similarities. Both emphasize the guilt of those wishing to do harm despite the Wise Man’s imperviousness to that harm (dial. 2

4 Note words in e(x)- or a(b)-: in editum verticem […] extra omnem […] emineat: 1.1; intervallo […] abductus (4.1); editissimas arces (6.4); excelsa (6.8, cf. 1.2); excedentia (7.1). 5 Note adjectives in in-: invictus (2.1, 5.7, 19.4); invulnerabile (3.3); inexpugnabilis (3.5, 6.8); [virtus] libera est, inviolabilis, inmota, inconcussa, sic contra casus indurata (5.4); indemnem (5.7); integrum incolumemque (6.5); integra inlibataque (6.7); inexsuperabilibus munimentis praecincta (6.8); intrepida (8.2).

126

r. scott smith

[= const.] 4.3, 7.3–6, 9.4; cf. benef. 2.35) and stress self-sufficiency and virtue as the sole factors in the Wise Man’s well-being (const. 5.4f.). The syllogisms in the discussion of iniuria point to the Old Stoa, but the priority of magnitudo animi in enduring contumelia indicates the influence of the so-called Middle Stoa, in particular Panaetius (André 1989: 1740, Abel 1967: 128f., Knoche 1935; cf. Cic. off. 1.61–92 with Dyck 1996 ad loc.). Seneca relies heavily on historical exempla, some of which he may owe to handbooks: Cato (1.3; cf. 7.1, 14.3; epist. 14.12–13), Stilpo/Demetrius Poliorcetes (5.6–6.8), Vatinius/Cicero (17.3), Socrates (18.6), and Antisthenes (18.6); cf. Chrysippus at 17.1. Others, such as the story of Corbulo calling Cornelius Fidus, Ovid’s son-in-law, a “plucked ostrich” in the senate, he probably experienced first hand (vidimus: 17.1); so too the emperor Gaius’s conduct toward Valerius Asiaticus, Cassius Chaerea, and Herennius Macer (18.1–5).

DE IRA

Maria Monteleone

Date The treatise De ira is composed of three books, viz. Dialogi nos. III, IV, and V. The only pinpoint for dating the work is a terminus post quem corresponding to Caligula’s death on January 24, ad 41, as suggested by repeated allusions to this Emperor’s violent temper and frequent outbursts (1.20.4; 1.20.5; 1.20.8 f.; 2.17.1; 2.20.1; 2.33.3–6; 2.36.3; 3.18.3f.; 3.19.1–5; 3.21.5). 1 Other hints that might determine the terminus ante quem do not provide such clear evidence.2 Seneca’s brother, to whom the treatise is addressed, is still called by his native name, Novatus, instead of his adoptive one, Gallio, which he definitely bore by ad52–53. Still, the exact year of the adoption is not known to us, and neither is the date of composition of the De Vita Beata—the first of Seneca’s works in which his brother is addressed as Gallio. Also, Seneca’s hint at his wife’s presence beside him (3.36.3) is no help, as we know nothing about his conjugal affairs. Nor can Claudius’s edict, in which he promises to control his temper,3 be accepted as evidence: we do not know when it was issued, and it cannot be ruled out that it was precisely in response to Seneca’s De ira that Claudius publicly admitted his being affected by such passion, and declared himself willing to mitigate it. Content Seneca’s De ira begins with a declaration of its author’s alleged aims: “You have asked me, Novatus, to write on how anger can be mitigated.” An introductory chapter follows, providing a phenomenological description of anger: among passions (adfectus), it is both the fiercest and the least human-like, so that

Notably 2.20.1, 2.17.1, 2.36.3 find confirmation in Suet. Cal. 50.8, 53.2, and 50.3, respectively. For an overview, see Herrmann 1937: 95f.; Coccia 1958: 34f.; Giancotti 1957: 93–150; Cupaiuolo 1975: 28–52; and Grimal 1978a: 270–276. 3 Cf. Suet. Claud. 38.1. 2

128

maria monteleone

many philosophers view it as some short-lasting insanity (brevis insania), beyond self-control and inaccessible to reason’s advice (rationi praeclusa). Even the appearance (habitus) of people consumed by anger reveals their insanity: just as animals make their looks more ferocious when about to attack, so people who are becoming irate (irascentes) show clear symptoms of insanity (furentium certa indicia) all over their body, which appears to be deformed (deforme) by passion. Among different adfectus, wrath is the most dangerous for mankind, bringing ruin upon whole communities as well as individuals: it is “craving to take revenge of an offence” (cupiditas ulciscendae iniuriae), “to punish the one whom we think has done us wrong” (cupiditas puniendi eius a quo te inique putes laesum); it can turn relatives and fellow citizens into enemies, and kings into tyrants. A distinction is made between ira, being a brevis insania, and iracundia, which is a constant inclination toward ira. The problem is then tackled as to whether anger has natural origins, and whether it can be of some use (1.5–21). Confronting the Aristotelian and peripatetic view that properly restrained ira may even be useful, Seneca holds it to be a passion against nature: in his view ratio and ira are, respectively, the change of the soul for the better and the worse, so that the former never needs the latter and can succeed in everything by itself. Book Two (especially 2.1–4) examines the process through which the adfectus of ira arises: the impression of having been wronged (opinio iniuriae) determines a first non-voluntary motion (primus motus non voluntarius) in the animus, a surge of indignation; next, the animus recognizes by a rational and willful act (iudicium) that such an impression is justified and approves (adsensus) the necessity of repaying the offence (punire, ulcisci, dolorem reddere, vindicare); ira then bursts out with an attack (impetus) aiming to take revenge (ultio, poena). The treatise next (esp. 2.18–36) proposes a prophylactic method to avoid falling prey to anger (de tuenda valetudine = ne incidamus in iram): young people should be shielded by proper education, while adults must get into the habit (consuetudo) of not believing in opinio iniuriae and not bestowing adsensus animi on it—that is, they must convey passion’s primus motus into a motus rationi parens by means of a iudicium aequum that should assess whether an offence has really taken place. If so, the iudicium aequum should impose a correct evaluation of those responsible (facientes iniuriarum), never forgetting that “no one of us is without blame” (neminem nostrum esse sine culpa) and that revenge is pointless, because poena is of some use only if it can prevent a scelus from being accomplished, and not when it comes later. Book Three is dedicated to what may properly be called therapy for anger (de restituenda valetudine = iram excidere animis aut refrenare), once it has

de ira

129

made its way inside a man’s soul. First of all, attacks must be prevented by reasonable behavior and the company of amicissimi, in order to preserve the mental equilibrium (tranquillitas animi) that is needed to restrain the insurgencies of passion. If preventive therapy does not work, anger can be restrained in many ways by reasoning and by recourse to aequum iudicium: opinio iniuriae can at first be confronted with a different view, when the causes of iniuria appear to be insignificant (levia); if the crisis has reached a further stage, one must take his time (se differre) and fight against himself (pugnare secum) to prevent anger from bursting out (exilire). Most of all, the thought of mankind’s common destiny, the awareness that death will soon make the same of us all, is the tool that will eradicate wrath from one’s soul, bringing him to neglect iniuriae and not to delight in other people’s suffering. Topics Among the most controversial issues of Seneca’s De ira has always been the problem concerning its compositional structure. Most scholars have agreed—albeit on different grounds—on the work’s lack of unity, stressing the substantial autonomy of Book Three, deemed to be either a much later addition to Books One and Two or a separate and independent work (the latter view was backed by Pfennig 1887, Rabbow 1914, Albertini 1923, NikolovaBourova 1975, and Castiglioni 1924, while Mueller 1912, Bourgery 1922a, Coccia 1958, Abel 1967, Boal 1972, Grimal 1978a: 410–424, Fillion-Lahille 1984: 283–290, and Ramondetti 1996 argue for unity despite the uneven composition).4 It cannot be denied, however, that in all three books Seneca devotes special attention to some definite themes, which makes the whole theoretical construction more solid and consistent. An outstanding place is taken by physiognomics (a very popular theme among Stoics, see Cupaiuolo 1975: 91, Evans 1950, and Ramondetti 1996: 56f., 67f.), as becomes evident in the three different descriptions of the angry man (1.1.3 f.; 2.35.2–5; 3.3–4.3): deformitas animi, entailed by anger as a worsening of one’s soul, is reflected in deformitas corporis. Anger is described in terms of decay, as some sort of deviation from good qualities, moral (ira = deformation of animus, collapse and loss of ratio) as well as bodily (ira = ugly, animal appearance; disease). 5

4 For a critical review of different positions, see Cupaiuolo 1975: 67–87 and (more recently) Ramondetti 1996: 9 f. 5 Comparisons of the angry man with the enraged beast (in the extreme case of a furens tyrant, Seneca actually uses the word feritas), with sick people (first and foremost insane

130

maria monteleone

Such deviation from the natural condition of man—i.e., from the norm— is viewed by Seneca in a basically “interactional” perspective: ira does not only affect the angry person, it is the kind of passion that displays itself and its worst effects on the level of human relationships. Seneca actually describes anger, the desire of repaying an offence, as a perverted model of the mechanism of reciprocity between men:6 anytime ira intervenes in the relationship between two subjects, instead of the proper exchange of mutual services (officia, beneficia, mutuus amor, fides, etc.) another is triggered, that of offence and punishment. An immediate consequence is the reversal of the normal terms of relationships, so that anger “prevents anybody falling prey to it from remembering what his own duties are: you infuse it in a father, he becomes a foe; in a son, he becomes a parricide; in a mother, she becomes a cruel stepmother; in a fellow citizen, he becomes an enemy; in a king, he becomes a tyrant.”7 Indeed, it is the tyrantwhom Seneca most frequently portrays as the representative par excellence of angry people, the one displaying most evidently what dangers may be engendered by this passion to the human race. The emphasis on the figure of the king who, blinded by anger, cannot make the punishments he inflicts on his subjects fit the gravity of their crimes and is therefore unable to administer justice in a rational way, has led some scholars (Cupaiuolo 1975: 7–17, Fillion-Lahille 1984: 278–282, and Viansino 1992: 119f.) to think that the work may actually be addressed to Claudius, or to rulers in general: Seneca warns them not to yield to passion as did Caligula, who turned himself into a bloodthirsty tyrant of the “oriental” fashion and an example of how a Roman emperor should not behave.8 Seneca’s representation of the tyrannus, as well as his remarks on the subversion of the social order that is brought about by anger, have triggered a profitable reflection on the relationship linking our treatise to Seneca’s tragedies, above all Thyestes (see especially Staley 1975 and 1981–1983; also Abel 1985a: 765):9

people), or with those categories (women, children, the elderly) who possess a lesser share of ratio compared to adult men, all obey the same logic. 6 For a discussion on the mechanism of reciprocity and mutual damage in general, see Courtois 1984; on Seneca and De ira especially, see Guastella 2001: 9–30. 7 Quemcumque obtinuerit, nullius eum meminisse officii sinit: da eam patri, inimicus est; da filio, parricida est; da matri, noverca est; da civi, hostis est; da regi, tyrannus est (1.3. fr. 3a). 8 It should be noted (Nikolova-Bourova 1975, Fillion-Lahille 1984: 274–278, and Viansino 1992: 120) that De ira already contains the theoretical basis of Seneca’s later work De clementia. 9 “Il personaggio di Atreo rappresenta una così metodica applicazione del meccanismo illustrato da Seneca nel De ira, che si sarebbe tentati di leggere il Tieste col commento dello stesso trattato” (Guastella 2001: 31).

de ira

131

the pattern of anger sketched in the former seems to provide the background on which the latter’s intrigues are constructed. Tragic characters actually seek revenge obeying the same logic that we find outlined in De ira: the uncontrollable urge for retaliation, which is engendered by suffering aniniuria, causes the affected person to cross the threshold of rational behavior and to attempt revenge with a totally disproportionateinuria—that aliquid maius that is neither poena nor castigatio, but definitely scelus—in order to appease the dolor inflicted on him by the suffered iniuria (Guastella 2001: 15 f.).10 Language and Style Among Seneca’s works, De ira is perhaps the one in which the author’s rhetorical training and declamatory practice are most evidently displayed, affecting the rhetorical pattern that underlies the whole treatment of the subject (propositio, argumentatio, conclusio): style, which is expressive and in constant pursuit of effect, and language, which is rich in figures of speech and variationes. Seneca also draws on the tradition of cynic-stoic diatribe, which becomes particularly evident in his neglect of the rules of decorum, in his brilliant style, which aims to impress the reader, and in his unusual language, which straddles tradition and renewal.11 Sources When Seneca wrote De ira, he probably had at his disposal, in addition to treatises on the passions in general,12 works devoted specifically to wrath by Theophrastus, Antipater from Tarsus, Posidonius from Apamea (generally deemed to be Seneca’s main source), Philodemus from Gadara, and Sotion from Alexandria. At least, he knew of such works, although in some cases only indirectly.13

10 The application of this mechanism in Seneca’s tragedies is discussed at length in Guastella 2001: 9–154. 11 On the language and style of De ira, see Coccia 1958, Cupaiuolo 1975: 118–161. 12 Works On passions were written by Stoics such as Zeno, Chrysippus, Ariston from Chios, Herillus, Hecaton, Posidonius from Apamea, and Sotion; by Xenocrates, an Academic; and among Peripatetics, besides Aristotle, by Theophrastus and Andronicus from Rhodes. 13 For discussion in detail on the sources of De ira see Fillion-Lahille 1984, Cupaiuolo 1975: 88–106, and Viansino 1992: 122–137.

132

maria monteleone

Seneca explicitly mentions Aristotle and Theophrastus,14 without precise reference to their work, to disclaim (especially in 1.5–21) some statements of theirs, first of all that anger is a natural phenomenon and might prove very useful if kept under reason’s control.15 It is precisely in the diatribe against Peripatetics that De ira betrays its author’s loyalty to the Stoic tradition, which considered reason to be fully enabled to act only if free from passion, and which condemned anger as being “against nature,” therefore denying that it might be of any use in any occasion. Book One appears to reflect Chrysippus’s “orthodox” Stoicism, acknowledging only the rational part of the human soul and regarding the passions as some sort of insanity (allowed by perverted reason), which is impossible to restrain and which must be eradicated. Books Two and Three, on the other hand, are affected by Posidonius’s more realistic views and less strict rationalism, reckoning with the existence of instinctual reactions that reason cannot control, although they are not properly called “anger”, and devoting greater attention both to an investigation of the causes and to therapy, which is centered on education, patience, and practice. As far as prophylaxis is concerned, there can be little doubt that Seneca was influenced by Sextius, whom he explicitly mentions in connection with exercises to be practiced every day—inner self-examination and looking in the mirror to see how passion disfigures one’s outer looks (2.36.1 f., 3.36.1). Seneca was also most likely well aware of the treatise written by Philodemus from Gadara, the Epicurean philosopher, with whom he shares both the perception of anger as some ruinous evil for mankind—a sort of insanity that must be cured—and his denial of the usefulness of passion; their views diverge concerning the nature of anger, deemed by Philodemus (and by Epicureans in general) to be a natural and ineluctable phenomenon, which ought to be accepted and kept under control, rather than altogether suppressed. The consonance between the attacks on Peripatetics in Book Four of Cicero’s Tusculanae disputationes and those in Seneca’s De ira is better explained by supposing that both drew on the same sources (viz. Chrysippus and Posidonius), rather than that the latter depended directly on the former.

Aristotle in 1.9.2, 1.17.1, 3.3.1; Theophrastus in1.12.3, 1.19.3. However, a comparison is usually drawn with remarks made by Aristotle on this subject in De anima, Ethica Eudemia, and, most of all, in Ethica Nicomachea and the second book of Rhetorica. For exact references, see Cupaiuolo 1975: 94–96, Fillion-Lahille 1984: 203–210, and Viansino 1992: 125–127. 15

de ira

133

Transmission In Ambrosianus C no. 90, the eleventh-century codex preserving the corpus of Seneca’s Dialogi, the beginning of De ira is missing up to tamquam (1.2.4).16 Section 1.1–2.3 was restored by a twelfth-century corrector (a) on the only page he wrote (14r, in Beneventan writing). A lacuna still exists (between capitis damna, 1.2.3, and tamquam, 1.2.4, where the Ambrosianus begins), which is usually patched with an epitome written by Martin from Braga in the sixth century (viz. 1.3.3 from ira omnia to tyrannus est)17 and a quotation from Seneca in Lactantius, De ira Dei, 17 (viz. 1.3.3b, from ira est to nocere voluit). Reception Seneca’s De ira enjoyed considerable success among Christians in Late Antiquity;18 notably, it seems to have provided an important theoretical basis for both Arnobius’s Adversus Nationes and Lactantius’s De ira Dei. The former author seems to have drawn on Seneca’s text in many passages of his work, for example, where he describes anger, its pernicious effects on the affected person, and the ruinous consequences of its displays, or when he depicts divinity as some being of higher nature, remote from passions and constant in virtue—in the mold of Seneca’s conception of the Stoic wise man.19 Lactantius reutilizes De ira in an even more conspicuous way, deliberately and systematically drawing from Seneca’s treatise the building material for his De ira Dei, and adjusting it to suit his own needs. On the one hand, he acknowledges Seneca’s merits as a moralist, and as being particularly close to the Christian mentality; on the other, he sharply criticizes him every time he appears to stray from this mentality.20 Nothing is heard about the text from the sixth century—when Martin of Braga composed his own De ira by summarizing the Senecan one—until the eleventh, when the Codex Ambrosianus turned up in the abbey of Montecassino. Some excerpts from

16 For relevant information about the manuscript tradition of De ira, see Reynolds 1968: 357, 368 f., Idem 1977: xii, 41, Idem 1983: 367, and Ramondetti 1999: 75f. 17 On Martin’s intervention, see Bickel 1905a: 535, 541 f. and Barlow 1937: 29–31. 18 See Traina 1987: 46–49, 195–198 for an overview of studies on Seneca’s Fortleben, and notably 171–192 for his influence on Christian writers. 19 For an overview of similarities (including those in form and style) between these two texts, see Mastandrea 1988: 12–33. 20 For a strict comparison between the two works, see Lo Cicero 1991: 1242–1261.

134

maria monteleone

the first two books of De ira can be found in a collection of sententiae drawn from the Dialogi, which circulated in France and England in the twelfth century under the title Proverbia Senecae per ordinem alphabeti disposita.21 In the fourteenth century, De ira was the only one among the Dialogi, along with De tranquillitate animi, to appear in Thomas de Hibernia’s Manipulus florum.22

21 For discussion, see Reynolds 1968: 360, Idem 1983: 367f.; Brugnoli 2000a: 230f.; and Munk-Olsen 1987: 163, 213. 22 Munk-Olsen 2000: 171 f.

CONSOLATIO AD MARCIAM

Jochen Sauer

Dates The Consolatio ad Marciam is addressed to the daughter of the senator and historian Cremutius Cordus; although three years have passed she is still in deep mourning over the loss of her son Metilius. The date of the dialogue is much debated. The only traceable terminus post quem is Marcia’s renewed publication of her father’s works (1.3). Most scholars follow Suetonius (Cal. 16.1), who says that this event took place during the reign of Caligula, i.e., not before March ad 37 (Lana 1955: 88 f., Giancotti 1957: 72 f., Abel 1958: 610 and 1985: 705f.).1 For the terminus ante quem most scholars are in favor of the beginning of Seneca’s exile in ad41. In this case the Ad Marciam could be the earliest of the preserved prose works of Seneca. Dating the consolatory address to the period of the exile (ad 41–49) leads to problems with 16.2 (in qua istud urbe, di boni, loquimur?) and with the description of the exile in 17.5 (Abel 1985a: 705, Bellemore 1992: 219 f.); dating it to the period after the exile leads to problems concerning Marcia’s age (2.2f.; Abel 1985a: 705, Griffin 1976: 397, Fabbri 1977–1978: 316f.). In view of the liberal tone of the piece the first months of Caligula’s reign would appear to be a most probable date for the composition of the work (Abel 1967: 159f. and 1985: 705f., opposed only by Griffin 1976: 397). Sources Our knowledge of Seneca’s predecessors in the genre of consolation literature is not sufficient to allow us to identify with certainty definite models. Plausible conjectures have been made suggesting Krantor’s Consolatio, the 1 Only Bellemore 1992, casting doubt on Suetonius’s reliability, places the renewed publication in the late Principate of Tiberius (ad34–37). In her argumentation she refers to the generous praise of Tiberius and to the fact that neither Caligula nor Claudius is explicitly mentioned. The problem with this date is that we would have to disregard our only direct testimonium (Suet. Cal. 16.1).

136

jochen sauer

prototype of the genre (Pohlenz 1906: 336, Favez 1928: xxvi, Stowell 1999: 15–20), and also Cicero’s Consolatio. A strong influence of the latter is likely, especially in the choice of examples and in the use of arguments from different philosophical schools (Helm 1939: 130 f., Abel 1967: 15f.). The recognition of possible lines of tradition is made difficult by the fact that visible parallels in thought often turn out to be commonplaces or topoi that are specific to the genre of consolation (cf. Gieseke 1891, Claassen 1999: 19– 26). The fact that many thoughts and arguments can be traced back to the repertoires of different Hellenistic schools of philosophy confirms the strong influence of the tradition of the genre where elements from rhetoric and popular philosophy abound (Grollios 1956, Kassel 1958). Regarding the choice of philosophical statements Seneca seems to show “a readiness to take help where it is offered” (Grollios 1956: 63f.). The background, however, to his weltanschauung is undoubtedly the Stoic doctrine2 (Abel 1985a: 711, Stowell 1999: 43). The undogmatic use of this, orientated toward the inward guidance of the addressee, is illustrated by the mild version of the apatheia ideal (nec te ad fortiora ducam praecepta: 4.1). One aspect worth mentioning is the striking similarity between the final sections of Seneca’s consolation and Cicero’s Somnium Scipionis (Armisen-Marchetti 2007).3 Topics and Content In contrast to the other two consolatory pieces, the Ad Marciam belongs firmly to the paramythetical genre. The main aim, the overcoming of pain, dominates the text (Abel 1967: 19f.). The choice and sequence of the arguments, however, seem to have been adapted to the character and mood of the addressee. Accordingly, the consolation begins with a piece of instruction by means of exempla (1–3), and this is followed by the praecepta. The beginning of the work is focused on Marcia’s personal exemplum on the occasion of her father’s suicide. By the re-publication of his works, writes Seneca, she has saved him from “real” death. The following exempla, Octavia and Livia, present protreptic and apotreptic models (cf. Shelton 1995). We then have philosophical statements from various sources as well as the lessons of experience on the transitory nature of mankind (5–11); these show the unnatural quality of excessive mourning.

2 On the philosophical significance of the consolatory works within Seneca’s oeuvre and the character of their thought in general, cf. Setaioli (infra, pp. 241–244). 3 Cf. Smith (infra, p. 359).

consolatio ad marciam

137

As is frequent in the literature of consolation, the selfish reasons for mourning (12.1–19.2) and the unselfish ones (19.3–25.3) are introduced in succession (Grollios 1956: 17f., Abel 1967: 16). This is done by considering first Marcia’s situation (refutation of the selfish reasons) and then Metilius’s (refutation of the unselfish reasons). Seneca gives the final words of the consolatory piece to Marcia’s father: the deceased Metilius, since he is now all knowing, is happy (26). Research One of the focal points of research is the examination of the structure of the piece on the basis of rhetorical and psychagogical considerations. The main point for Abel (1985a: 712) is the development of the motif “condicio humana,” whose major characteristic is subjection to death. In the fourstage development of this motif (9–11, 17f., 20.1–3, 25f.) he perceives a “crescendo,”4 which is intensified to the point of a “fortissimo” with Cremutius Cordus’s entrance (Abel 1985a: 712). 5 The ostensible contradiction between the recommendation of Peripatetic metriopatheia and Stoic apatheia as the suitable attitude of the soul is resolved by means of a persuasion strategy, with the help of which the addressee (and the reader) is led imperceptibly step by step to the desired attitude of apatheia (Abel 1967: 21f.). It is not Seneca’s goal to soften Marcia’s grief, but to achieve an end of it (Stowell 1999: 44). At the same time, it should be borne in mind that mourning itself, which for a certain duration was accepted (and indeed expected) in Rome, stands in contradiction to the strict observance of the apatheia ideal. To this ideal both the addressee and reader must first of all be gradually brought (Shelton 1995: 174). There are also numerous other elements in the presentation, e.g., the choice of exempla that are particularly close to Marcia (Shelton 1995), and these reveal Seneca’s skill in the psychagogical shaping of the piece to suit the needs of the addressee. Many arguments turn out to be familiar topoi of philosophical or rhetorical origin such as abound in the rhetoric of declamation (De Vico 1969: 139 f.). There are in addition some direct reminiscences of certain philosophers and

4 The central significance of “crescendo technique” as a principle of composition in the tragedies is noted by Steidle 1944: 257. 5 According to Stowell 1999: 103, Seneca “first tries to dismantle Marcia’s mistaken assent and impulse to grief. At 19.1, he announces the shift in consolatory direction” towards seeing what needs to be healed, and then, how it can be healed.

138

jochen sauer

works.6 However, recognizing these as such is in no way a precondition to understanding the text: these reflections will readily be understood both by the learned and the ordinary reader (Fillion-Lahille 1989: 1611). The question of a possible political intention behind the Ad Marciam has aroused the interest of some scholars. In mentioning Cremutius Cordus and the recovery of freedom of thought, Seneca, it is said, wishes on the one hand to exercise some influence on the young ruler Caligula (Borgo 1978: 71f.). On the other hand, taking up a position against Sejanus is interpreted as a desire to obscure an earlier connection with the latter (Stewart 1953, Fillion-Lahille 1989: 1613f.). When he praises Tiberius, he is detaching him from Sejanus, too, and putting himself firmly in the position of a loyal subject of the emperor. This has at times led to a charge of opportunism being leveled against Seneca. The prominent exemplum of Livia has continued to attract attention. Here, it is said, Seneca wants to keep before our eyes the picture of Augustus as the ideal ruler (Galdi 1928: 220f.). It may be, however, that Seneca means to flatter Caligula by praising his great-grandmother (Lana 1955: 88f.). On the whole, however, in view of our insufficient knowledge of the conditions of the period, a certain restraint with regard to political interpretation would seem to be advisable (Abel 1985a: 711). Language and Style In comparison with the other two consolationes the Ad Marciam reveals, both in matters of content and style, a closer connection to the rhetoric of declamation. Among the many features characterizing the style of this dialogus we may mention the abundance of the exempla given, the intensity of the exhortations, the recurring questions, the insertion of direct speech from the lips of fictitious persons, who frequently answer the questions of the main speaker (De Vico 1969: 140f.). A noticeable feature is the absence of a partitio. In its place there are several signpost expressions, which ensure a clear structuring of thought (12.1, 17.1, 19.3) and produce an impression of relaxed, easy-flowing speech (Abel 1967: 53f.),7 very different from the style of the consolatory piece Ad Helviam.

6 7

In 32.2, for example, Plato’s Phaidon (64A, 67D) is quoted indirectly. Cf. Tac. dial. 19.1 f.

consolatio ad marciam

139

Transmission The fundamental facts about the transmission of the text are to be found in Reynolds (1968 and 1977). The basis of the tradition is the Ambrosianus (Reynolds: A), together with a group of manuscripts (Reynolds: g), the best of which are independent of A and, although less reliable than the Ambrosianus, still preserve a useful tradition of the archetype.

DE VITA BEATA1

Fritz-Heiner Mutschler

Date and Circumstances of Composition Concerning the date we are better off with De vita beata than with the majority of Seneca’s dialogi.2 The first hint is provided by the name of the addressee, Seneca’s elder brother Novatus, who as an adult was adopted by the rhetor Iunius Gallio. Since Seneca addresses him as Gallio, the adoption is the terminus post quem of the essay’s composition. Unfortunately, it cannot be dated more precisely than to the years between the composition of De ira, in which Seneca addresses his brother still as Novatus and which presupposes the death of Caligula (ad41), and Gallio’s proconsulship in Achaia (ad51–52; cf. Act. Ap. 18.12f.). Other considerations, however, allow us to further narrow the time of composition. The author obviously lives in great state and has influence and power (cf. esp. 2.4). This seems to exclude both the years of Seneca’s exile (ad42–49) and the years after his retirement from the court (ad62–65) and, perhaps, to suggest that De vita beata came into being after the accession to power of Seneca’s alumnus Nero (ad54) and before the beginning of the estrangement between the emperor and his teacher (ad 59). Such a date is further supported by Tacitus’s report on a trial of the year ad58 (ann. 13.42f.) in which the accused, P. Suillius, directed exactly the kind of reproaches toward Seneca against which the philosopher defends himself in the second half of De vita beata. This does not mean that the essay is a direct response to Suillius’s accusations,3 but it shows that Seneca’s situation in the years between ad54 and 59

1 Commentaries: Grimal 1969, Viansino 1992, Kuen 1994; Studies: Pohlenz 1941 (1965), Ferguson 1958, Dahlmann 1972, Fuchs 1973, Griffin 1976 (1992): 286–314, Stroh 1985, Esposito 1988, Abel 1989 (1995), Chaumartin 1989a: 1686–1698, Asmis 1990, Mutschler 1990, Günther 1999, Blänsdorf 2005. 2 Cf. Gercke 1895 (1971): 299–306, Albertini 1923: 31f., Pohlenz 1941 (1965): 77, Giancotti 1957: 310–362, Grimal 1969: 17–21, Griffin 1976 (1992): 309f. and 431 n. 2, Grimal 1978a: 293f., Viansino 1992: 97–100, Kuen 1994: 21–24. 3 As Gercke 1895 (1971) has it.

142

fritz-heiner mutschler

and in particular in or around ad 58 would provide a fitting background for the essay’s composition.4 Structure The introductory paragraph (1.1) has been interpreted as an announcement of a division of the essay into two parts: the first dealing with the goal of human beings, beate vivere, the second with the path to achieving this goal (Grimal 1969: 7 and 82). This suggestion has been rejected for good reasons (Stroh 1985). When Seneca wants to give indications as to the disposition of a text he marks them clearly as such. That De vita beata is composed of two main parts is, nevertheless, correct, but it is divided in such a way that the first part is devoted to determining both the vita beata and the conditions of its realization (1–16), while the second is concerned with the problem of the relation of philosophical teaching and personal life (17–28). To which extent the two parts cohere is a matter of debate. On the one hand, it cannot be denied that the emphasis shifts from the philosophical issue of the definition of the vita beata to a defense against personal attacks (on philosophers in general and Seneca in particular).5 On the other hand, the problem of wealth, which is at the center of the second part, is also of importance within the more general question of what constitutes a happy life.6 The internal subdivision of the two parts blurs the bipartition of the whole because in both of them Seneca works with small units of one or two chapters’ length,7 which are connected with each other not in a strict logical sequence, but in a relatively free train of thought.8 Thus, after the introduction of the theme and a first approach ex negativo (1.1–3.1) there follows a series of definitions of the vita beata or the beatus, each of which links up with the preceding one and at the same time introduces a new aspect (3.2–6.2). Chapters 5 and 6 connect the definitions with a discussion of the objection of an epicurean interlocutor and thus form a transition to the next

4 This is now the opinion of the majority of scholars. For a more pessimistic view, see Giancotti, op. cit. 5 This shift has been noted by all commentators. 6 For other connections cf. Griffin 1976 (1992): 308–310. 7 See, e.g., the surveys in Albertini 1923: 76–78, Mutschler 1990: 199–201, Kuen 1994: 15–18, and Günther 1999: 25–28. 8 For interpretations of the dialogue, which follow this train of thought step by step, see esp. Pohlenz 1941 (1965) and Abel 1989 (1995). The examination of the logical coherence of the text is the main interest of Günther 1999.

de vita beata

143

sections in which the dispute with Epicureanism (and possibly with other doctrines, too) is in the foreground, while here and there new definitions are added (7–16). The second part is structured similarly. Short sections of one or two chapters follow one another, discussing first more general reproaches of disagreement between philosophical teaching and personal life (17–20), then more specific objections against the wealth of philosophers (21–24.4), before at the end Socrates, in an effective prosopopoeia, takes up the defense of the wise man and the repudiation of his opponents (24.4–end). Topics The subject of the essay is the vita beata. How does Seneca define it? What is his philosophical position on this issue? Concerning his affiliation to one of the philosophical schools Seneca leaves no room for doubt. Already in the first chapters he states that he is writing as a Stoic, though reserving the right to his own opinion (3.2, cf. Kuen 1994: 365f.). This statement is valid: Seneca’s position is the Stoic one, but individual accents are discernible. According to the orthodox Stoa, happiness (εὐδαιµονία) consists in ὁµολογουµένως ζῆν or ὁµολογουµένως τῇ φύσει ζῆν. Independently of whether both formulas go back to the founder of the school, Zeno, or only the shorter one stems from him and was supplemented by his successor, both mean the same: that happiness consists in the internal unity and harmony of a life, and that for achieving it accordance with nature is the necessary and sufficient condition. In full agreement with this doctrine, Seneca states:beata est ergo vita conveniens naturae suae (3.2) and idem est ergo beate vivere et secundum naturam (8.2) and later speaks of the internal unity and harmony of such a life as well (animi concordia, consensus, unitas: 8.6). The divine principle effective in nature is, for the Stoics, λόγος, ratio, reason. From this it follows that a life that is supposed to correspond with nature must be directed by reason. Seneca shares this point of view: potest beatus dici, qui nec cupit nec timet beneficio rationis (5.1, cf. also 5.3, 6.2, and 8.2). The area in which reason must above all prove itself is the judgment of goods. The position of the Stoa is clear, if somewhat complex. The only real good is moral goodness, that is, virtue, the only real evil is moral badness, that is, vice. All other things are neutral (ἀδιάφορα, indifferentia). Thus, a good life, in principle, must be directed toward virtus as the only good. However, the founder of the school already made certain distinctions within the domain of neutral things, some of which we are by nature disposed to prefer to their

144

fritz-heiner mutschler

opposites. Correspondingly, the reasonable man, whenever moral goodness and badness are not at stake, will decide for the preferable things (προηγµένα, praeposita) against their opposites (ἀποπροηγµένα, reiecta). Seneca adopts this view without restriction. He stresses repeatedly that the vita beata will be realized only by those who consider virtue the only good, vice the only evil, and all other things neutral: summum bonum est animus fortuita despiciens, virtute laetus; […] beatum dicamus hominem eum, cui nulllum bonum malumque sit nisi bonus malusque animus […] (4.2). On the other hand, he states several times that the reasonable man will strive for the preferable things, without binding himself to them: beata est ergo vita […] corporis sui pertinentiumque ad id curiosa non anxie, […] aliarum rerum quae vitam instruunt diligens sine admiratione cuiusquam […] (3.3, cf. also 8.2, 21.2, 23.4, and 26.3). Thus, Seneca’s statements are in general firmly based on orthodox Stoic doctrine. Individual accents are discernible with respect to two points. On the one hand, the emphasis placed on harmony, joy, and cheerfulness as components of the happy life is striking: hunc [scil. animum] ita fundatum necesse est, velit nolit, sequatur hilaritas continua et laetitia alta atque ex alto veniens (4.4, cf. also 3.4). 9 This seems to be in contradiction to Chrysippus’s doctrine that “virtue and reasonable sensation and reasonable drive and such are always present in all reasonable men; joy, however, and cheerfulness neither in all reasonable men nor always.” (Stob. II.77.6f. = SVF III.113). The least one can say is that Stoic doctrine was not uniform with respect to this point and that the intensity with which Seneca points to joy as a necessary component of the vita beata is in tension with other Stoic statements. Something similar can be observed with respect to the problem of wealth, with which the second part of the essay is to a large extent concerned.10 Apparently, the older Stoa qualified wealth simply as a προηγµένον without discussing it further, while representatives of the middle Stoa, like Diogenes of Babylon, Antipater of Tarsos, Panaetius, and Posidonius, seem to have treated the pertinent questions in more detail and emphasized the positive aspects of wealth more strongly. The Seneca of De vita beata is clearly closer to the latter and possibly goes further than they do: the statement that wealth

Cf. Pohlenz 1941 (1965): 58 f., Asmis 1990: 233 f., Mutschler 1990: 191f. See the detailed discussion of Seneca’s views on wealth in Griffin 1976 (1992): 294–314. Griffin points out both the closeness of Seneca’s position in De vita beata to the teachings of people like Panaetius and Posidonius and its uniqueness in comparison with Seneca’s other writings. 10

de vita beata

145

makes the wise man cheerful ut navigantem secundus et ferens ventus, ut dies bonus et in bruma ac frigore apricus locus (22.3) is singular in his own writings and did probably not have many parallels in other Stoic texts.11 Language and Style In terms of literary form, De vita beata is a typical Senecan dialogus. It is not a sober scholarly treatise as are, for example, the preserved works of Aristotle. But it is different from Plato’s or Cicero’s philosophical dialogues, with their imitation of real conversations, too. Instead, De vita beata is an essay of a middle scale, which treats its topic in a sequence of small sections and forces upon the reader a few basic ideas with penetrating persuasiveness. One important device of this persuasiveness is the technique of repetition with variation. It serves to make the fundamental points over and over again in new constellations and with new emphases and thus to imprint them on the soul of the reader as deeply as possible. The long series of definitions of the vita beata, which starts at 3.3, is a good example. Language and style, too, are directed toward psychagogical effect. Instead of long, wellbalanced periods, there are mostly short, aphorism-like sentences; easily understandable parataxis has preeminence over logically differentiating hypotaxis; punch lines, antitheses, and—sometimes bold—metaphors lead again and again to impressive formulations. In addition, there are numerous objections by an anonymous interlocutor, which add piquancy to the text and keep the essay going. And, finally, historical examples, often presented in anecdotal form, and prosopopoeiae give distinctness and emphasis to what is said.12

11 There is another point where Seneca may have put in a special accent by taking up the ideas of Panaetius. In his very first definition of the vita beata, he speaks of a life conveniens naturae suae (3.2). Asmis 1990: 224f. understands the suae as referring not to nature in general, but to each person’s individual nature—as it is taken into account in Panaetius personaedoctrine (cf. Cic. off. 1.107f.). She then interprets several features of De vita beata (as, e.g., the emphasis on joy and the justification of wealth) in this context, that is as adaptations to Seneca’s particular personality. This interpretation of suae, however, is less than certain and is not shared by many scholars. 12 All these devices are described, with lists of instances carefully documented, in Kuen 1994: 407–435.

146

fritz-heiner mutschler Sources

Seneca was, of course, well versed in Stoic literature (and in that of other schools, too). But in spite of the fact that the commentaries, for many passages of the essay, point out parallels to the writings of earlier Stoics (and of other schools, too), practically none of them is specific enough to prove a reference to one particular text. In the chapters that deal with the definition of the vita beata, that is, with the central topic of ancient ethics, this is not surprising, as this topic was treated not only in the writings of individual philosophers, but in compendia and handbooks as well.13 But this is also the case with the second half of the essay, which is concerned with the more specific problem of the proper attitude toward wealth: although it is clear that here the doctrines of the middle Stoa were of particular importance for Seneca, we are in no better position to determine specific dependencies. In De vita beata Seneca quotes passages of Roman poetry on four occasions. Both the selection and the kind of usage are typical: three quotes from Vergil (8.3: Aeneid 2.61; 14.4: Georgics 1.139; and 19.1: Aeneid 6.653) and one from Ovid (20.5: Metamorphoses 2.328); in all four cases Seneca isolates the citations from their original contexts and uses them in his own sense (Kuen 1994: 426–428, Tarrant 2006: 3). Textual Transmission The end of De vita beata presents one of the grave problems of transmission in Seneca’s dialogi. Although the index of works in the Codex Ambrosianus contains the entry ad /////// de otio, a work bearing this title is not found in the manuscript. Following the suggestion of Muretus, Lipsius, in his edition of 1605, proposed in chapter 28 a lacuna after cir and attributed what follows to (mutilated) De otio. This division of the two texts has found general approval among editors. The implication is that the end of De vita beata and the beginning of De otio are missing. As to De vita beata it is improbable that much has been lost. The prosopopoeia of Socrates gives a powerful ending to the dialogus, and it is difficult to imagine that after this Seneca once again spoke in his own name. 13 How difficult it is to determine the reference text(s) of a passage of Seneca can be seen from the ongoing discussion concerning chapter 15: It seems clear that Seneca is treating here the doctrine of another school, different from the Garden, with which he disagrees. But no consensus has been reached as to which school this is (cf. esp. Pohlenz 1941 [1965]: 62 f., Grimal 1967a, and Chaumartin 1989a: 1694–1697).

DE OTIO1

R. Scott Smith

Date The incomplete state of the text and the lack of any historical allusion make dating De otio speculative. In the absence of certainty critics have either conjectured specific dates from a biographical reading of the text or have attempted to establish a relative chronology based on the philosophical evolution of the addressee, Annaeus Serenus (to whom both De constantia sapientis and De tranquillitate animi are also dedicated). With the latter approach we are immediately confronted with a problem: Serenus may not be the addressee. Nowhere in the surviving fragment is the addressee identified, and the name in the index of the Codex Ambrosianus has been effaced. If Serenus is the addressee—perhaps indicated by the “ardent and stubborn” personality exhibited by the interlocutor, reminiscent of the other works dedicated to him (Griffin 1976: 354 n. 2; but see Dionigi 1983: 55–57)— two conclusions may be drawn: 1) a relative chronology of De constantia sapientis—De tranquillitate animi—De otio may be established based on Serenus’s conversion from Epicureanism to Stoicism, and 2)epist. 63 (dated to ad 63–64), in which Serenus’s death is reported, becomes a terminus ante quem. It is also commonly argued that De otio, with its encouragement to retire from public affairs, reflects Seneca’s political position vis-à-vis the court of Nero, i.e., around his retirement in ad 62. The preponderance of critics, drawn to this biographical approach, incline toward this position, but even within this large group opinion differs whether to view De otio as a prelude to his withdrawal from politics, as an ensuing defense of it, or as a concurrent document illustrating his reasons for it (see Dionigi 1983: 49–53). Parallels

1 Commentaries: Williams 2003, Dionigi 1983; Studies: André 1989: 1744–1747, 1772–1778, Schwamborn 1951; De otio and political participation, Griffin 1976: 315–366 (De otio: 328–334); otium in Seneca and Roman thought, Dionigi 1983: 66–77, André 1962, 1966, Grilli 1953; Stoics and choice of life, Joly 1956: 143–147.

148

r. scott smith

with the letters, esp. epist. 7, 8, and 68, argue for a late date, but Seneca may be returning to a topic treated many years earlier (see Giancotti 1957: 225–243, Griffin 1976: 316f.). Transmission and Content The index of works in the Codex Ambrosianus includes an entry ad […] de otio (as noted above, the name has been erased), but this work is nowhere found under a separate title in the manuscripts. In 1585, Muretus detected an inorganic suture in De vita beata 28 and posited that the material from there to the end was a fragment from another work. Lipsius, in his 1605 edition, separated this fragment from the preceding dialogue and added the titleDe otio aut secessu sapientis libri pars. What survives is a fragment of a text that is mutilated both at the beginning and the end. How much is missing from the beginning of De otio we cannot say for certain, but it is unlikely to be much. The opening lines have the color of an exordium; at 1.4 Seneca portrays an interlocutor raising a lengthy objection (a technique found early in other dialogues), and at 2.1 we have a formal divisio—all of which suggests that we are not too far along when the transmitted text begins. A reasonable speculation is that at most a few pages have been lost. For a host of reasons most hold that De otio is incomplete at the end (though see Dionigi 1983: 42–48). Given the shortness of the work (only 8 OCT pages) and the unlikelihood that the initial lacuna is extensive, it is probable that we have lost much of the end of De otio, perhaps, as Griffin notes (1976: 332), as much as half of the work. An overview of the work follows: Ch. 1 (exordium): Seneca urges withdrawal from the vicious crowd into otium (1.1–3); an unnamed interlocutor (Serenus?) objects that this violates the Stoic principle of an “active” life (1.4). Ch. 2 (divisio): After asserting his independence from his Stoic predecessors (1.5), Seneca sets out to prove his position is in fact orthodox, dividing his discussion into two probationes: one may choose otium 1) from an early age before entering public life (a prima aetate: 2.1), or 2) after a long career of public service (emeritis stipendiis, profligatae aetatis: 2.2). Ch. 3f.: Seneca first sets out a summary of the reasons (causae) why a Stoic might opt out of public service (ch. 3; cf. 6.1–4, 8.1–end), then seeks to

de otio

149

establish positive arguments for withdrawal (ch. 4 f.): first, there are two res publicae, the local (minor) and the cosmic (maior), the latter of which is served better in otio (4.2); second, Nature has fashioned us for contemplation in addition to action (ch. 5). Ch. 6f.: Contemplation is not inactive: the exempla of the three Stoic scholarchs, Zeno, Cleanthes and Chrysippus, prove that one can better benefit humankind in otio than by participating in government (6.4f.). In fact, all philosophical schools involve contemplation to some degree (ch. 7). Ch. 8: The lex Chrysippi allows for a vita otiosa if the res publica minor is not suitable (8.1). Close scrutiny reveals none is worthy of a sapiens (8.2); with the surprising statement that otium is necessary for all (8.3) the text ends awkwardly with an unanalyzed analogy. Topics At the center of this treatise stands a fundamental question: what is the appropriate life for a Stoic philosopher to lead? Although the Stoics’ position on this is by no means monolithic, they advocated political involvement in principle. Seneca reports their basic position: “the Wise Man will engage in politics unless something prevents him” (3.2; cf. Diog. Laert. 7.121). Despite their advocacy of an active life, the Stoics recognized that in certain conditions one might justifiably opt out of a public life. But since the Stoic principle of political engagement converged with Roman cultural norms— after all, the élite were expected to lead a political life—Stoicism, in the view of the typical Roman, became a byword for political participation. When set against the position of its rival philosophical school, Epicureanism, which espoused a purely private life (though in fact many Epicureans did engage in politics), the popular view was reinforced: Stoicism stood for the active life (negotium) in polar opposition to the Epicurean easy life of indolence (otium). Against this cultural backdrop, the (imaginary) interlocutor’s criticism of Seneca’s call to a vita otiosa is understandable but fundamentally misguided (cf. epist. 68.10). In rebuttal, Seneca himself provides a list of circumstances when the Wise Man will not participate in public affairs (3.3): 1) if the state is too corrupt to be helped and beset by evil men; 2) if he does not have sufficient influence or power to be effective; and 3) if he is in ill health. Thus, both Epicureanism and Stoicism allowed for a life of otium, though for

150

r. scott smith

different reasons; for the former it was an intentional choice (ex proposito), while for the latter it was a response to conditions (ex causa), like a temporary haven in a storm (7.4). Seneca then sets out to prove that otium, far from necessarily being equal to a sedentary life of pleasure, can in fact be active—provided that one moves from a narrow view to a cosmic perspective, one bounded neither by space nor time (see Williams 2003: 10–12). There are two res publicae (4.1 f.), the lesser (one’s local community), and the greater (the cosmos). One may work to benefit either or both, but Seneca argues that one can better serve the greater republic in otio, for there one can contemplate the fundamental questions of the universe and help all of humankind (cf. dial. 9 [= tranq.].3.3), not just those of our time and place. Such contemplation conforms to the more fundamental Stoic principle underlying political participation, that one will seek to benefit other humans and society at large (3.5). Indeed, the Wise Man will only retire into otium “in the knowledge that even then he will be performing actions through which he will benefit future generations” (6.4; cf. epist. 8.2). By expanding our view to encompass the entire cosmos, we realize that otium allows us “to set future generations straight and to hold court not with just a few but with all men of every nation, both those now and those to come” (6.4), just as the great Stoic philosophers had done (6.4f.). Of course, in perfect conditions contemplation is to be tied directly to action (6.2f.; cf. Cic. off. 1.153f.), but when the situation demands it, the Wise Man will withdraw in order to expend his efforts to some good end, that is, in serving the maior res publica.2 Language and Style Although one of Seneca’s more theoretical works, De otio is varied in terms of style and register, ranging from rigorous, logical argumentation in the body of the treatise (esp. 3.2–5, 4.1, 5.1, 7.1–4) to flights of poetic fancy (e.g., 5.2–8) and rhetorically vibrant displays at the fragment’s beginning and end (1.1–4 and the interrogatio at 8.2). The opening chapter, imbued

2 By tying contemplation to action Seneca is fully in line with the type of life advocated by the Stoics. At Diogenes Laertius 7.130 we find that, among the contemplative life, the active life, and rational life (bios logikos), the Stoics preferred the last since humans, as rational animals, were created by Nature both for contemplation and action (cf. Cic. off. 1.15–17, fin. 2.40; Seneca, Otio 5.1, epist. 94.45; Plut. lib. educ. 7f–8a). Posidonius (fr. 186 E.-K.) directly ties contemplatio to the telos (goal) of human life.

de otio

151

with political/electoral language (see Ingrosso 1988), vividly emphasizes the detrimental effect of the fickle crowd with frequent rhetorical devices: anaphora (tunc potest: 1.1; aliud ex alio: 1.2); antimetabole in isocolon and homoeoteleuton (petita relinquimus, relicta repetimus: 1.2); and correctio (1.3, twice). The rhetorically forceful language continues with the interlocutor’s imagined response (sermocinatio), beginning with staccato questions (quid agis, Seneca? deseris partes?: 1.4) and ending his objection with powerful alliteration with plosives and sibilants: quid nobis Epicuri praecepta in ipsis Zenonis principiis loqueris? […] si partium piget, transfugis potius quam prodis? Seneca similarly uses sermocinatio at dial. 2 (= const.).3.1 to establish a conventional opinion as a starting point and later in De otio to mark transitions and to raise obvious objections (6.1, 6.5, 7.2), as often elsewhere (e.g., dial. 1 [= prov.].2.1, 5.9, 6.1, 6.6; dial. 2 [= const.].1.2, 3.3, 4.1, 4.3, etc.). Throughout the work Seneca seeks to minimize the differences between Stoicism and Epicureanism as part of his rhetorical strategy to redefine otium (3.2f., 6.1, 7.1 deposita contentione, depositoque odio). In the long series of physical, theological, and ethical questions at 4.2, 5.5 f. (which neatly reflect the very act of contemplatio), he places both Stoic and Epicurean theories side by side without prejudice, employing Lucretian language liberally (e.g., gignuntur […] diducta et solidis inane permixtum; dei sedes: 4.2, gravia descenderint, evolaverint levia: 5.5). At 5.6 the personification of human cogitatio “bursting through heaven’s walls” (caeli munimenta perrumpit) recalls Lucretius’s encomium to Epicurus (1.70–79, esp. effringere ut arta | naturae primus portarum claustra cupiret: 70–71). Sources Given that Seneca strongly asserts his intellectual independence (1.5, 3.1), that no early Stoic text on political participation has survived completely intact, and that the topic was often debated in Greco-Roman rhetorical circles (Cic. top. 82, de orat. 3.112; Sen. epist. 14.13f.; Quint. inst. 3.5.6–8; cf. Cic. Att. 9.4.1 f.), it is difficult to judge whether, or the extent to which, Seneca drew on specific sources. Chrysippus’s Περὶ βίων (“On the Kinds of Life”) may have been influential, probably indirectly through some intermediary. Although attributed to Zeno at 3.1, the Stoic principle of political participation may properly belong to Chrysippus, who articulated precisely this position in the first book of his Περὶ βίων (Diog. Laert. 7.121). At 8.1 Seneca examines the lex Chrysippi that permits one to choose a vita otiosa, although what follows is attributed to Stoics (nostri) in general (cf. epist. 68.2), and the

152

r. scott smith

discussion breaks off with the end of the text. For Stoic exceptions to political participation (3.3) we rely mainly on testimony from the so-called Middle Stoa and later: see especially Arius Didymus (SVF 3.690); cf. Cic. off. 1.71–73 and 107–121 (the four personae, presumably translating Panaetius); Sen. dial. 9 (= tranq.).6; Epikt. diatr. 2.6.9; see also Dionigi 1983: 79–86. The content of chapter 5, concerning humankind’s innate desire for knowledge, ultimately derives from Aristotle’s lost Protrepticus, though not directly nor necessarily through Antiochus of Ascalon, as Dionigi (1983: 86–95) suggests based on close similarities to Cic. fin. 5.48–52.

DE TRANQUILLITATE ANIMI 1

Fritz-Heiner Mutschler

Date and Addressee The genesis of this dialogue can be dated only to a relatively extended period between Seneca’s return from exile in ad 49 and the death of its addressee, L. Annaeus Serenus, which occurred probably some time before ad 62 (cf. Griffin 1976 [1992]: 447f.). Several interpreters take the dialogue to reflect more precisely Seneca’s situation around ad 60 when, after the murder of Agrippina, the philosopher was already considering his retreat from the court.2 But with the text giving no specific indications of Seneca’s circumstances, this is only a conjecture. The data of the public career of the addressee—amant en titre of Nero’s mistress Acte in 55 (Tac. ann. 13.13.1) and praefectus vigilum probably from 55 until his death (Griffin loc. cit.)—do not help us either (Giancotti 1957: 165). That the relationship of Seneca and Serenus was a close one is indicated by the address carissime Serene (4.1 and 17.12) and confirmed by Seneca’s confession that he lamented the death of his friend immoderately (epist. 63.14). De constantia sapientis and (possibly) De otio are addressed to Serenus as well. As to the relative chronology of the three dialogues, it is mostly assumed that De tranquillitate animi, in which the friend appears as a Stoic proficiens, is preceded by De constantia sapientis, in which he seems to show epicurean inclinations, and followed by De otio, in which his Stoicism is consolidated. However, other theses have been argued as well.3

1 Commentaries: Cavalca Schiroli 1981, Viansino 1992, Lazzarini and Lotito 1997, and Parenti 2004; Studies: D’Agostino 1929, Barigazzi 1962, André 1989, Maurach 1991: 123–135, Abel 1992 (1995): 3–36, Motto and Clark 1993a, Blänsdorf 1997, Lefèvre 2003, and Nocchi 2008. 2 Münscher 1922b: 61, Albertini 1923: 37–39, Pohlenz 1941 (1965): 98, Lana 1955: 252, André 1989: 1730, Motto and Clark 1993a: 134, and Lefèvre 2003: 164. 3 Grimal 1978a: 286–293; Williams 2003: 16.

154

fritz-heiner mutschler Contents and Structure

Among Seneca’s dialogues, De tranquillitate animi stands out by its beginning, as it starts with a long statement of Serenus (1), to which the rest of the dialogue represents Seneca’s answer. People have spoken of a letter of Serenus, but there is no formal indication of this.4 Rather, for once in the dialogi, Seneca’s vis-à-vis is a partner with a voice of his own. The subject of Serenus’s utterance is his state of mind. It is characterized by his inability to keep to his intentions, by a continuous back and forth in the orientation of his life. His request is for a remedy against this fluctuatio and for the means to achieve tranquillitas (1.17). In his first response, Seneca asserts that Serenus is already on the way to recovery (2.1 f.). He determines the state of mind toward which his friend is aiming more precisely (2.3f.) and announces that he will investigate a means of achieving it in a way that Serenus can select what he considers appropriate for himself. Beforehand, however, Seneca gives his own description of the vitium, in order to enable everyone to recognize his or her own version of it (2.5–15). With the beginning of Chapter 3, Seneca starts his elaborate treatment of the various means of help, which extends to the end of the dialogue. The disposition of the main part of De tranquillitate animi has caused some discussion. Several suggestions have been made,5 but no agreement has been reached. This has its reason in the text, not in the idiosyncrasies of its interpreters. As in other writings of Seneca, the structure is one of variation and addition rather than of hierarchical order. Thus, it is unproblematic to distinguish small units of a few chapters, but difficult to point out more general structural principles. Seneca starts out by discussing critically Athenodorus’s advice to retreat from public life when difficulties arise (3–5). Next, he accentuates the need, before undertaking some business, to examine (a) oneself, (b) the task, and (c) the people with whom and for whom one is going to act. The following sections deal with the disruption caused by possessions (8f.), elaborate on adequate behavior in difficult situations (10.1–4), and treat the problem of desires (10.5–7). Shortly after the middle of the dialogus Seneca introduces— as nonaddressee of his text—the sapiens, which leads him to a description of the wise man’s being unperturbed by strokes of fate and a discussion of the

Cavalca Schiroli 1981: 9 f., Maurach 1991: 123 f., Abel 1992 (1995): 13f. See, e.g., Albertini 1923: 279–282, André 1989: 1765, Maurach 1991: 129 f., Abel 1992 (1995): 11–13, Lefèvre 2003: 155. 5

de tranquillitate animi

155

value of the praemeditatio malorum (11). Refering to what has been discussed before, Seneca insists on the necessity of avoiding useless tasks (12f.). The next section is mainly concerned with the mind’s independence from all that is external; this independence is illustrated with several examples, the most extensive being that of Canus Iulius in his confrontation with Caligula (14). As the right attitude toward the shortcomings of the vulgus, Seneca advises placid acceptance and composure (15). A graver problem is that of the (seemingly) evil end of good men, which dissipates, however, on closer inspection (16). In the last section, Seneca points out the advantage of being natural (17.1 f.) and the value of finding the right balance between sociability and solitude (17.3), exertion and relaxation (17.3–7), and sobriety and intoxication (17.8–11) before he concludes his presentation with the admonition that intenta and assidua cura is necessary if tranquillitas is to be obtained (17.12). Language and Style Its unsystematic nature and cumulative structure distinguish De tranquillitate animi from the other dialogues only in degree, not in principle. One observes here the influence of the Hellenistic-Roman diatribe and of declamatory rhetoric as it developed in the early imperial period. In both genres the aim is less to convince the audience by coherent logical argumentation than to affect and impress it by the application of multiple devices of—partly popular—rhetoric. Though the formal exchange with Serenus is restricted to the first two chapters—later on the friend is addressed only two more times (4.1 and 17.12)—and the “you” of the text becomes that of the general reader, bits of direct speech repeatedly enliven the text. Thus, in Chapter 11 there are statements in oratio recta of the imagined sapiens (11.3), Fortuna (11.5), and the animus (11.9), and Chapter 14 presents, on the whole, no fewer than nine instances of direct speech. Another means of attracting the reader’s attention is imagery.6 Pointed formulations, be they antitheses or simple sententiae, enliven the text throughout. Furthermore, it contains many quotations from both the writings and sayings of prominent people.7 Finally, De tranquillitate

6 See the extensive treatment by Armisen-Marchetti 1989. Cf. also Nocchi 2008: 71–100, esp. 87–93 on imagery connected with the sea. 7 3.2–8: Athenodorus; 8.7: Diogenes; 9.5: Livy; 11.4: Cicero; 11.8: Publilius Syrus; 13.1: Democritus; 14.3: Zeno, Theodorus; 15.4: Bion; and 17.10: Graecus poeta, Plato.

156

fritz-heiner mutschler

animi is full of exempla:8 Greek (Socrates: 5.2–4, 16.1, 17.4 et al.), Roman (Cato minor: 7.5, 16.1, 16.4, 17.4, 17.9 et al.), and others (like Croesus: 11.12), presented individually or in groups, simply named or elaborately evoked (cf. in particular Canus Iulius: 14.4–10).9 Topics “Taedium sui.” De tranquillitate animi opens with one of the most impressive passages of Seneca’s writings, a twofold description of the problematic state of mind of which Serenus wishes to be cured by Seneca. The description uses the ethical and psychological categories of contemporary school philosophy10 and illustrates the phenomenon both with reference to the social realities of its time11 and by a number of unusual images, which give the text a specific flavor.12 Most striking is, however, the long series of terms by which the phenomenon in question is characterized (animi infirmitas: 1.4; bonae mentis infirmitas: 1.15; fluctuatio: 1.17; fastidium [sui]: 2.5; sibi discplicere: 2.7; animi iactatio, cunctatio vitae: 2.8; taedium et displicientia sui, animi volutatio, sui tristis et aegra patientia: 2.10) and which finally lead to the climax fastidio esse illis coepit vita et ipse mundus et subit illud tabidarum deliciarum “quousque eadem” (2.15).13 “Tranquillitas animi.” Already within the first part of his description of the vitium in question Seneca outlines the positive goal toward which the therapy is directed (2.3f.). This small section is interesting for two reasons. First, although the formulation non concuti (2.3) seems to refer to Stoic ἀπάθεια or ἀταραξία, the following reference to Democritus’s more inclusive and less definite εὐθυµία shows that Seneca avoids confining himself to a single, strictly defined Stoic concept and tries to secure for his key term tranquillitas animi,

On both quotations and exempla, cf. Nocchi 2008: 101–125. Canus Iulius represents one of the exempla from Seneca’s own time, which are supposed to show the reader that exemplary behavior is still possible in the present (cf. Mayer 1991: 152). 10 Serenus distinguishes types of vitia (1.1), calling his present habitus a vitium (1.2). Seneca, too, speaks of a vitium (2.5) and contributes to its analysis by discussing the problem of the cupiditates (2.7–10). 11 Cf. Serenus’s description of his indecisiveness with respect to life style, public engagement, and intellectual and literary activity (1.4–15) as well as Seneca’s reference to the phenomenon of travel addicts (2.13 f.). 12 1.2: Serenus ~ patient between illness and health, 1.17: Serenus’ psychic state ~ nausea, 2.1 f.: Serenus’ affectus animi ~ tremor of the calm sea after a storm. 13 Cf. André 1989: 1767: “Sénèque a donné un vocabulaire à la pathologie mentale, au malaise de l’ âme.” 9

de tranquillitate animi

157

which he presents as a translation of εὐθυµία, a more comprehensive meaning (Hadot 1969: 136). Second, tranquillitas denotes a certain state of mind. On the other hand, shortly afterward Seneca programmatically asks the question: quomodo animus semper aequali secundoque cursu eat propitiusque sibi sit […] (2.4). That means that the problem is not simply how to reach a momentary state of mind, but how to gain and permanently retain inner satisfaction by leading one’s life in a proper way. For this reason it is understandable that the major part of the remedia-section consists of a series of “miniature-treatises” (André 1989: 1769) dealing with concrete problems of how to arrange one’s life. “Vita activa” and “vita contemplativa.” The problem that is treated first and most extensively is that of the relation between vita activa and vita contemplativa. This topic concerned Seneca repeatedly and his attitude toward it changed according to the changing circumstances of his life. 14 Here, he discusses critically the views of Athenodorus Calvus (3–6). Whereas Octavian’s teacher too readily advised retreat from public life, should difficulties arise, Seneca pleads for more persistence and for a retreat only step by step if no other possibility is left. The difference is one of degree, but the example of Socrates under the oligarchy of the Thirty is evoked with force (5.1–3), and the idea that the wise man may serve the “greater republic” of mankind better in retirement (De otio 4.2) has not (yet) been formulated. “Right mixture”. In the section on the vita activa and the vita contemplativa Seneca already suggests that the optimum might be the right mixture of both (4.8). In the last chapter of the dialogue, the idea of the right mixture is taken up and further developed. Seneca suggests an alternation between solitudo and frequentia (17.3), but he advises in particular a proper balance between intentio and remissio of the mind: after effort and strain the mens / animus must be granted the proper measure of relaxation, be it in the form of joking and playing, of a walk or trip, or even of intoxication, provided it occurs in moderation (17.4–9). And, as Seneca—hardly in agreement with Stoic orthodoxy—adds, at times one must be out of one’s mind if something truly sublime is to be achieved (17.10f.).15 The end of the dialogue responds in a peculiar way to its beginning: what Serenus had perceived as a painful symptom of his disease is now—in reflected form—presented as a means of curing it (Blänsdorf 1997: 88–91).

14 15

21f.).

For detailed discussions see Griffin 1976 (1992): 315–366, and Günther 1999: 113–172. The passage is also of poetological interest (cf. Schiesaro 1997b: 98f., and Schiesaro 2003:

158

fritz-heiner mutschler Sources

In the case of De tranquillitate animi we are in the rare position of knowing a concrete “source text”: In the passage in which Seneca points out that tranquillitas is a translation of the Greek εὐθυµία he also states that there is an “outstanding” book on the subject by Democritus (2.4). In view of this reference, it is more than probable that Seneca used this volumen directly, even if there was a Περὶ εὐθυµίας by Panaetius, too, which Seneca does not mention, but which Cicero refers to and quotes from in his De officiis.16 The assumption of direct use is confirmed when Seneca quotes Democritus literally in 13.1, perhaps with the opening sentence of Περὶ εὐθυµίας. This does not, of course, exclude the possibility that he used Panaetius as well. Another source text explicitly referred to is a treatise of Athenodorus Calvus, which Seneca reports and criticizes extensively in his discussion of the problem of vita activa and vita contemplativa (3–5). Less clear is Seneca’s debt in the broad exposition of the vitium to be cured at the beginning of the work, but it is obvious that in some passages of the final parts of Books 3 and 4 of Lucretius’s De rerum natura (3.1006 is quoted in 2.14) just as in some passages of Horace’s Odes (cf., e.g., 2.16 and 3.1) and Epistles (cf., e.g., 1.8 and 11) we discover a strand of tradition that Seneca knew of and that inspired him. There are a number of other quotations in the dialogue that do not come from thematically related texts, but simply show that Seneca was well-read and knew how to insert a quotation at the right time and in the right place (see, e.g., 11.4: Cicero, 11.8: Publilius Syrus, and 17.10: Aristotle). Textual Transmission De tranquillitate animi is transmitted as part of Seneca’s dialogi and shares with its partners the specific situation of transmission of this corpus. There are two particular problems. First, given that the INCIPIT of the dialogue in the Ambrosianus is written by another hand, where the first hand had left a space of two lines, it is possible that the very beginning of the work, which may have introduced the antilogy of chapters 1 and 2, has been lost together with the missing end of the preceding De otio. Second, the passage

16 The question of sources is discussed in D’Agostino 1929, Hadot 1969: 135–141, Grimal 1978a: 141–146 and 344–353, André 1989: 1741–1744, Viansino 1992: 207–214, Abel 1992 (1995): 7–11, and Blänsdorf 1997, 75–77.

de tranquillitate animi

159

considerandum […] labor est in 7.2 does not seem to be in its proper place, but to have been displaced here from somewhere in 6. As to its original position, several suggestions have been made, the most probable being between 6.2 and 6.3.17 Reception De tranquillitate animi is a case where we can trace the influence of a text of Seneca in a precise historical context of some significance. This context is the discussion and analysis of the phenomenon of ennui in nineteenthcentury France. At the center of this complex we find Charles Baudelaire (Bouchez 1973: 85), in whose Fleurs du Mal the figure of l’ Ennui looms large in the introductory poem Au Lecteur. As a note in his Journaux intimes indicates,18 Baudelaire’s conception of ennui might have been influenced, at least indirectly, by the portraits of Serenus by Seneca and of Stagirius by John Chrysostomus19 found in the essays of the doctor and psychiatrist Alexandre Brierre de Boismont.20 Both portraits play an important role in these essays21 and together with Cassian’s descriptions of tristitia and acedia22 also appear elsewhere in the extensive debate about the mal du siècle.23

17 Cf. Albertini 1923: 184–186, Reynolds 1977: ad loc., Cavalca Schiroli 1981: 35 and 85, and André 1989: 1733. 18 Fusées IX: “Brierre de Boismont. / Chercher le passage: Vivre avec un être qui n’a pour vous que de l’ aversion […] / Le portrait de Sérène par Sénèque, celui de Stagyre par saint Jean Chrysostome. / L’ Acedia, maladie des moines. / Le Taedium vitae” (Oeuvres complètes, ed. C. Pichois, Paris 1975, vol. 1.656). 19 Ad Stagirium a daemone vexatum. 20 Cf. Clapton 1931, Bouchez 1973: 27–32 and 84–91, and Mandelkow 1999: 261–271. 21 De l’ ennui (1851) and Du suicide et de la folie suicide (1856). 22 De institutis coenobiorum et de octo principalium vitiorum remediis, Books 9 and 10. 23 Cf. Mandelkow 1999: 115–128, 255–285.

DE BREVITATE VITAE 1

R. Scott Smith

Date From 18.5 (modo modo intra paucos illos dies quibus C. Caesar perit) we know that Caligula was dead when Seneca wrote De brevitate vitae. On the basis of modo modo Lipsius (1615) dated the treatise to the early 40s, but since the phrase does not necessarily imply immediacy, this argument cannot be sustained. This is especially true if the addressee, Pompeius Paulinus (likely Seneca’s father-in-law), was praefectus annonae at the time of writing, as seems certain. Although Seneca does not identify this specific position, all indications point in this direction: Paulinus “managed the accounts of the whole world” and “understood the accounting of the public grain supply” (18.3; cf. 18.4f.). If this is correct, the historical record allows for two periods when Paulinus could have served in this capacity, 1) ad 48–55 or 2) ad 62–71. Scholars have generally fixated on two dates within these periods, ad 49 and ad62, when Seneca’s political situation could allow him to advise Paulinus to retire without appearing hypocritical. Such attempts to tie Seneca’s essays to his own political situation are, as always, dubious; it would hardly have been hypocritical for Seneca, even at the height of his political involvement, to advise a close family member of somewhat advanced age—perhaps 60 or older—to retire from such a demanding position. Even so, arguments in favor of ad 62 are much weaker than those for ad 49 (overview: Griffin 1962). Critics who champion 49 usually point to a terminus ante quem suggested by Seneca’s failure to mention Claudius’s extension of the pomerium to include the Aventine (13.8), which on inscriptional evidence is securely dated to his ninth tribunate (January 49–January 50). The argument that Sulla was the last to extend the pomerium, however, is not

1 Commentaries: Williams 2003, Traina 1984, Grimal 1959, Dahlmann 1949, Duff 1915. Studies: Abel 1965, André 1989: 1738–1739, 1747–1756, Blänsdorf and Breckel 1983, Dionigi 1995a, Giancotti 1957: 363–445, Griffin 1962, Grimal 1949a, 1960, Hambüchen 1966. Seneca on time: Gagliardi 1998, Armisen-Marchetti 1995b, Goldschmidt 1979 (205–210 on Brev.).

162

r. scott smith

Seneca’s own but is attributed to a learned pedant, who, as Griffin has well argued (1962: 108–111; contra Abel 1965, Hambüchen 1966: 26–81; response by Griffin 1976: 401–407), is asserting the illegality of Claudius’s extension of the pomerium. Thus, the passage establishes not a terminus ante but a terminus post quem of ad49. Griffin’s arguments to this end are persuasive; her contention, that De brevitate vitae is Seneca’s public apology aimed at providing Paulinus the means to resign his post gracefully to make way for Faenius Rufus in ad 55, is attractive but cannot be corroborated. On balance, a date between ad50 and ad 55 seems likely. Content De brevitate vitae, one of Seneca’s most loosely organized treatises, relies not on a rigorously structured plan but, on a constellation of ideas orbiting around a central theme (Albertini 1923: 258–260). Attempts (Grimal 1959, 1960) to give a detailed analysis that presupposes rhetorical divisions are strained. There is no expressed organizational principle, transitions from topic to topic are not well marked, and the dialogue alternates freely between critical depictions of the occupati (those consumed by meaningless activity) and exaltations of the philosophical life. The following overview thus offers only a descriptive outline of its contents. See also Williams 2003: 19, 21–24; and André 1989: 1747–1749. The treatise begins with an exordium establishing the topic: most humans, both the inprudens vulgus and clari viri, complain that human life is too short; but it is long enough if put to proper use (1). To demonstrate this point, Seneca illustrates the myriad ways in which we squander life (2f.), providing exempla of powerful men unable to retire (4, Augustus; 5, Cicero; 6, Livius Drusus). Chapters 7–92 present a wide-ranging overview of why the occupati feel as if life is not long enough: they do not value time itself and therefore they do not realize its loss. At 10.1 Seneca expresses a sort of propositio, “the lives of the occupati are the shortest of all,” but the structural importance of this has been overstated (Grimal 1959: 5f.). Seneca thereupon presents a technical presentation of the three divisions of time (10.2) and argumentation as to why the occupati cannot employ the past (10.3–6) and barely enjoy the present (10.6). Seneca resumes criticism at 11.1 of the occupati who have

2 See André 1989: 1732 for a renewed argument in favor of Albertini’s (1923: 179) transposition of 7.1–7.2 to ch. 12.

de brevitate vitae

163

wasted their lives (balanced by an exaltation of otium at 11.2), then launches a vicious attack on those who live an otium occupatum or desidiosum (ch. 12– 13). Contrary to the idle and meaningless activity of these ignorant types, the philosopher enjoys true otium and has a life that is expansive regardless of its biological length—the first sustained construction of the positive aspects of otium in the treatise (ch. 14–15; cf. 7.5, 11.2). At ch. 16–17 Seneca returns to satirical illustrations of the occupati, before at last calling on his addressee to give up his position as the praefectus annonae and return to philosophy (18–20). Topics The treatise, an exhortation to practice philosophy, is aimed at exposing the meaningless activity that passes for life and providing a suitable alternative to both his addressee and his wider readership. Taking as his point of departure the common complaint that Nature is too stingy when it comes to the length of human life, Seneca offers a defense of Nature (= Stoic deus/ratio) by attacking human ignorance: life is not too short, but we make it so because we do not know how to properly use what time we are allotted (1.3). We are misguided on two counts: 1) human life is not to be measured by its duration, and 2) bald activity is not equivalent to living. Seneca offers a corrective to these common misconceptions. Seneca’s critical portrayal of the occupati consistently focuses on the antithesis between vivere and esse, that is “really living” as opposed to “merely existing” (7.3, 7.10). The art of living is a life-long pursuit (vivere tota vita discendum est: 7.3), consisting of knowledge (scientia), which is more difficult to acquire than any other (7.3; for scire cf. 2.1, 7.4, 16.3). Most humans, because they do not practice philosophy, cannot see the truth (caligo mentis: 3.1, 13.7) and so do not use their (biological) lives to the fullest. By presenting a catalog of types who merely exist (esse), their lives pulled apart by various pointless pursuits, Seneca invites his readers (us) to conduct—from an appropriate distance—an honest audit of their (our) lives (3.2–3, 7.2, 8.1–5). Inescapably, we realize how much of our lives have passed by without real meaning. Seneca identifies the source of our error: we do not appropriately value time because it is sub-sensory (8.1–5). But time, as one of the four incorporeals, is real, according to Stoic ontology, and Seneca goes so far as to argue that it is the most valuable of all possessions (re omnium pretiosissima: 8.1). Although we go to great lengths to protect more tangible possessions, such as land or money, we obligingly allow others to occupy our lives and take time

164

r. scott smith

from us (3.1). The Wise Man, however, understanding the value of time, is most protective of it (custos eius parcissimus: 7.5). Challenging the traditional Roman notion of the vita activa, Seneca unequivocally states that “only those who free themselves for philosophy are truly otiosi” (14.1). This is true otium (14.1), not the desidiosa occupatio (12.2) or the iners negotium (12.4) of those who spend their free time on idle pursuits, still less the indolent life of the man who does not even have the self-awareness to know whether or not he is sitting (12.7f.). Stoic otium—stripped here of all negative implications, as in De otio—becomes the arena where one, free from external control and in possession of the self, can engage in meaningful introspection and create an intentional life.3 Really living (vivere), then, can only happen when one is in control of time. For the Stoics this means seizing upon the present, because only it is “available.” Seneca urges his readers to “live immediately” (protinus vive: 9.2): one cannot put off starting to live because living always involves conscious attention to the present. But unlike the Epicurean carpe diem, the Stoic notion of “live immediately” means adapting one’s life as soon as possible to purposeful living in harmony with Nature (see Dionigi 1995b). Paradoxically, by embracing the present, one also unlocks both the past and future; all of time is available to the Stoic Wise Man since his consciousness—like that of the Stoic god’s—expands to encompass the universe both spatially and temporally (15.5; see Armisen-Marchetti 1995b). One’s life, and so one’s happiness, therefore depends not on the number of years lived but on the completeness of each present moment (reflecting the Stoic idea that time cannot add to one’s happiness, which depends solely on virtue: SVF 3.49–67). For Seneca, then, the focus of one’s attention should be not on the whole of life, but only on the present day (7.9, 9.2–3). At chapter 18, Seneca leaves his theoretical treatment and turns to his addressee, Paulinus, urging him to retire into a more tranquil harbor of life (otium) where he may provide more important service (ad haec sacra et sublimia accedas: 19.1) than the mundane and all-consuming job as praefectus annonae. This epilogue, at one and the same time detached from the rest of the work and yet the crowning testimonial in Seneca’s exposition, downplays the seeming importance of the annona in favor of the more important and fulfilling philosophical life, again turning the Roman ideal of the vita

3 For otium in Roman intellectual thought see André 1962 (37–42 on Brev.) and more generally Grilli 1953, André 1966; Griffin 1976: 315–366 treats Seneca and political participation (De brev.: 317–321).

de brevitate vitae

165

activa on its head. This all-consuming post seems to have resonated with imperial Stoics; see Epikt. diatr. 1.10, where the praefectus annonae is similarly represented as a distracted occupatus. Style and Language Forceful and vibrant, De brevitate vitae is aggressive in exposing the myriad reasons why human life appears so short. Rigorous argumentation is minimal, technical language avoided; rather, Seneca’s protreptic depends primarily on the brutal but honest exposure of humanity’s misguided ways. Satirical illustrations of the occupati therefore dominate the work, creating the impression that the excoriation of human ignorance is as important as the exaltation of philosophical reflection. The positive message of the treatise thus works in tandem with these negative portrayals of the occupati by creating a stark opposition that serves to highlight the benefits of a philosophical life. Readers are not coaxed but jostled into evaluating their own lives. Seneca’s vigorous approach here may reflect the influence of Seneca’s teacher Fabianus, whose therapeutic methods relied not on subtle argumentation but on full-on frontal assault (10.1). Seneca’s hard-line strategy aims primarily at overturning, or forcing his readers to rethink, traditional beliefs. Old men with white hair and wrinkles have not lived but merely existed for a long time (7.10). Likewise, Seneca ends the treatise with a stark image: those who squander time planning elaborate funerals should rather be buried at night by “torchlight and candle” like very young children (20.5)—they have not made any progress in “really living.” Seneca criticizes the Roman institution of clientela (2.1, 2.4f., 7.7, 12.1, 14.3 f.) but later transforms it into the more positive clientela of philosophers (14.5). This aim also plays out on a smaller scale: there is a high incidence of wordplay,4 polyptoton,5 and correctio6 aimed at demolishing conventional notions. We also find sermocinatio (3.2f., 3.5, 7.6, 7.8), impatient questions (esp. ch. 13), and other features of so-called diatribe used to similar ends. Anaphora is deployed here to a degree greater than in any other work, used to

4 E.g., 6.4 vita/vitia; 7.3 percepisse/praecipere; 9.1 pendet/perdit; 12.2 otiosa vita/desidiosa occupatio; 20.1 cum videris … ne invideris … cum in consummationem dignitatis per mille indignitates erepsissent. 5 E.g., 1.1, 2.4, 3.4, 4.6, 11.1, 17.4, 17.5, 20.5. 6 E.g., 1.3 non exiguum temporis habemus, sed multum perdimus; 7.10 non ille diu vixit sed diu fuit.

166

r. scott smith

accentuate the numerous distractions of the occupati7 as well as to emphasize the inability of others to take away a philosopher’s time.8 Conversely, the terse, asyndetic language at 15.5 reflects the Wise Man’s control of time;9 compare this to the lengthy anaphoric descriptions of the distracted lives of the masses. Sources Given the non-technical character of De brevitate vitae it is difficult to assess the extent to which—if at all—Seneca was reliant on specific philosophical sources. Seneca does not treat the philosophical debate on the three kinds of life as he does in the similarly aimed De otio (ch. 7), nor does he appeal to the concept of the two republics found there (ch. 4). Stoicism, though prefiguring the debate, is not outwardly privileged; Seneca mentions only one Stoic philosopher by name, Zeno, and only in passing alongside Pythagoras, Democritus, Aristotle, and Theophrastus. When Seneca does provide technical discussions of time (e.g. 10.2–6, for which see Williams 2003 ad loc.), it is not always possible to attribute specific doctrines to earlier Stoic philosophers, who, our evidence suggests, did not extensively treat time as an ethical issue. Seneca employs numerous domestica exempla from Roman history and pictures from Roman life—ones that would be familiar to his non-Stoic addressee Paulinus. Augustus (ch. 4), Cicero (ch. 5) and Livius Drusus (ch. 6) serve as illustrations of powerful men hoping to retire from their trying public lives. In the first two cases, Seneca draws on an otherwise unknown letter (4.3, 5.2); the last instance is perhaps drawn from a history (dicitur: 6.1). In all three cases, the historical material is extensive. Later, he returns with another historical triad, perhaps drawn from handbooks (17.6): Marius, Quintius (Cincinnatus), and Scipio. Only once does he utilize a foreign figure as an exemplum, the Persian king Xerxes (17.2).

7 2.1 alius 3 × (followed by variatio: quosdam … sunt quos … multos … plerosque … quibusdam); 2.4 quam multi 4 × (with hic 2 ×, ille 2 ×); 3.2 quantum 7×; 7.7 quot ille 5×; 7.8 quando 3 ×; 14.4 quam multi 4 ×. 8 8.5 nemo/nihil/nusquam 6×; 11.2 nihil 6 ×; 14.5 nemo 3 ×; 15.1 nemo/nullius 5×. 9 transit tempus aliquod, hoc recordatione comprendit; instat, hoc utitur; venturum est, hoc praecipit.

CONSOLATIO AD POLYBIUM

Jochen Sauer

Dates and Sources The Consolatio ad Polybium was composed during Seneca’s exile on Corsica and is addressed to a freedman who was in charge of the department of petitions at the court of emperor Claudius. The formal occasion for the piece was the death of this man’s younger brother. A clear terminus post quem is the beginning of Seneca’s exile at the end of ad 41, a slightly weaker one is the award of the title pater patriae to Claudius in January ad 42 (possibly reflected in 16.4: parentem publicum; Grimal 1978a: 227). The terminus ante quem is Claudius’s Britannic triumph at the beginning of ad 44 (Giancotti 1957: 87, Abel 1967: 163, Grimal 1978a: 227, Kurth 1994: 17), since in 13.2 the hope is expressed to be present at this occasion. It is probable that the work was composed toward the end of this period (Griffin 1976: 396, Grimal 1978a: 277, Abel 1985a: 707), particularly in view of the fact that 6.2 indicates that Polybius, after his appointment by Claudius, had already been in office for some time. Since the emperor was present in Rome (12.3 f., 14.1) a date of composition between the late autumn of ad43 (return from Britain) and the beginning of the year ad 44 is not implausible (Grimal 1978a: 278, Kurth 1994: 17). Topics and Content1 The beginning of the work is lost—probably not much more than the proem (Abel 1985a: 719, Kurth 1994: 25), since the main section gives an impression of completeness. It takes the form of an admonitory address: After the praecepta (1–12) there is a transitional passage (13) and then an exempla section (14–17); an epilogue forms the conclusion (18). The praecepta section proves to be in a ring composition. The rejection of the duty of mourning (4–7) lies at 1 On the philosophical significance of the consolatory works within Seneca’s oeuvre and the character of their thought in general, cf. Setaioli (infra, pp. 241–244).

168

jochen sauer

the center, around which the idea of the necessity of transitoriness is laid as a frame (1 and 11f.). The following exempla section is developed as a ninepart, asymmetrical series containing eight positive examples for dealing with personal mourning, reproduced as Claudius’s actual words (14.2–16.3), and, in contrast, one negative example portraying Caligula’s behavior on the death of his sister (Abel 1985a: 719). While in the other two consolationes Stoic “apatheia” is recommended for the purposes of overcoming mourning (in Marcia’s case in a rather mild form, in Helvia’s in unmitigated strictness), Seneca here, using the Peripatetic “metriopathy” seems to be following a fundamentally different plan (Abel 1985a: 718; for a different view cf. Studnik 1958: 33f.): The task is not to cut out the emotions but to master them and control them. Here Seneca distances himself explicitly from the Stoic ideal of apatheia (18.5). Here, even more clearly than in the Consolatio ad Helviam, we are dealing with a consolatory piece with the character of a cryptic petition to the emperor to repeal Seneca’s banishment (Abel 1985a: 718). The panegyrical depiction of Claudius is quite remarkable. It conveys a notion of power in which the clementia of the ruler and his care for individual citizens are seen in relation to his unlimited power, in this point an anticipation of the De clementia. This account, which may very well reflect the hopes of the author, is quite compatible with the Claudian ideology of power, in which clementia and cura occupied a central position (statement Hölscher in Döpp 1994: 305). Research The question of authenticity has been solved in favor of Seneca (Isleib 1906, Stephanie 1910, Galdi 1928). Some scholars objected to the positive depiction of Claudius, feeling that it is hardly compatible with the portrait in the Apocolocyntosis. The Ad Polybium was hereby identified with a work that is mentioned by Cassius Dio (61.10.2f.). According to him, Seneca sent a work to Messalina and to Claudius’s freedmen from Corsica, containing their praises, but later destroyed it due to his feelings of shame about the flattering tone of the piece. Another solution was offered by Momigliano (1932: 75, with n. 119 f.), who considered the panegyric to have a satirical character (further developed by Atkinson 1985: 872–879). However, the explanation that Seneca is here describing Claudius in the mirror of his own expectations (Döpp 1994: 302) seems to carry more conviction, particularly in view of the fact that Claudius had just begun to rule when this work was composed. The panegyrical tone is demanded by this genre in praise of the monarch. In

consolatio ad polybium

169

this sense the consolatio should be read and interpreted with reference to Claudius, just as Cicero’s Pro Marcello is understood with reference to Caesar (Grimal 1978a: 99). Language and Style The double intention of the work leads to the combination of two genres, the literature of consolation and the panegyric. Traditional arguments of consolation (Abel 1967: 70–96, Johann 1968: 85–88, 150–155, Ceccarini 1973: 12–16) can be found side by side with elements of the praise of the monarch, which are placed here to serve Seneca’s own interests (Döpp 1994: 295). The praise of the monarch is intensified by contrasting him to Caligula’s reign of terror. The picture of the emperor, both in content and style, is quite compatible with Claudius’s personal concept of himself; Claudius’s speech (14.2, 16.3) reproduces the emperor’s favorite stylistic and rhetorical expressions.2 Features that will later be used to satirical effect in the Apocolocyntosis are here part of the panegyrical structure.

2 A comparison with Tacitus’s account of Claudius’s speech in favor of the ius honorum for the Gauls makes this very clear. Common features are, for example, the tendency to be pedantically accurate when defining family connections or to give very long series of scholarly exempla (Dahlmann 1936: 374 f., Griffin 1990: 482).

CONSOLATIO AD HELVIAM

Jochen Sauer

Dates and Sources The Consolatio ad Helviam is addressed to Seneca’s mother Helvia and attempts to provide her with some consolation for his exile. A definite terminus post quem is thus determined by the beginning of Seneca’s exile at the end of ad41 and the terminus ante quem, the end of his exile at the beginning of ad 49, is equally certain. The remarks in 1.2 and 2.5 suggest that Seneca has already been in exile for some time. Since there is also an allusion to the customary period of mourning of ten months (16.1) and these do not seem to be quite over yet, a probable date of composition would be the spring or summer of ad 42 (Giancotti 1957: 74f., Abel 1967: 163, Grimal 1978a: 197, Abel 1985a: 707, Kurth 1994: 16 f.); only Griffin (1976: 397f.) casts doubt upon such a precise dating, seeing in the reference to the ten-month period of mourning only the intention “to exemplify the idea of limit.” Since the literature of consolation from the period before Seneca has not survived well, it is impossible to make reliable statements about possible sources (cf. the remarks on Ad Marciam). Seneca remarks that in this work he is trying out something new, since he has found no important work in the literature of consolation where the author himself is the person to be mourned (2.1). This novelty, however, is probably only of limited relevance to the contents and form: traditional elements of the genre are everywhere in evidence. Topics, Content and Research1 The piece possesses a clear structure (Albertini 1923: 64 f., 255f., Coccia 1959: 150f.): After stating his personal motivation (1) and giving an account of Destiny’s harsh treatment of Helvia (2f.) Seneca presents the partitio: The 1 On the philosophical significance of the consolatory works within Seneca’s oeuvre and the character of their thought in general, cf. Setaioli (infra, pp. 241–244).

172

jochen sauer

first task is to make clear that Seneca himself is not to be mourned and that Helvia accordingly need not mourn for his sake (4–13); secondly, that Helvia’s own situation does not necessarily demand mourning (14–19). The main section thus adopts a familiar scheme of consolation literature, i.e., the distinction between unselfish and selfish reasons for mourning. Seneca states his concrete intentions in 4.1: His aim is not just to soften the pain, but to overcome it completely. Accordingly, the Stoic ideal ofapatheia is present in all its strictness throughout the piece—quite in contrast to the other consolationes. The reason for this is not so much an alteration in the author’s fundamental principles but rather their adaptation to the addressee: Helvia has great strength of character and has proven so; this enables the idea of invulnerability in the face of Fate to occupy a central place within the argumentation and to appear quite openly as the goal of the inner attitude (Abel 1967: 54 and 1985a: 715). At the center of Seneca’s reflections we have first of all the argument that change of location (and what else, after all, is exile?) occurs everywhere (6f.); the correct spiritual attitude that produces happiness, on the other hand, is, like nature, not dependent on location (8f.). Neither on account of poverty, writes Seneca (10–12), nor on account of the alleged disgrace (13) is exile a miserable lot. Helvia’s reason for mourning is not the loss of the protection, which he might have given her (14), but rather the loss of their personal contact. In order to come to terms with mourning, Seneca warmly recommends to Helvia, who is of strong character, the study of philosophy (15–17);2 but he also recommends keeping in touch with relatives (18f.), particularly with her sister, who could be a good example. The work concludes with the assurance that he is cheerful and deriving happiness from his studies (20). In the Ad Helviam Seneca explores many different implications and side-implications of the idea of exile. The variety of figurations of exile distinguishes this work from its ancient precedents and as well from Seneca’s other consolations (cf. Williams 2006b). The attempt to bring out the political character of the piece has taken two different forms: one of these sees the work as a cryptic petition, begging the emperor to recall Seneca from exile (Ferrill 1966: 255, Abel 1967 and 1985, André 1995). Seneca’s emphasis on his own strict ideal of integritas and his

2 With the recommendation of the study of philosophy Seneca offers Helvia “the same philosophical safehaven that sustains him on Corsica” (Williams 2006b: 168).

consolatio ad helviam

173

high admiration for conjugal harmony are seen by scholars as connected with the accusation of adultery with Iulia Livilla, which was in fact the reason for his banishment to Corsica. The credibility of Seneca’s innocence is increased not so much by the discursive argumentation as by the fact that Seneca calls as witness to his defense his own fides and auctoritas. Following this intention, Seneca has chosen an addressee of great suitability, his mother being a person to whom he is bound by a special fides. Abel more than others (1967: 47 f.) has spoken strongly in favor of this basic tenor of self-defense, while Griffin (1976: 21f.) makes less of this, pointing out that the motif of integritas appears in closely antithetical connection with the motif of luxuria, and this latter has nothing to do with the accusation leveled against Seneca; furthermore the concept of pudicitia does not really play a particularly dominant role here in comparison with other works. The other interpretation expects the reader to connect the exemplum of Marcellus (9.4–10.1) with Seneca’s own fate. Marcellus retreated voluntarily into exile after Caesar’s victory at Pharsalos. The parallel is heightened by the fact that Marcellus’s situation is portrayed from Brutus’s point of view. Grimal (1978a: 97 f.), while underlining the fundamentally apolitical nature of the genre, considers the work to be a clear protest against the “tyranny” of Claudius.

DE CLEMENTIA

Ermanno Malaspina

Content De clementia is a treatise in two books, of twenty-six and seven chapters, respectively, dedicated to Nero. The only Latin text with Plin. paneg. reserved to the imperial ideology, is, with Cic. rep., leg., off., one of the very rare systematic discussions of Roman political thought. Scholars have examined, in addition to the dating, the original structure (two or three books?) and the presumed incompleteness. Today, the analysis of the sources or the genesis of the term clementia appears to be more productive field for research (Adam 1970, Borgo 1985, Mortureux 1989: 1658–1664, Carile 1999); little studied are the relationships between De clementia and the actual policy of the Quinquennium Neronis.1 The structure of De clementia is difficult to recreate, as attested by the different results reached by scholars 2 despite the presence of a divisio, with gaps and corrupted at 1.3.1, whose three partes3 probably refer to an original project in three books (like De ira), of which we have only the pars prima and the beginning of the secunda.4 Book 1 is constructed on the honestum—utile pair, typical of the rhetorical arrangement (Rhet. Her. 3.3–7). After the proem with divisio (1.1.1–1.3.1),

1 The political agenda (consilia et exempla capessendi egregie imperii) of the speeches at the beginning of Nero’s reign (Tac. ann. 13.3f., 10f.; Suet. Nero 10) appears antithetical to De clementia, because it is founded on a renewed proposal of the Augustan paradigm of the collaboration between prince and senate, whereas this latter body plays no role in De clementia (excellent Griffin 1992: 141, contra Grimal 1991b: 119–131, Chaumartin 2005: xlix). 2 See Vallette 1930: 688–691, Giancotti 1955: 36–61, Fuhrmann 1963: 491–500, Büchner 1970: 209–212, and Mortureux 1989: 1649–1655. 3 Nunc in tres partes omnem hanc materiam dividam: prima erit †manu missionis† ***; secunda quae naturam clementiae habitumque demonstret (text Malaspina 2005a: 193, 251–256). 4 It is impossible to ascertain whether the gap at 2.7.5 derives from a mutilation in the manuscript tradition or whether it was intended by Seneca himself (Malaspina 2005a: 111).

176

ermanno malaspina

Seneca discusses clemency as ornamentum.5 In section 8.6–19.9, characterized by careful historical exempla (9, 10, 15), clemency appears as salus, because it guarantees security and it distinguishes a good king from a tyrant. Lastly, some practical cases on how to punish wrongs (20.1–24.2) are followed by a rhetorical epilogue, which is dominated by the dark-hued portrait of the tyrant (25.1–26.5). The seven chapters of Book 2 are taken up by a complex terminological disquisition, aimed at establishing the conduct of the sapiens and the differentiae verborum in the semantic field of clementia (as opposed to crudelitas, close to severitas, and distinguished from misericordia and venia.) Dates In the absence of external clues, from the text one can glean, first, that Nero is already the emperor (De clementia is subsequent to 13.10.54); second, that De clementia was composed in the early days of the empire, when Nero had raised great hopes among the public.6 A parallel between the age of Nero and that of Octavianus at the time of the civil wars (1.9.1)7 is, unfortunately, contained within a passage with controversial punctuation. The most ancient reading8 places a period after movit: hence Seneca would have composed De clementia after Nero had turned eighteen (ad 12/15/55–12/14/56) and he would have praised his innocentia after Britannicus’s poisoning (shortly before ad 2/12/55, Tac. ann. 13.15–17), a cynicism that appeared implausible to many (Schimmenti 1997: 53 n. 31). However, the murder of Britannicus, which could fall within those justified by publica utilitas (clem. 1.12.1), did not formally besmirch Nero’s innocence, which he maintained as a pillar of imperial propaganda.9

5 Clemency is human and it behooves the sovereign, caput and animus of the empire; a noble and necessary virtue, it obligates one to a “noble servitude” (3.2–8.5). 6 The search for more precise chronological clues yields no persuasive results (Schimmenti 2001: 57–68 and Chaumartin 2005: xliv–lii). 7 It is intentionally without punctuation marks: divus Augustus fuit mitis princeps si quis illum a principatu suo aestimare incipiat in communi quidem rei publicae gladium movit cum hoc aetatis esset quod tu nunc es duodevicensimum egressus annum iam pugiones in sinum amicorum absconderat iam insidiis M. Antonii consulis latus petierat iam fuerat collega proscriptionis. 8 From Janus Gruter and Iustus Lipsius to Momigliano 1969: 250 and Griffin 1992: 133–136, 407–411. 9 Faider 1929, Lana 1955: 225, Griffin 1992: 134–136.

de clementia

177

To this punctuation is preferred, out of stylistic and compositional considerations,10 the proposal advanced by Calvinus to place a period between nunc es and Duodevicensimum. But this way the passage from “at your present age” to “At eighteen years just completed” can be interpreted both as an explicative reprise11 and as a chronological step on the before-after axis.12 The dating, therefore fixed in the two-year time frame ad12/15/54–12/14/56 ad, does not solve the problem of relative chronology with respect to the murder of Britannicus. Topics and Sources Seneca sets as a substrate of De clementia the utterly Roman virtue of clementia, first as a Republican, then Augustan, and finally Imperial justification for Rome’s domain; on it, he grafts the Hellenistic themes 13 that distinguish Book 1. With Book 2, less catchy and more speculative, Seneca ambitiously intends to give full citizenship to the Roman clementia of the sovereign in Stoicism, which was hostile to compassionate behavior. De clementia therefore arises not from encomiastic intents, but from the aspiration to offer a theoretical justification of principality (Griffin 1992: 139, 141), outlining the condition of an individual who, possessing absolute power, exercises it while spontaneously limiting himself by effect of a single virtue, which stands above the other virtues, which are inferior or ancillary to it. The decision to identify this extraordinary virtue in clementia14 derived from Cicero’s discussion of Caesar’s misericordia et liberalitas and from the then-established presence of clementia among the virtutes imperatoriae (Konstan 2005, Malaspina 2005b, Braund 2009: 27–38). However, it was difficult for Seneca to find grounds for his overestimation of clementia in terms of philosophical tradition, since in Greek sources praótes, epieíkeia and philanthropía are not given a predominance over the other virtues.15 Thus,

Discussion in Malaspina 2005a: 292–301. Hoc aetatis = Duodevicensimum egressus annum: “all’ età che tu hai ora brandì la spada. (Infatti), uscito dai diciott’anni (= all’età che tu hai ora), già […]” (Malaspina 2005a: 297), with the traditional dating ad 12/15/55–12/14/56. 12 “A diciotto anni (= l’ età che tu hai ora) brandì la spada. Uscito dai diciott’anni già […]” (Capocci 1954: 66), with dating ad12/15/54–12/14/55. 13 Fürstenspiegel: Delatte 1942, Adam 1970, Hadot 1970, and Bertelli 2002. 14 Clementia establishes “un patto di reciproca tolleranza o benevolenza fra il re ed i sudditi” (Lana 1955: 214), assured by Nero’s singular innate goodness. 15 This position is reserved for dikaios´ yne or sophros´yne not only in the historical system of 11

178

ermanno malaspina

moving from the practical-political to the philosophical-moral context of Book 2, Seneca abandons the prince’s (historical and political) uniqueness, making his case fall within the (moral) one of the sapiens and diluting the asserted imperial extraordinariness of clementia in the more ample concepts of juridical aequitas and/or of humanitas.16 This strong discordance between Books 1 and 2 may, however, also betray differences in chronology between the two independent writings: “the result of an incomplete synthesis between an address to Nero and the draft version of a technical treatise analysing the virtue of clementia”.17 Language and Style The style is less characterized than that commonly understood to be Senecan, as confirmed by the reduced number of citations in the main discussions of the author’s style, lexicon, and grammar (Bourgery 1922a: 206–305, Setaioli 1980–1981, Traina 1987). The syntax, while far from Ciceronian concinnitas, is, however, less reduced to sententiae; use of diatribe figures (statement and reply with a fictitious interlocutor, rhetorical question, polemic dialectic) is reduced, because the presence of a real interlocutor, like Nero (1.8.1, 2.2.2), reduces the opportunities for a vivacious dialectic confrontation; the fondness for antitheses, anaphors, repetitions and variations, while undeniable, only rarely provides the pages of De clementia with the epigrammatic and conceptual verve that is typical of Seneca’s other works. In short, it is a more restrained prose, with a language more typical of “predicazione” than of “interiorità,”18 perhaps because of the debt to the Hellenistic Fürstenspiegel, the lack of a final revision, or the alleged

the four cardinal virtues, but even in Perì basileías treatises (Ten Veldhuys 1935, Adam 1970, Griffin 1992: 144 n. 3, 166 n. 4, and Braund 2009: 17–19). 16 2.7.3: clementia liberum arbitrium habet, non sub formula, sed ex aequo et bono iudicat, et absolvere illi licet et, quanti vult, taxare litem. Nihil ex his facit tamquam iusto minus fecerit, sed tamquam id quod constituit iustissimum sit. Fuhrmann 1963: 503, Adam 1970: 39, 49, and Griffin 1992: 159–171 deem the juridical argument to be preeminent here, as a reference to the attenuating circumstance of the crime in view of a superior ideal of iustitia, connected with aequitas. According to Büchner 1970 and Bellincioni 1984a, to be clement consists instead of subordinating and sacrificing formal compliance with the law to a moral notion, which approaches humanitas and the amor mutuus feeling of epist. 95.52. 17 “… le résultat d’une synthèse, qui n’ a pas été menée jusqu’à son terme, entre un discours à Néron et l’ ébauche d’ un ouvrage technique, une analyse de la vertu de clementia” (Grimal 1991b: 121; see also Vallette 1930). 18 Traina 1987.

de clementia

179

rhetorical nature of Book 1 (see supra, p. 178). In any case, De clementia, too, is subjected to the rhythmic clause rules.19 Transmission The entire manuscript tradition of De clementia derives from Vatican City, Pal. lat. 1547, known as Nazarianus (N) and written in pre-Carolinian minuscule script in Northern Italy around the year 800. Next to this manuscript, which was passed to Lorsch and then to Heidelberg until 1623, must be placed in terms of antiquity the direct apograph Reg. lat. 1529 or Reginensis (R), which is slightly more recent and which gave rise in France to the recentior tradition.20 N and R are also the ancestors of De beneficiis, a treatise whose circulation was parallel to that of De clementia until the invention of the printing press. The sole topic of discussion remains the possibility of a direct filiation of N, without passing through R, during its stay in Lorsch: this filiation is presupposed by Mazzoli 1978 for De beneficiis, but the transmission of De clementia probably did not follow the same path (Malaspina 2001b). Reception After Seneca, the notion of clementia returns to the Augustan limits of the virtue among virtues: Plinius avoids the approach of De clementia (paneg. 3.4, 35.1, 80.1) and in subsequent panegyrics clementia is less frequently found than words like pietas and maiestas; the term has little weight in coinage and in the juridical vocabulary. Rare are also the allusions, aside from Octavia 440–592, in which Seneca, speaking with Nero, puts forth arguments patently deduced from De clementia21 In the twelfth century De clementia, often reduced to an anthology, began circulating again, almost always together with De beneficiis: the first traces are found in France (Hildebertus Cenomanensis, maybe Hugo of Flavigny, Alanus de Insulis, Vincentius Bellovacensis and Guillelmus de Conchis, see Mazzoli 1978: 92–97). The text was also read during Humanism and the Renaissance,

19 Hijmans 1991 (65% of clauses are constituted by cr+sp, cr+cr and tr+tr, with the related solutions); Malaspina 2005a: passim. 20 The identified descripti (over 250) are subsequent to the eleventh century: Buck 1908, Mazzoli 1978, Malaspina 2005a: 11–140. 21 Subsequently Flavius Merobaudes (9.13, 9.19 Vollmer), Sidonius Apollinaris (carm. 9.230), Martin of Braga: see Préchac 1925: xliii–lxxii, Manuwald 2002.

180

ermanno malaspina

being worthy in 1532 of the attention of young Calvinus (Battles and Hugo 1969);22 however, De clementia never had the publishing success of the epistles or of De brevitate vitae. In a place apart is the narration of Cinna in 1.9, which impressed Montaigne (Essais 1.23) and was the basis for Cinna ou la clémence d’Auguste by Corneille (1640), far more than of Clemenza di Tito by Metastasio (1734, see Questa 1998: 191–203). Interest in De clementia weakened with the decline of absolutism and the rise of constitutional states; significantly, the concept of Roman clementia is invoked by Steven Spielberg’s Oskar Schindler when he tries, unsuccessfully, to tame the brutal Amon Göth (Schindler’s List 1993).23

See also Prinz 1973: 421, Arend 2003a. “Power is when we have every justification to kill, and we don’t. […] That’s what the Emperor said. A man steals something, he’s brought in before the Emperor, he throws himself down on the ground. He begs for his life, he knows he’s going to die. And the Emperor […] pardons him [= clem. 1.5.4]. This worthless man, he lets him go. […] That’s power, Amon. That is power.” 23

NATURALES QUAESTIONES1

Gareth D. Williams Although Seneca embarked on the Natural Questions relatively late in life, in the early 60s ad (cf. senex: 3 pr. 1), his allusion at 6.4.2 to a youthful work De motu terrarum2 signals a long-standing interest in natural science. Of the eight surviving books of Natural Questions, six are complete, the end of 4a and the beginning of 4b lost. It is conceivable that Seneca planned or completed additional books that would either fill out his treatment of meteorological themes by covering the Milky Way, say, or the sea (both figure in the first three books of Aristotle’s Meteorologica, which largely set the agenda for later ancient treatments of the subject), or extend into non-meteorological areas such as astronomy; but there is no evidence to deny that the eight surviving books were conceived as a complete and self-contained whole. The title Naturales quaestiones,3 first attested in a ninth-century inventory of the library of the Benedictine monastery at Reichenau and subsequently corroborated in the extant manuscript tradition, corresponds to the Greek φυσικὰ προβλήµατα or φυσικὰ ζητήµατα (cf. SVF III p. 205.6–13 for φυσικὰ ζητήµατα attributed to Chrysippus). That the title is unremarkable enough is also indicated by the various parallels for the phrase naturales quaestiones that are to be found in Seneca himself and elsewhere (epist. 88.24; cf. Cic. part. 64, Vitr. 1.1.7). In contrast to the typical question-and-answer format of the Greek problémata and zetémata, however, with no necessary relation between successive questions, the Natural Questions is characterized by a more continuous flow of argument, and also by complex contortions (often through the intrusion of an interlocutory voice) in the substance of Seneca’s argument (further, Hine 1981: 28f.). In this respect Seneca’s choice of title might be said to raise expectations that are surpassed in the body of the work. In addressing the same Lucilius who is the recipient of his De providentia, Epistulae morales, and (probably) the lost Libri moralis philosophiae (cf. Vottero 1989: 21–24, 1998: 75), Seneca exploits a continuity of person to suggest 1 2 3

The standard text is now Hine 1996. Fr. 5 Haase = T55 Vottero 1998: 166 f. (with pp. 31–33). Hine 1981: 24–29 (with important attention to 6.17.3), Vottero 1989: 19f.

182

gareth d. williams

a continuation of philosophical(/therapeutic) treatment and purpose. The work is conventionally dated to between 62 and 64 ad—a dating that relies on detailed (and disputed) points of chronological interpretation within the text,4 but significantly also on Seneca’s self-presentation in the Natural Questions as a public figure no more, a senex who, in the preface to Book 3, belatedly devotes himself to a life of philosophical contemplation. The retirement enacted there may be partly symbolic in its emphasis on total philosophical absorption (cf. esp. 3 pr. 2) and in its radical rejection of the more ordinary negotia of life, as if challenging his audience with a vision of “true” liberation before he insistently repeats the mantra-like question, “What is important?” (quid est praecipuum?) in his interrogation of the unexamined life later in the preface (§§11–16). But this philosophical withdrawal also finds a convenient biographical reverberation in Seneca’s increasing estrangement from the Neronian court in and after 62ad (cf. Tac. ann. 14.53–56, 15.45.3)— a prolific period in which his output, including the Epistulae morales and the completion of De beneficiis, arguably challenges Cicero’s remarkable philosophical industry in 45–44 bc (Hine 2006: 54). Yet why delay until the preface of Book 3 this seemingly programmatic announcement of a new turning in life? An answer that has gained important ground in recent scholarship is that Book 3 was in fact the first in the original ordering of the books of the Natural Questions. But if so (and advocates of alternative orderings remain),5 how did Book 3 become displaced? The manuscript tradition presents the books in two main orderings:6 1–4a, 4b–7, known as the Quantum order after the first word of 1 pr. 1; and 4b–7, 1–4a, known as the Grandinem order after the first word of (what remains of) 4b. On the basis of these divergent orderings, Gercke (1907: v–xlii) divided the MSS into two groups, Φ (Grandinem) and ∆ (Quantum); but he undervalued the independence of a twelfth-century MS, Geneva lat. 77, known as Ζ, which was tentatively proposed by Vottero (1973: 264–267) as a third branch in addition to Φ and ∆. It has since been convincingly demonstrated by Hine,7 however, that despite their different orderings of the books, Φ and ∆ derive from a common hyparchetype in a bipartite stemma, with the other hyparchetype represented by Ζ. The latter has the Grandinem ordering: from this agreement

See Vottero 1989: 20 f., and now Hine 2006: 68–72, with Gauly 2004: 19–28. E.g., Gross 1989: 306–320, with a convenient summary of different scholarly proposals on pp. 310 f. 6 On book order, see Hine 1981: 2–23, 1983: 376 f., 1996: xxii–xxv; Codoñer 1989: 1784–1795; Vottero 1989: 109–113; Parroni 2002: xlvii–l; Gauly 2004: 53–67. 7 Hine 1978, 1979, 1980, 1981: 2 f., 1983: 376 f. 5

naturales quaestiones

183

between the two hyparchetypes, the Grandinem ordering can be traced back to the common archetype. Internal considerations, including crossreferences between books (for which Hine 1981: 6–16), themselves indicate that that the original order was 3 (I), 4a (II), 4b (III), 5 (IV), 6 (V), 7 (VI), 1 (VII), 2 (VIII). A plausible explanation (Hine 1981: 17, 1983: 377) for the disruption of that sequence to produce the Grandinem ordering is that a codex was broken late in Book 4a—a rupture that resulted in the permanent loss of a section spanning the end of 4a and the beginning of 4b. The two parts were subsequently reconstituted in the wrong order, with Books 3 and 4a following after Book 2 and renumbered as Books IX and X; hence the numeration in the archetype from three to ten, and hence the movement of Book 3 from pole position to penultimate placement. This reconstruction of the original order allows a loose thematic pattern to be drawn across the different books, the four elements providing its substratum.8 Book 3, on the waters of the earth, and 4a, on the Nile, form an initial grouping; then 4b, on clouds, rain, hail, and snow; 5, on winds; and 6, on earthquakes, treat phenomena consisting of or caused by air; and fire loosely connects 7, on comets, 1, on lights in the sky, and 2, on lightning and thunder. Another elemental design can be contrived out of the Grandinem ordering, to the effect of Vottero’s grouping (1989: 112f.) of 4b and 5 (air), 6 (earth), 7, 1 and 2 (fire), and 3 and 4a (water). But if we persist with an original order of 3, 4a, 4b, 5, 6, 7, 1, 2, the distribution of the elements also creates a secondary design that is vertical in structure, beginning at water level and then ascending in 4b to the intermediate region of sublimia (cf. inter caelum terrasque: 2.1.2). True, this upward progress is not always consistent, with the winds of Book 5 anticipating the parallel action of air underground to explain earthquakes in Book 6; and after Seneca rises in Book 7 from “lower” to “higher” (and ultimately celestial) levels of cometary explanation that culminate in his speculation on comets circling in unknown orbits, he descends in Books 1 and 2 to phenomena at the intermediate, atmospheric level. In terms of the broad elemental arrangement across the books, however, from aqua to aer and then to ignis, the work takes on a suggestive symbolic property in distancing us ever further from ground level: as Codoñer (1989: 1800) puts it, “à mesure qu’ils s’éloignent de la terre [les éléments] acquièrent une plus grande transcendance.”

8 On the arrangement of material, see Waiblinger 1977 (related but contrasted pairs of books in the order 1–7), Hine 1981: 29–34, Codoñer 1989: 1799–803, Parroni 2002: xlix–l, Gauly 2004: 69 f.

184

gareth d. williams

An elaboration of this elevating vision loosely relates Seneca’s tripartite division of the universe at the opening of Book 2 into lowly terrena, intermediate sublimia, and lofty caelestia to different levels of cognition—the literal, the more abstract, the purely conceptual—a scheme that centers the Natural Questions on the atmospheric/meteorological zone of sublimia.9 Late in Book 1 that spectacular sexual deviant, Hostius Quadra, is pictured relishing the sight of his every bodily exploit in a chamber of mirrors that reflects not just his actions but also his vile inner character (1.16).10 If Hostius here symbolizes an obsessive form of “terrestrial” vision that stands in contrast to the heightened awareness of the “celestial” philosopher, and if the former’s hall of mirrors serves as a claustrophobic antithesis to the latter’s free-ranging habitation of the universal whole, the region of sublimia represents an intermediary place of contingency and accident, of always changeable conditions and ephemeral phenomena (e.g., lightning, hail, shifting winds, sudden earthquakes, etc.); in that region the naked eye strains to see clearly (cf. “Nothing is more deceiving than our eyesight”: 1.3.9) as we hover with Seneca between observation and conjecture, “reality” and illusion, sight and insight. In the imaginative world of Senecan science, our efforts adequately to grasp and explain these sublimia raise us from a level of “terrestrial” cognition to a more elevated plane of inference and speculation, while the permanent movements of the heavenly bodies at the celestial level (comets among them, at least on Seneca’s preference for an orbital/planetary interpretation of them in Book 7) symbolize a region of regularity and epistemological certainty—in strong contrast to the provisionality (as if a loose approximation to dóxa of a Platonic kind) that characterizes the atmospheric region.11 On this approach, the Natural Questions may be concerned not only, or even primarily, with the “true” causes of the phenomena Seneca investigates, but also with the hierarchy of different forms of world perception, which range from the lowly and “terrestrial” to the liberated heights of the “view from above.”12 It is this idiosyncratic Senecan stamp, this sense of a coherent artistic vision of the physical world, that makes the Natural Questions so much more than a useful if largely unoriginal rendering in Latin of theories mostly extrapolated from the Greek scientific tradition. The extent of Seneca’s direct

For this approach, see Williams 2005a: 147. On Quadra, see Leitão 1998, Bartsch 2000: 82–87, Berno 2002 and 2003: 31–63. 11 Cf. Inwood (2002: 125) on Seneca’s concern in the Natural Questions with “the epistemic limitations of human nature.” 12 On this “perennial motif in ancient philosophic writing,” see Rutherford 1989: 155–161, Hadot 1995: 238–250. 10

naturales quaestiones

185

acquaintance with many of the authorities on whom he draws is unclear; even in the case of a work such as Aristotle’s Meteorologica, which he seems to render almost verbatim in places, it remains uncertain if he knew it at first hand or relied instead on an intermediary source such as Posidonius or on doxographical compendia.13 Few scholars would now accept, however, especially after the important interventions of Setaioli (1988: 375–452) and Gross (1989), that Seneca depended on a single dominant source such as Posidonius or Asclepiodotus, or on a single compendium source.14 A versatile reliance on a medley of sources better explains internal contradictions and unevenness across the work, and is arguably more in keeping with the freeranging doxographical approach that he habitually applies, weighing now this (sequence of) opinion(s), now that, before eventually asserting his own view. Whether contending against a given source who “speaks” viva voce (e.g., Diogenes of Apollonia at 4a.2.28 f., Artemidorus of Parium at 7.13.1–3), or in brief or extended debate with interlocutory voices that stubbornly follow their own committed agenda (e.g., 1.5–8, 7.24–27), Seneca brings his doxographical mode to life by injecting personality and drama into the proceedings15—an idiosyncratic approach enhanced by at least two other distinctive features of Senecan doxography. First, in certain passages where Seneca may at first appear to be at his most scrupulous in cataloguing earlier theories (e.g., 6.5–26, on earthquakes explained by reference to each of the four elements in turn) or, more broadly, in surveying different types of a given phenomenon (e.g., wind types at 5.7–13), artistic considerations can be seen not just to influence but also to dictate the arrangement of his scientific material. So in Book 5 his coverage of different wind types progresses from pre-dawn breezes to winds of a sturdier but still consistent seasonal character until cloudbursts and whirlwinds finally erupt in the Senecan text, shattering that initial pattern of consistency. The careful ordering of the winds here, rising from calm to wild, offers a suggestive paradigm for “normative” and then transgressive human conduct as pictured later in the book: on this approach the pure science of anemology is subordinate to Seneca’s artistic orchestration of the winds for a symbolic, and ultimately moral, purpose in Book 5 (further, Williams 2005b). In Book 6 his coverage of received theories of earthquakes

For more on Seneca’s (in)direct(?) use of Aristotle, see Hall 1977: 410–416, Parroni 2002:

xxvi. 14 15

xxxv.

Concisely on the source problem, Vottero 1989: 24–39, Parroni 2002: xxii–xxvi. More generally on “Il linguaggio ‘drammatico’ di Seneca scienziato,” Parroni 2002: xxvi–

186

gareth d. williams

progresses from explanations based on simple visual analogy to a more inferential level of conjecture until, at a purely abstract level of speculation, we are guided solely by the mind’s eye. In moving from a visual, literal level of engagement with the world toward a more detached and speculative mode of inquiry, Seneca here suggestively experiments with a variation on his broader purpose in the Natural Questions of transporting us from a local to a cosmic, from a “terrestrial” to a more abstract, level of awareness (further, Williams 2006a). From a purely philosophical perspective this different awareness— what might be termed cosmic consciousness—may represent progress in a Senecan/Stoic direction. From a contemporary political angle, however, it arguably offers a form of psychological protection from the vagaries of life under a Nero by cultivating a form of detachment from the here-and-now, or by shifting the primary focus of our identity from immersion in a localized Roman context to identification with the cosmic whole. Secondly, in collecting the theories of so many enquirers over the ages, ranging from the Presocratics down to his own times and extending from the Greeks to the Egyptians (cf. 3.14.2, 7.3.2f.) and Chaldaeans (cf. 7.4.1, 28.1) to the Romans (e.g., Caecina, 2.39.1, 49.1, 56.1; Papirius Fabianus, 3.27.4; Varro, 5.16.3), Seneca constructs “a virtual community of scholars” (Hine 2006: 59). In picturing their collective contribution across time to what amounts in Book 7 to a Senecan concept of gradual scientific discovery and intellectual progress (cf. esp. 7.25.3–5, 29.3–30.6), he obliquely inscribes himself into this “virtual academy” (Hine 2006: 58), which may also be tangentially influenced by the familiar Stoic idea of a transcendent community of the wise (Hine 2006: 59). In a post-Ciceronian context, however, this Senecan construct may also be designed to assert a new confidence in the status of Roman philosophy relative to its Greek past. If Cicero took the pioneering step of consolidating a Roman philosophical vocabulary and medium separate from the Greek tradition that was their source, Seneca came of age in a Roman generation for whom philosophy was no longer essentially Greek or necessarily practiced in Greek. He was engaged in “primary philosophy (rather than exegetical or missionary work) in Latin,” thinking and writing creatively as “a rare example of first-order Latin philosophy” (Inwood 1995: 68, 75; my emphasis). Despite Cicero’s efforts in his philosophical dialogues to portray a cultured Roman familiarity with Greek philosophical ideas, that picture was in part idealized (Hine 2006: 58f.). The majority of sources on which Seneca draws in the Natural Questions continues to be Greek, but a strong Roman presence asserts itself by appealing to influential sources such as Varro, by “naturalizing” Greek material and terminology (e.g., “also the Eurus [wind-name] has already been granted citizenship and does not come into our speech as if it were a

naturales quaestiones

187

foreign word”: 5.16.4), and, more generally, by rationalizing the natural world through Roman technological, legal, and political language and metaphor (Hine 2006: 50–53, 54–56, Vottero 1989: 52f.). In this respect Seneca follows an aggressive agenda as part of a broader movement in the first century ad toward Roman cultural/scientific “ownership” of the world—a tendency that finds an alternative approach in the elder Pliny’s Roman appropriation of nature through serial cataloguing in his Natural History. But whereas Pliny constructs an encyclopedic vision of the world item by item, list by list, the Senecan mindset that brings the world to order in the Natural Questions relies on a pre-conceived, “whole” (Stoic) vision of the sympathetic correlation of the universal parts. From this viewpoint, the different phenomena that he explores from one book to the next are based on variations of elemental action, with the elements themselves significantly cast as interchangeable in the introductory Book 3 (cf. fiunt omnia ex omnibus: 3.10.1). If the oneness of reality is from the earliest moment, before we can perceive the fragmentation, lost at the ordinary level of experience through the dividing of time, the compartmentalization of different parts of existence (e.g., work, leisure; childhood, adolescence, adulthood), the partitioning of history, the habitual classifications that systematically categorize life, on offer in the Natural Questions is an imaginative vision of the original oneness—a vision that is acted out in the seamless incorporation into the work of different but (in Seneca) related meteorological phenomena, in the combination of scientific exegesis and “literary” elaboration (extending to the easy intermingling of prosaic exposition and poetic quotation), and, perhaps above all, in the fusion of ethics and physics. A central controversy in modern assessments of the Natural Questions is how, if at all, to reconcile Seneca’s main scientific agenda with the moralizing emphasis imported in prefaces (Books 1, 3, 4a), epilogues (cf. 1.17, 2.59, 4b.13, 5.18, 6.32, 7.31f.), and seeming digressions within the text (cf. 1.16, 3.17f., 5.15).16 If few critics would now endorse a polarizing approach to the problem, to the effect that the scientific or the moralizing emphasis is subordinate to and fully detachable from the other, more sympathetic responses that reconcile the two planes in a unified work nevertheless differ significantly in emphasis and strategy. To take but two recent examples, Gauly (2004: 73–85) enterprisingly applies the Bakhtinian notion of “Dialogizität” to set the (Roman) moralizing sections in productive dialogue with the main (Greek) doxographical content; while

57.

Conveniently on the history of the problem, Codoñer 1989: 1803–1808, Scott 1999, esp. 55–

188

gareth d. williams

Berno (2003) sharply analyzes the detailed network of intratextual linkages, operative within a larger scheme of oppositional pairings in the different books (e.g., visual reality and illusion in 1, natura/luxuria [and more] in 4b), that integrate the digressions within their respective contexts or whole books. While differences of method and outlook may separate interpretations that nevertheless share this integrating approach, the fact remains that the ethical and physical branches of philosophy are for Seneca closely intertwined: the study of nature elevates us above the vicious impurities (sordida) of life, fortifying the soul from corruption in its bodily prison (cf. 1 pr. 11, 3 pr. 18). Hence the sympathetic correlation drawn between contemplation of nature and the self at, e.g., dial. 12 (= cons. Helv.).20.1 (Seneca on the benefits of his Corsican exile): “[…] my soul, free of all preoccupation, […] now amuses itself with lighter studies, and, pressing eagerly after truth, now rises to the contemplation of its own nature and the nature of the universe (suam universique naturam).” More generally, given that to the Stoic imagination the three parts of Hellenistic philosophy, physics, ethics, and logic, are mutually informing and involving (cf. Hadot 1998: 77–79), the Natural Questions would in a way be incomplete without a significant moralizing emphasis. The study of physics inevitably implicates ethics, not least because the rational functioning of the physical world that Seneca charts in the Natural Questions establishes a paradigm of normative behavior that is overthrown by the human excesses that he features in his moral excursuses. As in the case of Senecan tragedy (albeit there with different dramatic expression), the order of nature is here permanently in tension with disorderly human nature; the scientific portions of the Natural Questions themselves construct a vision of rational nature that has significant implications for moral interpretation of her as always benign, even when humankind is destroyed by a disaster as indiscriminate and total as the cataclysm that overwhelms the end of Book 3 (§§27–30).17 Relatively few traces of the Natural Questions are found in later Classical antiquity.18 In his De bello civili Lucan, Seneca’s nephew, shows a direct familiarity with the work (Stok 2000: 350 and n. 6); and there are clear verbal overlaps with the pseudo-Virgilian Aetna, but the exact chronological relationship between the two remains uncertain.19 It is possible that the

17 Cf. Long 1985: 25 for the Stoic world conflagration as part of “a rational and beneficent plan for the good of the whole” (my emphasis). 18 In general on the work’s sopravvivenza, see Waiblinger 1977: 1–8, Vottero 1989: 54–69, Stok 2000, Parroni 2002: xxxv–xl, Trovato 2005. 19 On the whole question, see De Vivo 1989 (adjudging the Aetna the later work).

naturales quaestiones

189

elder Pliny drew on the Natural Questions in his Natural History, especially in Book 2, on cosmology (Vottero 1989: 56), but he makes no direct mention of the work. It is not until the late fourth century that fresh echoes are heard in Ammianus Marcellinus, albeit possibly via an intermediary (Stok 2000: 351; Parroni 2002: xxxvi–xxxvii). Apart from traces in the Etymologies of Isidore (Ross 1974: 130), the Natural Questions disappears from view in the Latin West until the twelfth century, while in the Greek East it resurfaces briefly in the sixth century, in the fourth book of John the Lydian’s Perì menôn/On months. Locating the Nile’s annual flood in July, John offers a doxographical survey of its causation that is derived, very possibly via an intermediary, from Seneca’s treatment of the theme in nat. 4a (Vottero 1989: 59 f., Parroni 2002: xxxvii). This survey is of special interest because of its coverage of sources (4.107, pp. 146.3–147.6 Wünsch) on which Seneca apparently drew in the portion of 4a now lost to us. While manuscripts can be traced to Reichenau and elsewhere in the ninth century, there is evidence to locate the rediscovery of the Natural Questions in northern France by the early twelfth century (Hine 1983: 377). The first writer known to draw upon it is William of Conches (ca. 1090–1154) in his Philosophia mundi and Dragmaticon, and other allusions are found in the twelfth century before its wider diffusion in northern Europe in the thirteenth century, when it was used by such savants as Robert Grosseteste (ca. 1175–1253), Vincent of Beauvais (ca. 1190–1264), Roger Bacon (ca. 1214– 1294), and Albertus Magnus (between 1193 and 1206–1280).20 By the end of the thirteenth century the Natural Questions reached Italy, where copies of it quickly proliferated. With the rise of Renaissance humanism in the fourteenth century, however, Seneca’s moral writings drew greater scholarly interest than the Natural Questions; it is significant that the editio princeps of his philosophical works (Naples 1475) predates by fifteen years that of the Natural Questions in the 1490 Venice edition of his opera. Already late in the twelfth century the diffusion in the West of Aristotle’s Meteorologica in a Latin version rendered from an Arabic translation by Gerard of Cremona (Haskins 1924: 14f., Lindberg 1978: 65f.) provided a new foundation for meteorological study. Despite the proliferation of editions in the sixteenth to eighteenth centuries, this Aristotelian presence, and the advances made in technical instruments and observation, contributed to the gradual marginalization of what, from a scientific standpoint, came to be viewed as “una raccolta di ‘curiosità’” (Vottero 1989: 64). This reputation has contributed to a relative

On these and other influential figures, see Nothdurft 1963: 161–181.

190

gareth d. williams

neglect in modern scholarship, albeit with a welcome resurgence in the later twentieth century. If in the nineteenth and twentieth centuries scholarship on the Natural Questions was centered primarily on matters of text, the ordering of the books, and source identification (cf. Waiblinger 1977: 5), recent scholarship continues to grapple with the problem of reconciling its scientific and moralizing portions. Or to state the position differently: just how to define and articulate the “true” nature and meaning of Senecan (literary) science remains an important (even the main) object of debate in contemporary scholarship.

EPISTULAE MORALES*

Aldo Setaioli The Epistulae morales ad Lucilium, “the highest and most mature” of Seneca’s works,1 were composed near the end of his life. One of the earliest letters, which makes it clear that he is writing after his retreat from public life,2 must probably be dated in the spring of ad62.3 References to historical events that can be securely dated are extremely rare; only one letter can be safely connected with a precise date, namely epist. 91, which mentions the fire of Lugdunum (Lyon), which took place near the end of the summer of 64.4 For the rest, we must be content with a few scattered hints at the month or the season: epist. 18 is written in December, during the Saturnalia;5 epist. 23 during a cold spring following upon a mild winter;6 epist. 67 toward the end of a similarly cold spring;7 epist. 86 toward the end of June.8 Consequently, if the spring mentioned in epist. 23 is the same as that in epist. 67, it must be the spring of ad 64, and the December of epist. 18 must be placed in ad63. This is the so-called “short” chronology. Conversely, if epistles 23 and 67 refer to the two successive springs of 63 and 64, epistle 18 obviously refers to December of ad62—the so-called “long” chronology. Although several scholars have attempted calculations based on more or less plausible rhythms of exchange of correspondence between Seneca and Lucilius,9 there is too much we do not know to do so in any credible way. However, despite several authoritative *

Submitted for publication in 2007. Abel 1985a: 745: “dasjenige Werk, in dem er sein Reifstes und Höchstes gab, die ‘Epistulae morales’.” 2 Epist. 8.1: in hoc me recondidi et fores clusi, ut prodesse plurimis possem […] 2: secessi non tantum ab hominibus, sed a rebus, et in primis a meis rebus. 3 Tac. ann. 14.52–56. 4 Cf. Tac. ann. 16.13.3. 5 Epist. 18.1: December est mensis […] Saturnalia. 6 Epist. 23.1: hiemps […] et remissa fuit et brevis, […] malignum ver […], praeposterum frigus. 7 Epist. 67.1: ver aperire se coepit, sed iam inclinatum in aestatem, quo tempore calere debebat, intepuit nec adhuc illi fides est; saepe enim in hiemem revolvitur. 8 Epist. 86.16: Iunius mensis est quo tibi scribo, iam proclivis in Iulium. 9 Binder 1905 offers such calculations in support of the “short” chronology; more persuasively Grimal 1991b: 219–233, 443–456 uses them to support the “long” chronology. But, as Mazzoli (1989b: 1853) rightly remarks, “i ritmi elaborati da Grimal hanno il merito di essere realizzabili, piuttosto che reali.” 1

192

aldo setaioli

supporters of the “short” chronology,10 it is difficult to imagine that during just a part of the spring of ad64 Seneca had the time to send at least11 45 letters to Lucilius; therefore, epistles 23 and 67 must refer to two different springs, although both cooler than usual: those of 63 and 64. The “long” chronology appears to be all but necessary12 to account for any chronological frame in which the letters may plausibly fit.13 According to Grimal’s formulation, the Epistulae morales ad Lucilium are a sort of “diary”14 that Seneca kept in the latter part of his life. We shall see how this might be the case after a brief survey of the collection’s content. First of all, it must be said that the Epistulae are meant as spiritual direction and as a guide to ethical education (including self-education) and moral progress, engaging both the addressee and the writer himself. This involves an initial stage that aims to win over the addressee to Seneca’s educational program, followed by a further stage consisting of the actual teaching of the moral tenets of Stoicism—although the previous stage (the admonitio) will never be completely superseded. As we shall see, these two stages entail different linguistic and stylistic approaches, and are roughly reflected in the two parts of the collection, the second (epist. 89–124) turning increasingly to theoretical questions rather than moral paraenesis. The general theme of the letters is that virtue—or moral good—is the only good, and vice the only evil; what is commonly regarded as “good” and “evil” is in reality “indifferent” (indifferens, adiaphoron). So, for example, death is no evil; in fact, suicide is the guarantee of the wise man’s freedom. All must engage in the attainment of “right reason” (recta ratio, orthos logos); this entails the free and willing acceptance of the cosmic order: fate is perfectly equivalent to providence. If we are so disposed, nothing can prevent us from attaining virtue. Human will is crucial in this regard.15

10 These include such scholars as Abel, Griffin, and others. See the clear review in Mazzoli 1989b: 1851 f. 11 We must in fact admit the (likely) possibility that not all the letters written by Seneca to Lucilius found their way into the collection: cf. Grimal 1991b: 443. 12 Grimal 1991b: 222 n. 553 speaks of “la quasi-nécessité de la chronologie longue”; cf. Cugusi 1983: 197, Mazzoli 1989b: 1853. 13 As I say this I am, of course, assuming that the letters do record a real correspondence between Seneca and Lucilius. See below. 14 Grimal 1991b: 219: “les Lettres à Lucilius nous donnent une sorte de journal du philosophe.” According to Schönegg 1999, Seneca is an artist portraying himself in the Epistulae; but this interpretation is based on symbolisms arbitrarily “discovered” in the letters. Schönegg takes it for granted that the correspondence is fictional. 15 Cf., e.g., Voelke 1973: 161f., Abel 1985a: 749, Mazzoli 1989b: 1874f. According to Inwood (2005a: 132–156), Seneca failed to discover will as a separate moral faculty.

epistulae morales

193

One theme, which is particularly developed in the letters, is Seneca’s reflection on time, whose fleeting transience must be conquered by appropriating it as an ideal present, snatched from contingency. When one has reached moral perfection, one moment does not differ from eternity.16 Space does not permit us to go into further detail;17 we shall only remark that the theme is posed from the very first epistle, and point out that there is a further dimension of Senecan time that has not received the attention it deserves: the subjective conception of memory, which might remind one of Proust.18 The first problem the student of the Epistulae must address is whether the collection reflects a real correspondence between Seneca and Lucilius or whether the epistolary form is just a literary fiction. Both positions have been defended by authoritative scholars,19 but the burden of proof rests of course with those who deny that Seneca’s letters are what they purport to be. The letters are clearly arranged in chronological order, as even those who consider them to be fictional must admit,20 and several contain lively descriptions of details of daily life. Taking both these characteristics as a literary device aimed at giving “a pleasing depth to the illusion of epistolarity”21 and,

16 E.g., epist. 93.8: quaeris quod sit amplissimum vitae spatium? usque ad sapientiam vivere; qui ad illam pervenit attigit non longissimum finem, sed maximum; 73.13: Iuppiter quo antecedit virum bonum? diutius bonus est: sapiens nihilo se minoris existimat quod virtutes eius spatio breviore cluduntur. Cf. also epist. 53.11, dial. 1 (= prov.).1.5. See Setaioli 1988: 94–96. 17 We may refer to the bibliography collected in Viparelli 2000: 183–188, to which add Lévy 2003. 18 I am thinking especially of epist. 49.1–4. Cf. Viparelli 2000: 41, 49, 54–59. 19 See the clear review offered by Mazzoli 1989b: 1846–1850. Supporters of the fictional nature of the correspondence (first asserted by Lipsius) include Hilgenfeld, Bourgery, Cancik, Maurach, Griffin, and Abel. After 1989, this position was defended by Graver (1996: 10–32), who however must conclude (p. 29) that “no fictive correspondence in prose had been attempted on anything like the scale of the Epistulae Morales”—which hardly supports her thesis. Seneca’s letters are decidedly a novum, but in a different sense. See below. Cf. also Hachmann 1996: 393. For Schönegg 1999, see supra (n. 14). The correspondence is real for Schultess, Binder, Albertini, Delatte, Lana, Cugusi, and Mazzoli himself. 20 E.g., Bourgery 1911: 41. In addition to the usual arguments (seesupra, on epist. 18, 23, 67, and 86), I would like to point out that the chronological order is confirmed by the internal references found in the collection. Ep 20.13 quotes epist. 18.5; epist. 33.1 calls epist. 1–29: priores epistulae; epist. 75.9 quotes epist. 71.4; epist. 76.20 quotes epist. 74.21; epist. 8.1 harks back to epist. 7.1, as well as epist. 57.1 to epist. 53. In some cases, the reference is to a later letter (epist. 45.13 announces epist. 48 and 49. Cf. also epist. 94.52 ~ epist. 95.1 and epist. 36.11 ~ 71.12f.). But this hardly implies a rigorous preconceived plan; promises are not always fulfilled: a discussion on the problem of free will is promised (nat. 2.38.3, epist. 16.6), but never appears in Seneca’s writings. 21 See Graver 1996: 16; but it is a common line of argument with the supporters of fictionality; cf., e.g., Bourgery 1911: 46, 52. The figure of the addressee, Lucilius, has been interpreted as no more than a “fictional interlocutor” by the supporters of fictionality, whereas

194

aldo setaioli

in the final analysis, at deceiving the reader, clearly amounts to begging the question. The common point of the line of argument of many supporters of the fictionality of the letters is that they tend to take for granted the very point to be demonstrated.22 What does appear from the collection is that the Epistulae were really written to Lucilius but were also meant as “open letters” to be published and made available to a wider public, 23 including posterity—which Seneca considered his ultimate addressee.24 If we deny their authenticity, we give up the opportunity to understand their specific literary and philosophical import, which we shall now try to elucidate. As Margaret Graver rightly remarks,25 an important reason for resorting to the epistolary form was “its potential for creating a strong authorial presence.”26 But an equally important ground was that it implied a constant mutual relationship with an addressee and therefore it especially suited Seneca’s peculiar way of thinking and expressing his thought, constantly fluctuating between the inner self and the outside world.27 As aptly remarked by Foucault,28 whereas in his letters Cicero “recounts himself” as acting or deliberating in the outside world, what Seneca describes to Lucilius is his own relationship to himself—and not merely through his philosophical meditations, but also through the ordinary everyday events that prompt them. We shall soon see the importance of these “frame effects.”29

the opposite view has been defended by some of their opponents. Cf. Mazzoli 1989b: 1853–1855. In my opinion, no conclusion can be drawn from this argument either way: apart from a few biographical facts, the only Lucilius we know is the portrait sketched by Seneca—he is his “creation” (meum opus es: epist. 34.1) in more ways than one. 22 Abel (1985a: 745) unambiguously states that the fictionality of the letters is not generally recognized only because Seneca was a clever forger (“wenn die Wahrheit [!] sich so schwer hat durchsetzen können, dann darf man darin vornehmlich einen Triumph der Senecanischen Darstellungsweise erblicken, der es gelungen ist, dem vorgetäuschten Schein das Aussehen echtesten Seins zu geben”). See also Abel 1981a. Bourgery (1911: 51) goes as far as to state that the real letters of Epicurus prompted Seneca to write his fictive ones to Lucilius. 23 Cf., e.g., Cugusi 1983: 200 f. 24 Epist. 8.2, 21.5, 22.2, 64.7; cf. 79.17. Interestingly, Lana (1991a: 270 f.) points out that Seneca’s interest in posterity appears only after his retirement. 25 Graver 1996: 30. 26 The letter is in fact an “image of one’s own soul.” Cf. Demetr. eloc. 227 and Graver 1996: 30. 27 Cf. the felicitous formulation of Traina 1974: 41: “linguaggio dell’interiorità […] linguaggio della predicazione.” 28 Foucault 1983: 16–18. 29 According to the felicitous definition of Mazzoli 1991. As rightly observed by Rosati 1981: 9, the connection of the letter with everyday reality makes it an ideal genre for the daily practice of philosophy.

epistulae morales

195

But the letter has other advantages, too. Seneca was obviously familiar with Artemon’s definition of the letter as a “halved dialogue” reported by Demetrius:30 if living together is best,31 the letter is the next best thing.32 This was a traditional view; but Seneca proceeds beyond it: communication by letter is actually better than being physically together, in as much as it is not impaired by occupations and the neglect induced by the very nearness and availability of the friend.33 And, most of all, the very impossibility of offering ready advice makes the letter the ideal vehicle for imparting universal moral instruction valid for everyone, including posterity, as well as for the individual addressee.34 This brings us to Seneca’s conception of the letter (his letter) and its place in the epistolary writing of antiquity. A number of predecessors and possible models for Seneca’s letters have been proposed,35 but only four have been considered to be most relevant: the letters of Plato and Epicurus on the Greek side, those of Cicero and Horace in Latin. Contacts with Plato and Horace have been emphasized especially by supporters of the fictionality of the letters;36 but Seneca never mentions either one in this connection, while making it clear that the models he has in mind are the letters of Epicurus and those of Cicero to Atticus.37 As far as the latter are concerned, Seneca’s conception of his own letters makes it clear that what he meant to achieve was something quite different from Cicero’s letters to Atticus.38 It would be hasty, however, to view these merely as Seneca’s “anti-model.”39 We should not forget that a key element in Seneca’s collection is taken from Cicero, namely the single addressee,40

30 31 32

Demetr. eloc. 223. Epist. 6.5 f. Epist. 67.2: tecum esse mihi videor […] quasi conloquar tecum. Cf. 75.1. Thraede 1970:

65f. 33

Epist. 55.8–11. Epist. 22.1 f.; cf. 64.7. 35 Cf., e.g., Graver 1996: 27–29. 36 E.g., Maurach 1970: 183 f., 188–190 for Plato; Cancik 1967: 54–58, Maurach 1970: 196f., and Graver 1996: 12 for Horace. 37 Epist. 21.3–5. 38 Even before we find this explicitly stated by Seneca (epist. 118.1f.). 39 Cf. the remarks by Thraede 1970: 65–88. 40 Seneca specifically mentions Cicero’s letters to Atticus, not the ad familiares, nor any other letters of his, for that matter. As Cugusi (1983: 200) remarks, there were other published letters of Seneca’s to another addressee (cf. Mart. 7.45.3f.)—which proves that a collection with a single addressee follows a definite literary pattern. According to Cugusi (1983: 203), the planning of the letters as replies to Lucilius and the intimate tone of the correspondence are also derived from Cicero. 34

196

aldo setaioli

whereas Epicurus’s letters have several, as Seneca well knew.41 We should also bear in mind Seneca’s attitude of aemulatio toward Cicero, which can be traced in his writings from very early on, at both the literary and philosophical level.42 True, Cicero’s letters to Atticus were not meant for publication like Seneca’s, but under the empire they were obviously considered a literary classic,43 to the point that Fronto could write: epistulis Ciceronis nihil est perfectius.44 When Seneca states that he will be able to grant immortality to Lucilius just as Cicero did with Atticus and Epicurus with Idomeneus,45 and later adds that his letters will contain useful moral teachings rather than idle gossip,46 his challenge to Cicero involves both literary glory (i.e., stylistic form) and content—whose philosophic worth will grant him primacy in the Latin epistolary genre. As for Epicurus, we can be sure that Seneca knew at least some of his letters,47 although he drew many of his quotations from anthologies. They were a suitable model for letters of ethical instruction, especially in view of Epicurus’s attitude to his pupils: the same atmosphere of familiarity is indeed found in the letters to Lucilius—although Seneca never presents himself as an accomplished and infallible master, but always as a seeker of truth trying to progress toward virtue just like his pupil. Though Seneca, by referring both to Cicero and Epicurus, makes it clear that he intends his work to belong in the epistolary genre, regardless of content and approach, it is equally clear that he is quite consciously attempting something new in the literary panorama.48 This is made apparent by the very title of the collection, which is at least as old as Gellius,49 and possibly goes back to Seneca himself: Epistulae morales. According to

Cf. Setaioli 1988: 171–182. Cf. Setaioli 2003: 61–75. 43 Cicero’s letters to Atticus had surely been published for a long time when Seneca wrote his to Lucilius. Cf. Setaioli 1976, refuting contrary views. There are reliable signs of Seneca’s awareness of the letters to Atticus from the time of his exile (dial. 12 [= cons. Helv.].1.2). Cf. infra, p. 241 n. 27, and Setaioli 2003: 63 f. 44 Fronto, II, p. 158 (Haines). Epist. 21.4 proves that Seneca viewed Cicero’s letters to Atticus as literature. Cf. Thraede 1970: 67. 45 Epist. 21.3–5. 46 Epist. 118.1 f. 47 Cf. Setaioli 1988: 171–182. 48 Cf. von Albrecht 2004: 2. Rosati (1981: 3) rightly emphasizes Seneca’s play with the usual formulas of greeting to stress the novelty of his letter writing (e.g., epist. 15.1); cf. Mazzoli 1991: 74 n. 15, Spina 1999: 18, 22–28, who also points out “metaliterary” beginnings and endings in several letters. 49 Gell. 12.2.3. 42

epistulae morales

197

Lana, it is the first literary collection of letters in Latin prose,50 meant to be published and made homogeneous by the common purpose to impart useful ethical teaching.51 But whereas Lana sees little difference between Seneca’s letters (at least the longest ones)52 and his treatises, Mazzoli rightly points out that opposing Seneca’s “epistles” to the lively “letters” of Cicero or anybody writing his friends about the events of his life fails to do justice to Seneca: a so-called “epistle” can be just as real and lively as a “letter.”53 Mazzoli finds the specific character of Seneca’s letters in the often outwardly irrelevant details of daily life, which prompt the writer’s ethical reflection in several cases.54 Through a shrewd analysis of several such “frames,” he points out the links connecting the details of everyday life with the philosophical developments making up the main part of each letter— a feature distinguishing Seneca’s Epistulae from his treatises. Actually, as Mazzoli remarks, the tight unity of “frame” and philosophical reflection is explicitly theorized by Seneca.55 Often these factual “frames” are marked by humor, especially self-irony. This aspect has been well illustrated by Armisen-Marchetti.56 The humorous anecdotes or self-portraits reveal to Seneca as well as to the reader his miserable physical and/or spiritual conditions, of which he had previously not been fully aware, and help diminish the distance between master and

50 In verse there are, of course, Horace’s Epistulae. Lana (1991a) devotes several pages (258– 268) to Seneca’s letters in relation to other epistolary collections; the other standard treatment is Cugusi 1983: 196–206. 51 Lana 1991a: 268, 271. 52 Seneca clearly distinguishes between a letter and a book (epist. 45.13, 85.1, 89.17) and often hints at excessive length (e.g., epist. 30.18, 47.21, 51.13, 52.15, 86.21, 108.39). Only in one case (epist. 108.39) do we find such a remark in the second part of the collection, where the average length of the letters increases (though he does call epist. 95 an ingens epistula: 95.3). The length of the letters ranges from 2 paragraphs (epist. 38) to 74 (epist. 94). Cf. Lana 1991a: 292–295 (epist. 62 has 149 words and 17 lines; epist. 94 has 4,164 words and 503 lines). However, in spite of the “objective” indication of epist. 45.13 (non debet sinistram manum legentis implere), the length of the letter is largely a matter of subjective judgment; there are short letters, like epist. 38, 62, and 122, and immensely long ones, like epist. 94 and 95. 53 Mazzoli 1989b: 1856f., polemicizing against Cancik and Thraede and quoting epist. 40.1. 54 Mazzoli (1991: 73–75) classifies all the letters that have come down to us in relation to the presence and type of such a “frame.” At least 25 contain what Foucault terms “écriture de soi.” A slightly revised list is given by Spina 1999: 21f. 55 Epist. 55.3: ex consuetudine tamen mea circumspicere coepi an aliquid illic invenirem quod mihi posset bono esse. Cf. Mazzoli 1991: 82. 56 Armisen-Marchetti 2004. Earlier, Grant (2000) had especially stressed the theatrical and comic elements found in these Senecan descriptions.

198

aldo setaioli

pupil,57 although the former, as already mentioned, never pretends to have reached or even approached moral perfection. The letter is by nature “unsystematic” and allows the treatment of single problems detached from a wider doctrinal context, 58 but not all of Seneca’s letters are limited to a single theme.59 However, we must leave aside the investigation of the structural problems posed by the individual letters and restrict ourselves to the collection as a whole. What should never be forgotten is that we do not possess the complete collection. In addition to the twenty books that have come down to us60 at least two more existed, as testified by Gellius, who quotes from Book XXII.61 We must refer to Mazzoli’s clear overview of the several divisions and articulations that have been proposed.62 All scholars agree that the first three books (epist. 1–29) form a compact unit marked as such by Seneca himself,63 and that in the second part of the collection (epist. 89–124) the average length of the letters increases, although shorter ones are interspersed here and there. Hildegard Cancik64 is right when she stresses the interaction of structural references, which form a close-knit network linking the letters to one another and unifying the collection. This unity, however, is not the result of a compositional plan previously elaborated and later developed in a collection of fictional letters; rather, it is the spiritual direction common to all the epistles that causes the basic ideas connected with it to appear and reappear when needed to foster the accomplishment of the spiritual director’s task. This is what really confers unity on the collection. As Mazzoli 57 Armisen-Marchetti (2004: 322) remarks that this is probably not a strategy consciously pursued by Seneca. 58 Seneca himself opposes the moral letter to his organic treatment of ethics (themoralis philosophiae libri): cf. epist. 106.1f., 108.1. In epist. 81.3 he offers a more in-depth treatment of a problem already discussed in the De beneficiis. Cf. Rosati 1981: 11f. 59 Abel 1985a: 750 distinguishes between “monothematic” and “polythematic” epistles. 60 The letters are transmitted by a double tradition. A first group includes epist. 1–88 (Books I–XIII), a second epist. 89–124 (Books XIV–XX). These separate traditions go back at least to late antiquity. The first group is further divided (epist. 1–52: Books I–V, and 53–88: Books VI–XIII). The divisions between Books XI–XIII and XVII–XVIII are not marked in the tradition. Book XXII may have been part of a separate group that included posthumous letters. Cf. Reynolds 1965a. More recently, Jeannine Fohlen has studied the tradition of the Epistulae. We shall only mention Fohlen 2000. For more details, see Marshall, supra, p. 43. 61 Gell. 12.2.2–13. 62 Mazzoli 1989b: 1860–1863. Lana (1991a: 283f.) denies the existence of a structural plan in the collection, with the exception of the first three books. 63 Epist. 29.10: ultimam pensionem; 33.1: his quoque sicut prioribus. As Abel (1985a: 751) rightly remarks, the first passage seems to indicate that the book division goes back to Seneca himself. 64 Cancik 1967: 6.

epistulae morales

199

rightly remarks,65 it is a “work in progress”—at the ethical as well as at the literary level. The ethical object of the Epistulae ad Lucilium is matched by Seneca’s peculiar linguistic and stylistic resources. The scope of this article does not allow us to dwell on his stylistic praxis,66 but we should at least mention Seneca’s theoretic discussion of the style appropriate to philosophical works like his own, and note that his theory finds practical application in his own writings, as Michael von Albrecht has rightly remarked.67 According to Seneca, the philosopher must be able to master both a style that appeals to the emotions, in order to convince the pupil to undertake his moral reformation, and a plainer one, to be used later for actual instruction. In his discussion of style, Seneca reaches some surprisingly “modern” conclusions: every writer has his own personal style and establishes his own individual rules, and his relationship to the models is no passive process of reproducing their style, but rather amounts to a cultural formation pre-eminently aiming to elaborate the contents by bringing one’s own, original contribution. In Seneca we witness the fruitful encounter of the innovative rhetoric of the first century ad and his own philosophical background.68 We have already seen how important Epicurus’s letters were as a model for Seneca’s Epistulae morales. Epicurus is indeed the most frequently quoted philosopher, although his presence declines after the first three books, in which most letters end with a “quotable quote” borrowed from him.69 As Seneca makes clear, these fulfill a propaedeutic function;70 even more important, although he must be given credit for rejecting the widespread disparagement of Epicurus, he is only interested in some of the latter’s ethical ideas, totally detached from their philosophical context.71 65

Mazzoli 1989b: 1863. We must refer to von Albrecht’s contribution, infra, pp. 699–744; but we should at least mention Traina’s epoch-making work (Traina 1974: cf. supra, n. 27). Cf. also Mazzoli 1989b: 1863–1868. For Seneca’s peculiar use of the sermo cotidianus see Setaioli 1980–1981, now collected and updated in Setaioli 2000: 9–95, 393–397. 67 Von Albrecht 2000b: 228 f., 245 f. 68 See Setaioli 1985, now collected and updated in Setaioli 2000: 111–217, 397–408. The critics who see a contradiction between Seneca’s stylistic theory and his praxis (e.g., Rozelaar 1976: 345–404) disregard not merely the need for a style that appeals to the emotions in the first stage of spiritual direction, but—more important—the internal agreement of Seneca’s theory and praxis at a deeper level: Seneca’s style is the reflection of his own personality. 69 For Seneca and Epicurus, see Setaioli 1988: 171–248. For Seneca’s knowledge of Metrodorus, see Setaioli 1988: 249–256. 70 Epist. 33.1 f. 71 Setaioli 1988: 171, Mazzoli 1989b: 1872. Significantly, Seneca considers such ideas to belong to general common sense (eiusmodi vocibus referta sunt carmina, refertae historiae. itaque nolo 66

200

aldo setaioli

As there is no “Epicurean Seneca,” so there is no “Platonic” one, although such a view has been repeatedly defended.72 Seneca was surely sensitive to Platonism,73 but did not subscribe to it. He surely knew at least some of Plato’s writings directly, but probably drew many “Platonic” ideas not from the master himself, but from Middle Platonism.74 There are also scattered allusions to other philosophical “sources,”75 but there can be no doubt that the Letters’ basic philosophy is Stoicism76—with, of course, the peculiarities and personal innovations we can expect from Seneca’s own statements of autonomy.77 For lack of space I must refer to my book on Seneca and the Greeks for Seneca’s complex attitude to the several masters of Stoicism.78 The influence of Seneca in general and the Epistulae in particular was remarkably powerful in his own time and has remained so to this day— although it has naturally fluctuated over the centuries. Space does not allow us to even begin sketching a summary picture,79 and we must limit ourselves to a few bibliographical recommendations for further discussion.80

illas Epicuri existimes esse: publicae sunt: epist. 33.2). For the question of Lucilius’s possible leanings toward Epicureanism see infra, p. 245 n 48. Mazzoli 1989b: 1872f. rightly points out that Epicurus was the most readily available philosophical support for Seneca’s retirement from public life. 72 This has been done most notably by Donini 1979, 1982: 181–210. Donini has been refuted by Timpanaro 1979, Setaioli 1988: 505–510, Mazzoli 1989b: 1870f. More recently, Seneca’s “Platonism” has been asserted again by Schönegg 1999; see the convincing refutation by Armisen-Marchetti 2002. 73 We should not forget that his philosophical apprenticeship was under the standard of “Pythagoreanism” (epist. 108.17–22). 74 Cf. Setaioli 1988: 141–164. 75 For this, see Setaioli 1988. 76 Cf. also the sensible remarks on the influence of Cynicism and diatribe made by Mazzoli 1989b: 1873 f.; see also Setaioli 1988: 165–170. 77 E.g., epist. 33.7–11, 80.1; and the whole of epist. 84; cf. also dial. 7 (= vit. beat.).3.2. An example of this are Seneca’s ideas about the divine: see Setaioli 2006–2007. Others are probably his reflections on time and literary style (cf. supra). 78 Setaioli 1988: 257–365; cf. also Mazzoli 1989b: 1874–1877. 79 We must refer to Laarmann’s contribution, supra, pp. 53–71. 80 Blüher 1969, Trillitzsch 1971, Mastandrea 1988, Dionigi (ed.) 1999, Martina (ed.) 2000, Citti and Neri 2001. Many papers on Seneca’s Fortleben may also be found in Giornale Italiano di Filologia 52 (2000) and Aevum (ant.) 13 (2000).

DE BENEFICIIS

Mario Lentano

Dates and Sources De beneficiis is an extensive treatise in seven books about benefits and how to bestow them, about the right way of returning them, and how to deal with ingratitude. It is undoubtedly a work belonging to Seneca’s full maturity: the terminus post quem is ad56, whereas the first six books were already composed before the spring or summer of ad 64 (Préchac 1926: xv); more precise dating is controversial (ad57–58, Herrmann 1937: 99f.; ad59–60, Friedrich 1914 and Grimal 1978a: 459; ad 60–61, Chaumartin 1985: 194; ad 59– 61, Chaumartin 1989a: 1702–1709; before ad62, Griffin 1992: 399; Nero’s first years, Veyne 1999: 48; between ad 62 and the first months of ad 65, Lo Moro 1976 and Letta 1997–1998) and it depends on interpreting the allusions to Nero either as an attempt to go on guiding the princeps’s policy, or as a recognition of the final tyrannical drift of his power. It is possible to distinguish three sections in the treatise (Fowler 1886): Books 1–4 deal with the doctrine of benefits; Books 5–6 propose a wide and detailed case record, further developed in Book 7, which possibly was added later. It is not possible to come to definitive conclusions on the sources of De beneficiis (see Chaumartin 1985: 21–154, 1989a, 1989b). Hecaton, Panaetius’s disciple, who is mentioned four times in the treatise and who is the author of a Perì kathékontos, has an importance that is difficult to evaluate, maybe the prevailing one. It has been debated whether Seneca drew material from this treatise, even if very freely, or from a Perì cháritos or Perì charíton (not attested). The influence of other Stoic sources (Chrysippus, Cleanthes, Posidonius) is unprovable and seems marginal; the Cynic Demetrius, who is mentioned in Book 7, was known directly to Seneca. The influence of the declamation is more extensive than is thought. It can be observed above all in Book 3: here Seneca writes about the opportunity for a law on ingratitude, provided for by the existing school laws (with a possible allusion to the contemporary debate on liberti ingrati, see Manning 1986); the distinction between beneficium and officium in 3.18 probably dates

202

mario lentano

back to Hecaton, but its most likely counterpart lies in a text by Seneca the Elder (contr. 2.5.13), which deals with an actio ingrati; even the debate on the possibility for a son to bestow a benefit on his father echoes themes already widespread in the schools of rhetoric (Lentano 1999). Topics and Content Seneca’s fundamental theoretical move consists in distinguishing between the act of giving (beneficium) and the content of the service (materia beneficii): the latter may be returned or not, but the fact remains that the benefit possesses an intrinsic moral value and is in its deepest nature a res that animo geritur (1.5.2). The same distinction applies to gratia, which has become a matter of conscience in Seneca: the one who receives a benefit must never forget his debt (1.4.5), but the return of it may take different forms, from gestures of gratitude to the words used to acknowledge one’s debt (reddit enim beneficium qui debet: 1.1.3), up to the open will of returning the benefit: at the end of this series there is the Stoic paradox according to which qui libenter accepit beneficium reddidisse (2.31.1). Around this essential kernel, which runs through the treatise, there are other questions: how and to whom to bestow a benefit, how to choose the beneficiary, what kind of benefit to bestow, whom to accept a benefit from, and what sort of manifestations to associate with it; giving without expecting a return is emphasized, showing the gods as models, who bestow benefits on ingrates, too, and give benefits beyond any possibility of returning them. A varied case record is added to the illustration of the general principles: whether to bestow a promised benefit on an ingrate, whether to return a benefit in any case, whether it is possible to bestow a benefit on oneself, whether one is under an obligation to someone who bestowed a benefit unwittingly, and so on. Research Chaumartin (1989c: 1580f.) sums up the studies about De beneficiis from 1945 to 1985 in just one page; even the major monographs (Marchesi 1934, Sørensen 1976, Grimal 1978a, Maurach 1991, Griffin 1992, and Veyne 1999) devote just a few pages to the work. Lack of method, too much space given to case record, repetitiveness, and lack of organization are the most repeated charges and they partly explain the limited attention that scholars have devoted to the work.

de beneficiis

203

A turning point in scholarship was the masterly monograph by Chaumartin (1985). Far from being just an exercitatio ingenii aiming at oblectamentum (Fuhrmann 1997: 289f.), De beneficiis seems an “œuvre de combat” to Chaumartin (1985: 261), which aims at “restaurer la confiance dans les rapports humains” in an atmosphere dominated by the fear and uncertainty caused by the imperial regime. Among the relationships to be reformed there is that between master and slave—a theme that is dear to Seneca—and above all that between clients and patrons (including the emperor), where the good use of benefits can create true bonds of gratitude and friendship between the parties. The aim of the treatise is therefore both moral—a deep reform of the relationships between superiors and inferiors—and political— a warning to the princeps not to transform an autocratic regime into the worst tyranny. Other less convincing interpretations prefer the moral purpose of the treatise, holding that De beneficiis as a whole “constitue une doctrine de bonté” (Préchac 1926: xxxv) or that a sort of “société nouvelle” is prefigured in it, where friendship based on beneficium will replace existing relationships, which have been spoiled by injustice (Grimal 1978a: 181–183 and 305f.; see also Grimal 1976: 176f. and Veyne 1999: 193f.). In recent years a sociopolitical reading of De beneficiis has been privileged. Seneca is interested above all in the relationships within the élite: he underscores that the advent of the Principate introduces a strong innovation in the practice of benefits. It is expected that the princeps behaves as any other member of the élite and that the élite itself involves him in the social code developed in the Republican age. In theory, the princeps was a primus inter pares and all the parts involved were concerned in keeping this theory. The language of the beneficium helps create an appearance of equality between the partners in the exchange, an appearance that gains a new and even greater importance in the new context of the Principate (Griffin 2000, 2002, 2003a, 2003b). Seneca aims at adapting the model of the exchange of beneficia to the new reality of the Empire, where the “power of giving” is concentrated in the princeps’s hands. Separating the notion of gratia from that of remuneratio, Seneca suggests the attitude of a new reciprocity, which allows people not to give up the code of the exchange of gift and counter gift, but which gives an up-to-date pattern to this code (Lentano 2009). In a society with strong inequalities, individuals might in fact be reluctant to accept a benefit, if they know they could never return it, and this might undermine the social bonds produced by beneficia. The emphasis placed by Seneca on the facility of gratia is liberating, then, because it promotes

204

mario lentano

an ethics of the sheer intention that in fact consolidates political and social bonds. Seneca teaches his peers to give freely even if they might come up against ingratitude; he teaches the others to accept serenely the condition of debtors of a beneficium, with which one can live with confidence and dignity (Inwood 1995). De beneficiis offers much material to investigate Seneca’s idea of the imperial regime and that of individual princes (see the analytic discussion in Chaumartin 1985: 157–206) or in general to reconstruct the relationship between ethics and power (Bellincioni 1984b: 101 f.). In the opinion of some scholars, Seneca remains faithful to the system, even if he is critical of the emperors (Mayer 1991: 162f.); others see a radical pessimism in De beneficiis as regards the imperial regime, by then viewed as a brutal submission to a dominus (Letta 1997–1998). De beneficiis revolves around a theme—the gift—that has become central in anthropology since Marcel Mauss’s studies. In addition to the general analyses of the notion of beneficium and of gratia in the Roman cultural code (see the bibliography in Lentano 2005), there are also some attempts to apply Mauss’s categories to Seneca’s treatise, sometimes with some rigidity (Griffe 1994), or to detect Seneca’s contribution to a discourse about gifts, which lies somewhere between the societies described by the anthropologists and the modern age (Goux 1996). The question of gift exchange between father and son is very interesting from an anthropological point of view (3.29f.): in Roman culture the father seems to be the benefactor par excellence (Lentano 2005); but Seneca rejects the widespread opinion that the son’s services are no more than a mere return of the benefits received, and are therefore not praiseworthy (Lentano 1999 and Marchese 2005: 29–62). The section about the relationships between slave and master is undoubtedly the most studied in the entire treatise (3.18–28). The consonance with Stoic doctrine is generally acknowledged, but with different suggestions (Richter 1958, Grimal 1976: 176f., Mantello 1979: 17–182, Giliberti 1984, Bradley 1986, Manning 1989: 1525–1529, Martini 1989, and Rist 1989: 2008f.); here again Seneca’s position shows some originality. Language and Style Of the two types of language—the “linguaggio dell’interiorità” and the “linguaggio della predicazione” (Traina 1987)—the latter prevails in De beneficiis, the style of which is less characterized by typical Senecan elements,

de beneficiis

205

such as the sententia. The treatise contributes to the specialization of key terms from the Western philosophical tradition, such as conscientia (Molenaar 1969) and persona (Bellincioni 1981 [1986]: 70–73). The display of exempla is rich, often taken from the history of Rome (Mayer 1991); the recourse to the figure of the imaginary objector is extensive. Transmission The main data on the textual tradition of De beneficiis are easily summed up by Reynolds (1983a: 363–365; a new, accurate discussion is found in Malaspina 2001a: 13f.): the oldest manuscript is the Nazarianus (N: Vaticanus Pal. Lat. 1547), probably copied in the Milan area at the beginning of the ninth century and then moved to Lorsch monastery. The Reginensis codex (R: Vaticanus Reg. Lat. 1529), dating back to the second quarter of the ninth century and probably written in Northern Italy as well, is regarded as a descendant of N (a “caso esemplare di antigrafo e apografo,” Busonero 2000a). It is still considered controversial whether the remaining tradition can be traced back to R (Reynolds 1983a; contra Mazzoli 1978: 96 f. and Brugnoli 1998: 86). There are almost three hundred recentiores and there are also shortened versions, epitomes, excerpta, and collections of sententiae, which are useful for following the late medieval and modern reception of the treatise but not very meaningful to the constitution of the text. Reception A discussion of the reception of De beneficiis can be found in the general reviews of Seneca’s Fortleben (Faider 1921: 135–151, Bourgery 1922a: 150–186, Gummere 1922, Ross 1974, Chevallier and Poignault 1991, Dionigi 1999, Citti and Neri 2001). Among the Christian writers, Minucius Felix, Tertullian, Jerome, Arnobius, and Lactantius quote De beneficiis or hint at it (Mastandrea 1988: 43f., Lo Cicero 1991: 1258–1260, and Brugnoli 2000a: 238–241). The treatise was eclipsed in the early Middle Ages: the first quotation since antiquity perhaps dates back to the third decade of the eleventh century (Mazzoli 1978: 91, Brugnoli 1998: 80, Brugnoli 2000a: 239). The twelfth century is a real aetas Senecana; the number of manuscripts rapidly increases (18 copies of De beneficiis) and the treatise is widely reflected in florilegia, epitomes, and excerpts (Munk Olsen 2000). De beneficiis “wurde im Mittelalter als

206

mario lentano

Traktat der politischen Ethik gewertet”1 (Blüher 1969: 77) and, together with De clementia, it establishes the genre of Fürstenspiegel (Nothdurft 1963: 100f., Smiraglia 2000: 275f.). Dante quotes De beneficiis just once, maybe indirectly (Conv. 1.8.16 = ben. 2.1.3, see Mezzadroli 1990: 43f. and Dionigi 1999: 120f.), Petrarch knows the treatise well (De remediis utriusque fortune 1.93). After the editio princeps of 1475, the editions by Erasmus (1515 and above all 1527–1529, including the Epistulae) and by Justus Lipsius (1605) are decisive for Seneca’s success in European culture. Montaigne’s Essais contain hundreds of quotations from Seneca; De beneficiis is represented by eight mentions (Blüher 1997: 627 n. 6). The Grand Siècle enhances Seneca’s reputation as a tragedian, above all but in the Caractères by La Bruyère (1688) De beneficiis is widely quoted. The treatise was the first of Seneca’s works to be translated into English (1578); later, Ben Jonson transposed a great deal of De beneficiis into verse in his Underwoods (1640). In the eighteenth century Diderot expressed a keen appreciation of De beneficiis (Essai sur les règnes de Claude et de Néron, 1782). Here, Seneca becomes the symbol of the French philosophes and of the difficult relationship between intellectuals and power on the very eve of the Revolution. It is very difficult to follow Senecan reception in the nineteenth and twentieth centuries: De beneficiis is quoted by both writers and philosophers, but Seneca’s current reputation is assured by his letters. In an unusual anthology for a non-specialized audience published in Germany and France, Seneca for Managers, De beneficiis has no place: a work written for the ancient prince seems to have nothing to say to the modern Caesars of economy and finance.

“In the Middle Ages, [De beneficiis] was regarded as dealing with political ethics.”

LOST AND FRAGMENTARY WORKS

Anna Maria Ferrero

Content Among Seneca’s lost works,1 there are a judicial peroration given in Caligula’s times,2 the request to Nero to be allowed to retire from political life (Grimal 1967b: 131–138), some documents written for Nero: the laudatio funebris for the death of Claudius, the prince’s first speech in the Senate, and the message to the senators about Agrippina’s death.3 We know of numerous letters sent to Caesonius Maximus4 (to be identified, perhaps, with the Maximus mentioned in epist. 87.2), of a libellus sent by Seneca from Corsica to Messalina and to Claudius’s freedmen, and of a petition sent from exile to influential persons.5 We also know that he wrote ethnographic monographs: De situ et sacris Aegyptiorum, De situ Indiae,6 and philosophical-scientific treatises: De motu terrarium7 and De forma mundi.8 We have fragments of a De matrimonio,9 where Seneca argued in favor of the natural and spiritual need for marriage against those Stoics who advised against it, illustrated the valid reasons for contracting it, and provided negative examples of infidelity and positive examples of virtue; the peroration on modesty closed the work, which also mentioned Indian women’s custom of climbing onto their husbands’ funeral pyres. We have fragments of De immatura morte,10 where the very concept of a

1 Only authentic works are examined. No consideration was given to extracts and excerpta of various kinds or of anthologies, put together by editors of late Roman and medieval times, among which we should mention De remediis fortuitorum, which, in view of its success in the Middle Ages was, wrongly, included in the editio princeps of Seneca the philosopher (Vottero 1998: 7–9). 2 Epist. 49.2. 3 Tac. ann. 13.3.1–4, 14.10.3–11, 53f. 4 Mart. 7.45.1–4. 5 Cass. Dio 61.10, Birt 1911b: 596–601. 6 Serv. Aen. 6.154, 9.30. 7 Nat. 6.4.2. 8 Cassiod. 2.6.4. 9 Hier. Iov. 1.41–49, Grossgerge 1911, Bickel 1915: 382–394, Frassinetti 1955: 186–188. 10 Lact. inst. 1.5.26, 3.12.11.

208

anna maria ferrero

premature death is denied; of De superstitione;11 of De officiis,12 on the duties of daily life; of Quomodo amicitia continenda sit, on how to make up with friends after quarreling and how to distinguish friends from flatterers (Bickel 1905b: 190–201, Studemund 1969 [1888]: 13–24, 26–32); of a moral philosophy treatise: Libri moralis philosophiae;13 and of De vita patris.14 We also have quotes taken from a protreptic to philosophy, Exhortationes,15 in which, according to a codified modulus, the premise on the need to study philosophy as the sole path to happiness was followed by adversaries’ objections and their rebuttal, an adage (Omnes odit, qui malos odit).16 We have the memory of his last words, pertaining to considerations on the inevitability of fate, on firmness when confronted with adversities, on contempt for death, some added provisions to his will and testament, and codicilli containing instructions for his own funeral.17 It is not possible to identify the two works De piscium natura and De lapidum natura, which are mentioned only by Plin. nat. 9.53.167 and in the bibliographical index of nat. 36.18 Dates The only known element is the date of the Libri moralis philosophiae, late ad64 to early 65. For the other writings, the following chronology can be established with a certain approximation: De situ et sacris Aegyptiorum and De situ Indiae (ad 17–19); De matrimonio (ad 38–39); De motu terrarum (early exile years); De forma mundi (during the exile); De officiis (ad 60); De amicitia (ad62–63); De immatura morte (ad63–64); De superstitione (summer of 64); Exhortationes (fall of 64). Topics and Sources Both as an orator and as the writer of Nero’s speeches, Seneca, according to Quint. inst. 10.125–131 and Tac. ann. 12.69–13.4, was very favorably received

Aug. civ. 6.10 f., Diom. 379.15–19 K. Diom. 336.9 K. 13 Epist. 106.1–3, 108.1.39, 109.1.14.17, Lact. inst. 1.16.10, 2.2.14f., 6.17.28. 14 Cf. Winterbottom, infra, p. 695. 15 Lact. inst. 1.5.27, 1.7.5.13, 3.15.1.11–14, 16.15, 23.14, 25.16, 5.13.20, 6.24.12.16–17, 25.3. 16 Aug. epist. 153.14. 17 Tac. ann. 15.60–65, 67.3, Treves 1970: 507–524, Gnilka 1979: 5–21, Abel 1991: 3155–3181, Fabbri 1978–1979: 409–427. 18 Vottero 1998: 89–91. 12

lost and fragmentary works

209

by the public, whereas he was the object of much criticism as the emperor’s master of eloquence (Avery 1958: 167–169, Gelzer 1970: 212–223). Nothing definite can be said of Seneca as a poet: the poemata and the carmina mentioned by Quint. inst. 10.129 and Tac. ann. 14.52.1–3 refer to the entire Senecan poetic production including the tragedies and the collection of the epigrams;19 the poetic pieces mentioned by Plin. epist. 6.3.2.5 (perhaps epigrams, in view of the Plinian context) have not survived. Of the letters sent by Seneca to Caesonius Maximus, at least some were written from Corsica, since Mart. 7.45.1–4 places the information about the correspondence in a context taken up exclusively by the theme of exile and by that of faithfulness in friendship. Sources of the monograph on geography and the rites of the Egyptians, and of the one on India, could be the studies of Posidonius and the ethnographic digressions present in Sallustian works, but it should be noted that Seneca always had a lively interest in Egypt, as attested by the book nat. 4a, whose topic is the Nile floods. While India had long been known through reports by Alexander the Great’s generals, it apparently remained, in Seneca’s opinion, a region inhabited by barbarians and the subject of naturalistic curiosity (André and Filliozat 1986). The fragments of De matrimonio are all preserved by Hier. Iov. 1.41–49, an argumentative response to the lost Commentarioli by the monk Iovinianus. Of the treatise on earthquakes we are informed by Seneca himself in nat. 6.4.2 and, since we have no other information about it, the title is traditionally derived from this context. In it, Seneca probably followed the pneumatic theory, which in the Stoic school was held by Posidonius. From Cassiod. 2.6.4 we learn the title of the work De forma mundi, typical theme of Greek cosmologies, which discussed the cosmos in general, its shape, and the various problems connected to it. The Stoics attributed a spherical shape to the cosmos and Seneca shared this opinion. On the De officiis we only have the report of the grammarian Diomedes, who, in discussing the verb praestare, after observing that it is mostly used in the sense of melius esse or antecedere or superare, mentions among the exceptions an example taken from the work in question. Treatises of this kind already exist in Zeno, Cleanthes, and Chrysippus. The issue was taken up again by Panaetius and by Posidonius; the Stoic model also inspired M. Iunius Brutus, whose treatise on duties is quoted and used by Seneca in epist. 95.45. The fragments taken from De amicitia, in spite of some generic consonance, do not reveal a specific source in any of the ancient texts on the topic known

Traina 1976: 19.

210

anna maria ferrero

to us (Panaetius, Hecaton). The friendship theme was already present in the first letters to Lucilius, then procurator in Sicily; this period seems to have an echo in the ample space occupied in the fragments by the absence of friends and faithfulness to them, as Lucilius had demonstrated in those difficult years. The problem of death, the subject of De immatura morte, is one of the most widely debated themes in Hellenistic philosophical treatises. When confronting it, Seneca’s position stands out not so much for its originality, as for the persistence, conviction, and effectiveness with which he expounds some fundamental concepts: death as a law of nature, equal for all; death as liberation; the need to despise the terror it strikes; the ease with which it can hit us at any time; the praemeditatio mortis as the philosopher’s foremost duty and essential step in his path toward wisdom; voluntary death as supreme choice and conquest of freedom. Seneca constantly repeated these thoughts, and similar ones, from his first to his last surviving works: hence, it is not groundless to hypothesize their presence in De immatura morte as well.20 The title, on which the manuscripts agree, is consistent with Senecan use. Of the treatise De superstitione we have a mention by Tert. apol. 12.6, who, discrediting the statues of pagan gods (the materials used, the manner of execution, and their cold and nearly corpse-like appearance) refers to the similar criticisms with which Seneca targeted the cult of images. The ten actual fragments are preserved by Aug. civ. 6.10f., who points out the contradictions between Seneca’s deepest convictions and his external behavior, but, since he merely draws almost exclusively on theexempla and mentions neither the inner structure of the dialog nor the arrangement of the subject matter, he deprives us of the theoretical and doctrinal part connected to it (Herrmann 1970: 389–396, Funke 1974: 149, Traina 1977 (1987): 171–192, and Lozza 1989). The grammarian Diomedes provides us with the locution versa templa, which, out of context, adds little to our knowledge of the treatise. As his certain source we can point out only Varro Reatinus (Pépin 1956: 265–294) and, in general, the tradition of the Stoic school, which pointed to superstition as one of the components of fear, viewed by the Stoics as one of the fundamental passions.21 The Exhortationes belong to the mother lode of philosophically oriented protreptic literature, which, originating from Socrates’s teachings, opposes the Sophists’ didactic practice. It is difficult to reconstruct the context of the work, because the fragments that have reached us are short, sparse, and dispersed in the various books of Lact. inst., who only quotes the passages 20 21

Lausberg 1970: 153–167. Scarpat 1983: 98, André 1983: 55–71, Mazzoli 1984: 953–1000.

lost and fragmentary works

211

functional to his argument against pagans, which perhaps explains the high number of fragments about God (Mazzoli 1977: 7–47). Protreptic features are present throughout Seneca’s work. A comparison between epist. 16.1–6 and our fragments allows us to suppose that a premise on the need for sapientiae studium would be followed by the adversaries’ topical objections on the inconsistency between the philosophers’ words and deeds and the reply by Seneca, who in conclusion must have argued that only through the philosophical ars does human free will meet and identify with divine will. Among the sources of the work we can include, for its affinity to Seneca’s themes, Cicero’s Hortensius, a lost work, but one that can be reconstructed through numerous quotes by Lactantius, Augustinus, and Nonius Marcellus.22 It is impossible to specify the influence of other protreptics, both Greek and Latin, of which we often have only the title, a few mentions, or some fragments. The content of the Libri moralis philosophiae cannot be specified from the surviving fragments, whether it pertains either to the divinity or to the sapiens and his opposite, themes that are present throughout Seneca’s works (Leeman 1953: 307–313). Of some tenability is the work’s subdivision of ethics dating back to Eudorus of Alexandria, in the first century bc, of which we also partly know a text, On the Division of Philosophical Discourse. Thus latter text apparently had some influence on subsequent tradition and above all on the Stoics, who divided ethics into theory, impulse doctrine, and action doctrine (Lieberg 1973: 63–115). Thus, we can hypothesize that Seneca’s treatise contained a theoretical part and a practical part, linked together in a unitary discourse through one, or more, mediating sections. Language and Style Nothing can be said about the fragments’ language and style because, as is well known, in ancient rhetorical practice each author made the quote his own and gave it a personal expression. Transmission In 1605 Iustus Lipsius published in Antwerp the first edition of all Seneca’s writings, including all the fragments, which he placed before the surviving writings, at the end of the introduction. This was followed by F. Haase’s

Grilli 1976.

212

anna maria ferrero

new edition (Leipzig 1853), which included the fragments in volume III, pages 418–445. There were 128 fragments, distributed under twenty different indices without any apparent logical or chronological order: poems were first, followed by scientific treatises; philosophical works were interspersed with orations and letters. The edition lacked both a critical apparatus and comments. In 1902 the fragments were reprinted as a supplement to a new edition of the entire Senecan works. In 1971, Trillitzsch collected and published the fragments, still without providing critical apparatus or commentary. D. Vottero’s 1998 edition is distinguished by its scholarly introduction, critical apparatus, Italian translation, rich commentary, and ample bibliography. Reception We know that De forma mundi was present in Cassiodorus’s library, accompanied by the advice to read it, which was given to the monks, and that De officiis inspired Martinus, Archbishop of Braga, around ad572 his Formula vitae honestae.

EPISTULAE SENECAE AD PAULUM ET PAULI AD SENECAM

Alfons Fürst The pseudepigraphic correspondence between Seneca and Paul consists of fourteen short letters, eight allegedly by the philosopher, six by the apostle. They were written by an unknown author in the second half of the fourth century before ad 392/93, when they were first mentioned by Jerome in his De viris illustribus (vir. ill. 12). The most striking peculiarity of these letters is the lack of nearly any content. Only one letter deals with a historical topic, namely the great fire of Rome and the subsequent persecution of the Christians under the reign of Nero in ad64 (epist. 11). But, in contrast to what one might expect from an exchange of letters between these famous men, the author is concerned neither with Seneca’s philosophy nor with Paul’s theology. There are only a few scanty hints at philosophical, theological, and biblical matters, and all of them demonstrate the author’s inability in this regard (Fürst 1998: 80–88). Since the author is obviously not interested in such topics, scholars have proposed divergent hypotheses to explain the intention of his pseudepigraphic writing. Barlow (1938: 89–92), the editor of the first critical edition based on twenty-five manuscripts from the ninth to the twelfth centuries, considered it an exercise of style in a school of rhetoric in Late Antiquity, because many rhetorical terms are used in all the letters. Since, in some letters, “Seneca” exhorts “Paul” to use a good style in writing his letters (epist. 7, 9, 13), Bocciolini Palagi (1985: 13–16), who twice edited a revised version of Barlow’s edition supplemented with a useful philological commentary, interpreted the text as a pleading for the use of classical language in Christian writings (see also Natali 1995: 35–40). Bickel (1959b: 95) suggested that the reason to write these letters was the fascinating idea that “Seneca” should preach the Christian gospel to the Roman emperor (epist. 3, 7, 8, 9, 14). Westerburg (1881: 30, 37) detected an anti-Paulinian intention: The author wanted to bring disrepute upon Paul in connecting him with Nero, his infamous second wife Poppaea, and his mentor and minister Seneca who, as the “teacher of a tyrant” (Cass. Dio 61.10.2; cf. Auson. grat. act. 31), was in bad repute among the pagans of Late Antiquity. On the contrary, following Harnack (1893: 765, 1904: 458f.), many scholars (for example, Bocciolini Palagi 1985: 15f. and Malherbe

214

alfons fürst

1991: 417f., 421) surmise that this exchange of letters was meant to recommend the epistles of Paul to pagan intellectuals interested in Christianity. The question as to the intention of this curious text without any content remained unsolved, however. As Malherbe (1991: 417–421) has noticed, it consists of letters of friendship in which traditional commonplaces were used extensively. Therefore, Fürst (1998: 88–94, 2006: 13–16) argued for considering its genre. In such letters of friendship, there is no need for any content. Hence, the only intention is to portray Seneca as a friend of Paul’s. The reason for this is to be seen in Seneca’s Nachleben in Late Antiquity. Among Latin Christian theologians, Seneca was highly esteemed as a philosopher whose opinions often corresponded with Christian beliefs—Seneca saepe noster, as Tertullian said (De anima 20.1). For Lactantius, Seneca was almost a Christian (cf. inst. 1.5.26, 1.5.28, 6.24.14). This is exactly the relation of Seneca to Christianity put forward in the last letter of the exchange where “Paul” writes that “Seneca” “has nearly reached the irreprehensible wisdom” (epist. 14). By means of a fictitious exchange of letters between Seneca and Paul, the author underlined the affinity between the pagan philosopher Seneca and the Latin Christian theologians of the fourth century, inventing an apostolic tradition: Seneca, while not yet a Christian, was nevertheless supposed to have been a close friend of Paul’s (Fürst 1998: 103–117, 2006: 18–21). Although the style of these letters is awkward, with many passages that are difficult to understand, the fiction that Seneca was a friend of Paul’s became highly influential. During the Middle Ages, several theologians referred to the friendship between the two men, as was testified by their alleged correspondence (testimonies are gathered by Barlow 1938: 110–112 and Fürst 2006: 68–82). From the twelfth to the fifteenth centuries, the text was transmitted in more than 300 manuscripts, in most cases together with the twelfth chapter of Jerome’s De viris illustribus at the beginning of collections of Seneca’s works. In the times of early humanism, as Momigliano (1950: 336) and Sottili (2004: 676–678) have shown, the letters, especially the last one, gave rise to the legend that Seneca was a Christian, a legend that had proponents until the nineteenth century (Walter 2006: 129–132). After Lorenzo Valla and Paulus Pompilius had impugned the correspondence’s authenticity in 1440 and 1490, respectively (Faider 1926: 116), Erasmus of Rotterdam (epist. 2092), editing Seneca’s works in 1515, demonstrated that this exchange of letters was a fake.

PART TWO

PHILOSOPHY Topics

ONTOLOGY AND EPISTEMOLOGY*

Mireille Armisen-Marchetti It is futile to seek a systematic treatment of the “science of being” in Seneca: what we do find in his work are merely scattered, more or less developed remarks appended to moral or scientific discussions, or, in the Epistles, in answer to a question asked by Lucilius—and, nearly every time this happens, Seneca apologizes for straying into the field of dialectics. What we know from other sources about the positions of Stoicism in this regard, however, permits us to view Seneca’s remarks as parts of an organic whole and to appreciate their consistency. Still, we should be very careful and proceed from Seneca to Stoicism, rather than from Stoicism to Seneca; in other words, we must not read our philosopher merely in the light of external references, but rather start from his own text, and only after try to explain it through what we know—or believe we know—about Stoic doctrine. If we do so, we will realize that Seneca is well versed in Stoicism and the way it dealt with these problems, but also that he has no qualms about taking a position when he is faced with opposing doctrinal stands, or even proposing a personal view. On the other hand, the very area of the question to be treated—ontology—poses some problems, as it does not fit any Stoic doctrinal partition. What the moderns call “ontology” covers a field that falls astride Stoic logic and physics, and the demarcation of this field is largely arbitrary. The definition of “epistemology” poses the same problem. In order not to encroach on other sections of this volume, we have thought it advisable to understand it not as the study of the methods of knowledge, but rather as the description of the domain of knowledge: a description undertaken by the Stoics, including Seneca, through the concept of the “parts of philosophy.”

* My warm thanks to my friend and colleague Aldo Setaioli for translating this chapter from French into English.

218

mireille armisen-marchetti 1. Ontology 1.1.

According to the Stoics the world and the beings it contains result from the combination of two principles, one active, the other passive:1 Sen. epist. 65.2: Dicunt, ut scis, Stoici nostri duo esse in rerum natura, ex quibus omnia fiant, causam et materiam. Materia iacet iners, res ad omnia parata, cessatura, si nemo moveat. Causa autem, id est ratio, materiam format et quocumque vult versat, ex illa varia opera producit. Our Stoics say, as you know, that there are in nature two [principles] from which all things proceed, cause and matter. Matter lies inert, ready to become anything, but idle, if not moved by anyone. As for cause, that is reason, it imparts form to matter and turns it whatever way it wishes, thus producing different things from it.

Stoici nostri: this is in fact an essential doctrine of their system. It was upheld by all Stoics, who designate these principles by the term ἀρχαί.2 In the text quoted above Seneca uses a periphrasis (duo ex quibus omnia fiant), but elsewhere he translates the Greek term by principia (epist. 93.9) or initia (epist. 65.19, 90.29). The first principle (the order has no special meaning beyond Seneca’s choice of enumeration) is matter, materia, which, further down, will be identified as the passive principle.3 Materia corresponds to οὐσία, which denotes the sum total of matter in the cosmos.4 By contrast, the matter individual things are made of is designated by the Greek Stoics with the term ὕλη, which Seneca translates either by substantia, or, again, materia.5 The materia is iners (ἄποιος), that is, devoid of qualities in and by itself; it is res 1 In addition to the SVF, Long and Sedley (1987: chap. 44) provide a review and commentary of the main Stoic texts concerning principles. For principles in Stoicism, cf., among others, Duhot 1989: 73–86, Muller 2006: 66–68. As far as Seneca is concerned, the most recent treatment of the question is Wildberger 2006: I 3–7. 2 See, among others, Diog. Laert. 7.134, who attributes it in turn to Zeno, Cleanthes, Chrysippus, Archedemus, and Posidonius. 3 Sen. epist. 65.3: materia patiens dei. Cf. SVF II 300, 301: πάσχον, πάσχειν. 4 Cf. SVF II 300. In Epistle 58.6 Seneca defines οὐσία in the following way: οὐσία, res necessaria, natura continens fundamentum omnium; he proposes to render οὐσία by essentia, referring to precedents in Cicero and Papirius Fabianus; neither of these texts has survived. This translation by essentia, however, remains isolated, and Seneca resorts more often to materia or substantia. For materia, cf. epist. 65.12, 65.19, 65.23f.; dial. 8 (= de otio).4.2, 5.6, nat. 1 pr. 3, 16, 2.2.1. 5 ῞Υλη: SVF II 300, 301. Substantia: Sen. dial. 7 (= vit. beat.).7.4, epist. 58.15, nat. 1.6.4, 1.15.6. Materia: Sen. benef. 6.2.2, dial. 1 (= prov.).5.9, dial. 6 (= cons. Marc.).26.6.

ontology and epistemology

219

ad omnia parata, undifferentiated and displaying no preference concerning the active principle’s action.6 The word res is an approximation, since matter is not, properly speaking, a “thing,” a body, inasmuch as it is unable to exist by itself cut off from the qualities imparted to it by the active principle: this is how we must understand Seneca’s last remark, cessatura, si nemo moveat. The second principle is the active one; at §12 of the same Epistle 65 it will be denoted as ratio faciens, a translation of ποιοῦν. This imparts form to matter and fashions it according to its wish (quocumque vult versat, which echoes and reverses what is said of matter, res ad omnia cessatura): this means that it bestows qualities upon matter, thus creating, as we shall see below, the differentiated bodies. Seneca designates this principle through different appellatives, which, taken together, permit us to gain a complete notion of the idea. Here, in Epistle 65, where Seneca’s purpose is to emphasize the singleness of the Stoic cause as compared with the plurality of Platonic and Aristotelian causes, it is called causa:7 by this, Seneca points out that Stoic ontology is grounded on a single causality, represented by the active principle. This principle, on the other hand, may also be viewed as the divine reason immanent in the world: therefore, still in the same letter, Seneca calls it ratio and deus, thus conforming to Stoic habit.8 Finally, when it is viewed as acting within the differentiated beings in the world, it may also be associated with natura: “what else is nature but god and divine reason mixed with and within the world as a whole as well as with and within its parts?”9 1.2. The interplay of the two, active and passive, principles gives rise to all beings, first to the elements—fire, air, water, earth10—then to the differentiated bodies resulting from the combination of the elements. The way in which the

SVF II 301 τὴν δὲ ὕλην πάσχειν τε καὶ τρέπεσθαι. Cf. also Sen. epist. 89.16. SVF II 311: “the substance of what exists, being unmoving and amorphous by itself, must be moved and receive form from some cause.” 8 SVF II 300: “the active principle is the reason acting within it (viz. matter), god.” Ratio: Sen. epist. 65.2, 65.12, benef. 4.7.1, nat. 1 pr. 16. Deus: Sen. epist. 65.23, 58.27, benef. 4.7.1, nat. 1 pr. 16, dial. 8 (= de otio).4.2. Cf. SVF II 301, 311, 312, 313. 9 Sen. benef. 4.7.1: quid enim aliud est natura quam deus et divina ratio toti mundo partibusque eius inserta? 10 Sen. epist. 89.16: ipse elementis locus, ut quidam putant, simplex est, ut quidam, in materiam et causam omnia moventem et elementa dividitur; dial. 4 (= de ira 2).19.1: elementa [sunt] quattuor, ignis, aquae, aeris, terrae. The theory of the elements falls within physics: cf. Long and Sedley 1987: chap. 47, Wildberger 2006: I 60–79. 7

220

mireille armisen-marchetti

multiplicity of being becomes organized by grouping into species (species) and genera (genera) is treated in Epistle 58, where we find a complex ontological outline, which is somewhat puzzling, as it appears to be isolated within the Stoic tradition known to us. In this letter Seneca reproduces—so he tells us—a lecture by a scholarly friend concerning the six modes of being according to Plato.11 Within this report he inserts a section that he presents as his own (si indicavero: epist. 58.8), with the alleged intention of discovering “the first genus […] from which all other species depend, whence all division derives, within which all things are comprised.”12 In order to discover this first genus, Seneca proceeds by an upward path, as was customary in the schools, gradually going back from the species “man” to the supreme genus “what is,” quod est (translating τὸ ὄν, in accordance with the linguistic equivalence posed in Epistle 58.7). The quod est subsumes the bodies and the incorporeals, with no third class of beings, and—Seneca insists—undoubtedly amounts to the supreme genus.13 We thus get this first ontological outline:

This outline, it must be emphasized, is the one Seneca adopts as his own, but it is poorly attested in the remaining tradition.14 In fact, in the Stoic testimonies concerning the genera of being, the supreme genus subsuming bodies and incorporeals is called τί.15

11 In order to appreciate to what extent this section of the letter can be considered to mirror real Platonic thinking, cf. Setaioli’s analysis (1988: 137–140); also Gersh 1986: 188–194, Chaumartin 1993b. 12 Sen. epist. 58.8: primum illud genus […] ex quo ceterae species suspensae sunt, a quo nascitur omnis divisio, quo universa comprensa sunt. 13 Sen. epist. 58.11: “quod est” aut corporale est aut incorporale; 58.14: “quod est” in has species divido, ut sint corporalia aut incorporalia: nihil tertium est. This is no doubt the supreme genus: epist. 58.12: illud genus “quod est” generale, supra se nihil habet; initium rerum est; omnia sub illo sunt; 58.13: illud genus […] merito primum poni. 14 For τὸ ὄν as supreme genus one may quote Diog. Laert. 7.61: “supreme genus is what is genus without itself belonging to a genus, as is τὸ ὄν”; this is a weak testimony, however, since “as is τὸ ὄν” is not well attested in the manuscripts. On the other hand, Philo (de agr. 139) similarly divides τὰ ὄντα into bodies and incorporeals, but although this passage has been collected in SVF II 182, we are not sure it is Stoic; it is interesting that this division of beings is combined with a downward division of bodies, as in Epistle 58.12. 15 SVF II 330 and 331. Elsewhere, τὸ ὄν is only referred to bodies: SVF II 329ab.

ontology and epistemology

221

Notwithstanding, Seneca adds, some Stoics want to superpose on the genus quod est “another still higher and more essential genus,” aliud genus magis principale (epist. 58.13), the quid (clearly the translation of τί). This quid comprises not merely beings, but also non-beings: Epist. 58.15: In rerum, inquiunt, natura quaedam sunt, quaedam non sunt, et haec autem, quae non sunt, rerum natura complectitur, quae animo succurrunt, tamquam Centauri, Gigantes et quicquid aliud falsa cogitatione formatum habere aliquam imaginem coepit, quamvis non habeat substantiam. In nature, so they say, some things are (quaedam sunt), others are not (quaedam non sunt); nature comprises also the things that are not, those that appear to our mind, such as the Centaurs or the Giants, and everything else that received a form through false imagination and began to present an image, though being devoid of substance.

Quid (τί), as we just said, is in fact attested as the supreme genus in Stoic texts. Seneca, however, has said that the quid superposed itself 16 on the quod est: we must therefore understand that quaedam sunt is equivalent to the former quod est, and subdivide quaedam sunt into bodies and incorporeals, which results into the following outline:

The most striking originality of this second outline is the integration of the quaedam non sunt into the genus quid, and the identification of these non-beings with quae animo succurrunt, themselves explicated by Centauri, Gigantes et quicquid aliud falsa cogitatione formatum. These non-beings, therefore, are identical with imaginary mental representations, which are not bodies as they lack substance. They appear to correspond to what the Stoics called the νοούµενα, “concepts” formed through mental operations based on material provided by sensible experience: we do in fact find, among the examples of νοούµενα, Tityos and the Centaur17—which, however,

Sen. epist. 58.13: Stoici volunt superponere huic. Diog. Laert. 7.52f. = SVF II 87: Diogenes distinguishes the concepts (νοούµενα) formed through contact, resemblance, analogy, transference, composition, and opposition. As an 17

222

mireille armisen-marchetti

poses a problem, inasmuch as, in the Stoic system, the νοούµενα are not non-beings.18 On the other hand, the words “everything else that received a form through false imagination (falsa cogitatione) and began to present an image (habere aliquam imaginem coepit), though being devoid of substance” seem to us to have a parallel in a passage of the De tranquillitate animi. There the idea is expressed not in ontological, but in psychological terms, by appealing to the theory of mental representation. Seneca is speaking about the hallucinations of the insane: “false images of things perturb the insane; […] they are aroused by the appearance of something, whose vanity is not apprehended by their flawed mind.” 19 The terminology is the same as in Epistle 58.15 quoted supra: falsae (imagines) / falsa (cogitatione); imagines / imaginem. What follows, proritat illos alicuius rei species, proves that what Seneca has in mind here is the type of false perception that the Stoics called φάντασµα.20 The origin of Seneca’s second outline cannot be ascertained. According to an old hypothesis,21 τὸ ὄν might have been the supreme genus in the primal Stoic outline (corresponding to Seneca’s first outline), and later replaced by the τί through Chrysippus’s agency. But J. Brunschwig, in a convincing essay,22 establishes a quite different chronology, according to which the primal Stoic outline posed the τί as the supreme genus subsuming bodies and incorporeals, the latter being conceived of as non-beings. This was replaced by a second outline, illustrated by Seneca in Epistle 58.11f. and 14, with τὸ ὄν / quod est subsuming bodies and incorporeals, the latter being conceived of as a mode of being. Finally, a third, later outline appeared: the one mentioned by Seneca under the genus τί (epist. 58.13 and 15), trying to reconcile the two previous ones with each other by re-establishing, besides the beings, quaedam sunt, a class of non-beings, quaedam non sunt, made up of the νοούµενα conceived through mental operations. example of concepts by analogy, he offers Tityos; as an example of concepts by composition, the Centaur. 18 For the question of the genera of Stoic being, and for the problem posed by Seneca’s own outline in epist. 58, cf. Rist 1969: 152–158, Hadot 1968: 156–162, Pasquino 1978, Brunschwig 1988, Wildberger 2006: I 92–99. 19 Sen. dial. 9 (= tranq.).12.5: insanos falsae rerum imagines agitant; […] proritat illos alicuius rei species, cuius vanitatem capta mens non coarguit. 20 SVF II 54: “the phantasma is what we are drawn to through an attraction devoid of content (διάκενον ἑλκυσµόν) due to the imagination. This happens to the melancholic and the insane.” 21 Zeller 1904: 94, appealing, besides our Seneca text, to Diog. Laert. 7.61 (cf. note 14). 22 Brunschwig 1988.

ontology and epistemology

223

On her part, J. Wildberger23 appears to emphasize a meaningful point when she attributes to Seneca the intention to polemicize with Plato: with the first outline, the one in Epistle 58.11 f. and 14, Seneca’s object is, according to her, to stress the contrast between the multiple modes of being he attributes to Plato in the rest of the letter and the simplicity of his own quod est. As far as the distinction of genera, which Seneca appends to the quid of “certain Stoics” (epist. 58.15) is concerned, Wildberger believes that it is meant to utterly deny ontological status to the cogitabilia (epist. 58.16), which for Plato represent the veritable form of being. 1.3. Continuing our survey of the different branches of the ontological outline in Epistle 58, we shall now consider the quae(dam) sunt, the beings, and their subdivision into bodies and incorporeals. How does Seneca define the substance and meaning of the genus “body”? In Epistle 58.9–11 he adopts at first an upward procedure of classification, as was customary in the schools, proceeding from the species “man” and “dog” to the genus of “animals” (animalia), then from the genus of animals to the genus of “animate beings” (animantia), from the genus of animate beings to that of “bodies” (corpora), and finally, as we have seen above, from the bodies to the quod est. A downward classification appears in Epistle 58.12: starting from the genus generale (the quod est), Seneca goes back down to the individuals, Cato, Cicero, Lucretius (we should not miss the Roman nature of these examples, which testifies to the fact that Seneca has deeply assimilated this classification, whose peak is occupied by the quod est). Bodies, too, receive their own definition, which falls within the fields of both ontology and physics. The most meaningful texts are Epistles 106 and 117, which present elements totally in keeping with what we know from other sources about Stoic doctrine. A body is, in the first place, what acts and is acted upon: this statement, which goes back to Zeno, returns several times in the Epistles.24 This definition functions as support for several argumentations, often developed in the form of syllogisms. Such is the case with Seneca’s demonstration of the corporeal nature of good: good acts, since it is useful; now, what acts is a body.25 Besides, good acts upon the soul, and this is the 23

Wildberger 2006: I: 99. Sen. epist. 117.10: et quod fit et quod facit corpus est; also epist. 106.4, 117.2. The definition goes back to Zeno (SVF I 90, 146) and Cleanthes (SVF I 518). Conversely, according to Chrysippus, incorporeals are unable to act and be acted upon: SVF II 363. 25 Sen. epist. 106.4: bonum facit; prodest enim; quod facit corpus est. 24

224

mireille armisen-marchetti

hallmark of a body.26 The homogeneous nature of body and soul is proved by the fact that both reciprocally act upon each other; therefore passions are also bodies, inasmuch as they can affect the expression of a face, making it blush or blanch.27 By the same token and for the same reasons, virtues are bodies too.28 A body is also characterized by its capability to touch and be touched: “Can it ever be doubted that that by which something can be touched is a body? In fact, ‘nothing except a body can touch and be touched,’ as Lucretius says.”29 Therefore, what is capable of affecting a body is itself a body: “Nothing can befall without contact; what touches a body is a body.”30 A body has some futher characteristics: being able to move (“what is endowed with movement is a body”)31 and—as a natural consequence—possessing a vis enabling it both to move and to resist movement: “Besides, all that possesses a force permitting it to shove, constrain, withhold, and restrain is a body […]. What rules a body is a body, what acts by force upon a body is a body.”32 We recognize the Stoic idea according to which a body is characterized by tridimensionality and resistance, ἀντιτυπία, all bodily action consisting of a play of shoves, shocks, and resistance, without which neither affecting nor being affected could ever take place.33 1.4. The second branch of the genus quod est, according to the ontological outline of Epistle 58, are the incorporalia: this division, firmly established in Epistle 58, is reaffirmed in Epistle 89. 34 The term incorporale, in the Stoic technical sense, appears repeatedly in Seneca’s writings.35 The philosopher 26 Sen. epist. 106.4: bonum agitat animum et quodammodo format et continet, quae [ergo] propria sunt corporis. 27 Sen. epist. 106.5: quid ergo? tam manifestas notas corporis credis imprimi nisi a corpore? 28 Sen. epist. 106.7: corpora ergo sunt quae [scil. virtutes] colorem habitumque corporum mutant, quae in illis regnum suum exercent. 29 Sen. epist. 106.8: numquid est dubium an id quo quid tangi potest corpus sit? “tangere enim et tangi nisi corpus nulla potest res,” ut ait Lucretius [Lucr. 1.304]. 30 Sen. epist. 117.7: nihil enim accidere sine tactu potest; quod tangit corpus, corpus est. 31 Sen. epist. 117.7: quod motum habet corpus est. SVF II 387: “all that moves is a body.” 32 Sen. epist. 106.9f.: etiam nunc cui tanta vis est ut inpellat et cogat et retineat et inhibeat corpus est. […] quod imperat corpori corpus est, quod vim corpori adfert, corpus. 33 SVF II 315, 343: “what acts in a bodily fashion and what is acted upon is acted upon in the same way, since they [what acts and is acted upon] need a shove, resistance and shocks, and it could not be otherwise”; II 381, 501. 34 Sen. epist. 58.11: quod est aut corporale est aut incorporale; 58.14, 89.16. 35 Sen. dial. 10 (= brev.).8.1, benef. 6.2.2, epist. 117.3 and 10. Two further occurrences, appearing not in passages where Seneca expounds his own ideas, but in doxographical listings, do not fall within the area of Stoicism: at dial. 12 (= cons. Helv.).8.3 reason is qualified as incorporalis,

ontology and epistemology

225

nowhere offers a list of the incorporeals, but the Stoic sources inform us that canonically it included vacuum, place, time, and λεκτόν, that is, three physical entities and a logical one.36 Seneca refers to the incorporeal “time” at dial. 10 (= brev.).8.1, where he characterizes time as res incorporalis, which sub oculos non venit. It is possible to reconstruct Seneca’s reasoning, which is founded upon perfectly Stoic notions: vision is a sensible contact; but only bodies can touch and be touched;37 therefore an incorporeal cannot fall under the eye. Epistle 58 associates vacuum and time: “the sixth genus [of being] is that of quasi-existing things, like vacuum or time.”38 But although the coupling inane/tempus and the expression quae quasi sunt appear to be reminiscent of the Stoics’ description of the incorporeals, this passage, in which Seneca purports to describe the modes of being “according to Plato,” can hardly pass as Stoic. The incorporeal “place” does not appear in Seneca. By contrast, the incorporeal “λεκτόν” is the object of two important treatments, both related to a moral problem. The first one appears in the De beneficiis, in connection with the question of whether a benefit can be withdrawn: Benef. 6.2.1 f.: An beneficium eripi posset quaesitum est. Quidam negant posse; non enim res est, sed actio. […] Aliud est beneficium ipsum, aliud quod ad unumquemque nostrum beneficio pervenit. Illud incorporale est, inritum non fit; materia vero eius huc et illuc iactatur et dominum mutat. […] Ipsa rerum natura revocare, quod dedit, non potest. Beneficia sua interrumpit, non rescindit. The question has been posed, whether a benefit can be withdrawn. Some deny this to be possible, since a benefit is not a thing, but an action. […] The benefit itself and what each one of us receives through the benefit are different things. The former is incorporeal, and cannot be nullified; it is the material object received through the benefit that can be tossed here and there and change owner. […] Nature itself cannot repeal the fact of having bestowed a gift.39 It may cut off its benefits, not repeal them. and at nat. 7.25.2 the expression incorporalem potestatem defines godhead. But in Stoicism both reason and God are corporeal. 36 SVF II 331. For the evolution of the list of the incorporeals, cf. Bréhier 19704, Isnardi Parente 2005. On the Stoic incorporeals, see, besides these two works, Rist 1969: 273–288; Pasquino 1978; Long and Sedley 1987: chap. 33, 49, 51; Brunschwig 1988; Duhot 1989: 87–100; Viparelli 2000: 14–22; Muller 2006: 68–71, Wildberger 2006: I: 91–201. 37 SVF II 863–871: vision is explained by the contact established between the eye and the object through a cone of tense air; and, according to Sen. epist. 106.8: tangere enim et tangi nisi corpus nulla potest res; 117.7: quod tangit corpus est. 38 Sen. epist. 58.22: sextum genus eorum quae quasi sunt, tamquam inane, tamquam tempus. 39 I translate quod dedit as “the fact of having bestowed,” not “what has been bestowed”: this latter interpretation would run counter to Seneca’s argument.

226

mireille armisen-marchetti

Seneca is borrowing somebody else’s argument (quidam negant), but he makes it his own, as shown by the fact that he adopts, and diffusely comments upon, its conclusion, namely, that benefits cannot be taken back (benef. 6.2.3f.). We cannot determine the Stoic authors of this argumentation, but we do know the theorem whose application it is. What is incorporeal? The actio, says Seneca, “action”: this should be taken to refer to the fact of bestowing, to the abstract relationship established through language between the benefactor and the object bestowed. Whereas the object of the benefit is a res, a body, the benefit relationship, isolated and made abstract by way of language, and considered as such in and of itself, falls within the sphere of the λεκτά, or logic incorporeals. But although he uses the technical term incorporale, Seneca does not endeavor, as we have just done, to clarify the nature of this incorporeal or the ontological status of the benefit relationship. The word incorporale is only used in order to support his contention that a benefit cannot be withdrawn, inasmuch as it is not identical with the material object related, materia eius. The materia (the object bestowed) can change hands or be changed itself; an incorporeal, by contrast, is not exposed to modifications affecting bodies, inasmuch as it is not a body. The roundabout path through which Seneca strives to prove his point by way of the concept of “incorporeal” is only aimed at rescuing the actio of benefit from the scope of materiality. Epistle 117, by contrast, tells us more. The question is: is “being a wise man,” sapere, a good? Seneca expounds the Stoic answer: Epist. 117.2 f.: Placet nostris quod bonum est, corpus esse. […] Sapientiam bonum esse dicunt: sequitur ut necesse sit illam corporalem quoque dicere. At sapere non putant eiusdem condicionis esse. Incorporale est et accidens alteri, id est sapientiae. The philosophers of our school believe what is good to be a body. […] They say that “wisdom” is a good: it follows that it is corporeal. But they do not believe “being wise” to have the same status. This is an incorporeal and an accident of another being, namely “wisdom.”

The reasoning is the same as at benef. 6.2, quoted supra: Epistle 117 distinguishes the “wise man” and “wisdom”—which are bodies—on the one hand, and the relationship uniting them, “being wise”—which is an incorporeal— on the other. But here Seneca adds a further point, missing atbenef. 6.2: the incorporeal sapere is an “accident” (accidens) affecting “wisdom” (this term, accidens, as we shall see below, refers to the logical funtion of the verb, which the Stoics consider to be a predicate).40 It follows, Seneca goes on at § 4, that 40

Accidens is the translation of the Greek συµβεβηκός, which the Stoic texts associate with

ontology and epistemology

227

the Stoics are forced to declare, velint nolint, that although happiness is a good, living happily is not. Living happily cannot therefore fall within the class of desirables: it is not an expetendum (translation of αἱρετόν), but only an expetibile (translation of αἱρετέον), which is not sought as a good in itself, but merely adds itself to the good sought after.41 Several Stoic testimonies coincide in informing us that these αἱρετέα are incorporeal predicates (ἀσώµατα καὶ κατηγορήµατα), like, for instance, “acting prudently” (τὸ φρονεῖν) and “acting moderately” (τὸ σωφρονεῖν)—and, we could add, “being wise,” the sapere of Epistle 117. Summing up, “being wise” is an incorporeal because it is a predicate, that is to say, a λεκτόν. This is even clearer further along in the Epistle, where Seneca again quotes the Stoici. Just as there is a difference, he says, between a field and owning a field, so there is a difference between sapientia and sapere. Sapere is “what happens to the person possessing wisdom”:42 we recognize the notion of accidens, which has been mentioned above. “Being wise” is included in the “movements of the soul that state something about bodies.”43 These movements of the soul are not bodies: so, when I say “Cato walks,” I do speak about a body, but what I say, quod nunc loquor, is not a body, but “a declarative element concerning the body, which some call a proposition, others an enunciation, others a saying.”44 What is thus defined in the most explicit way is surely the incorporeal that goes by the name of λεκτόν. This is confirmed by the very example chosen by Seneca, “Cato walks,” a mere Roman version of the canonical Greek example: “Dio walks” or “Socrates walks.”45 But the incorporeal nature of the λεκτόν sapere entails a serious drawback, namely, an ontological and moral devaluation, in that—as we have seen—it forces the Stoics to state that beate vivere is not a good, and that sapere is not κατηγόρηµα, predicate: cf. Posidonius F 95 K. (Stob. ecl. 1.13.1C). The Stoic κατηγόρηµα is a verb: in the canonical example “Socrates walks,” the predicate is “walks”; here, in Epistle 117, sapere, therefore, admits of functioning as a predicate. And this predicate is an incorporeal because of its very linguistic nature. SVF I 448 and III Arch. 8: “Cleanthes and Archedemus give κατηγορήµατα the appellative of λεκτά.” Cf. Long and Sedley 1987: chap. 33, especially the texts J and M. 41 Sen. epist. 117.5: [expetibile] non petitur tamquam bonum, sed petito bono accedit. Cf. SVF III 89 and 91: the αἱρετέα are “predicates […] correlates of goods,” κατηγορήµατα […] παρακείµενα δ’ ἀγαθοῖς: Sen. epist. 117.5: petito bono accedit. For expetibilis/expetendus cf. Setaioli 1988: 297 f. nn. 1378 f. 42 Sen. epist. 117.12: id quod contingit perfectam mentem habenti. 43 Sen. epist. 117.13: motus animorum enuntiativi corporum. 44 Sen. epist. 117.13: enuntiativum quiddam de corpore, quod alii effatum vocant, alii enuntiatum, alii dictum. 45 “Dio walks”: SVF II 204 (Diog. Laert. 7.70); “Socrates walks”: SVF II 205 (S. Emp. adv. math. 8.96).

228

mireille armisen-marchetti

an expetendum but a mere expetibile (epist. 117.4f.): a devaluation, Seneca resolutely declares, that goes against the praesumptio omnium hominum,46 the consensus that sees a good both in sapientia and in sapere (epist. 117.6). Seneca will therefore engage in an effort to demonstrate through a whole series of dialectical arguments that sapere is not an incorporeal, but a body and a full-fledged good (§6–11 and 14–17); but this demonstration is not relevant for this chapter.47 Let us now try to outline Seneca’s personal position as far as the Stoic incorporeals are concerned. It is in fact clear that in several instances he detaches himself from the Stoic vulgate. In the first place we witness this at the strictly ontological level. If we refer to the outlines in Epistle 58, it should be remembered that Seneca includes the incorporeals in the genus of “being,” the quod est, by the same token as the bodies—not, as in the Stoic canonical outline, in the genus of quid, simply the “something”: this imparts an assured ontological dignity, inasmuch as they are sharply distinguished from pure “non-being,” the quaedam non sunt, which are, as we have remarked, the νοούµενα of imagination. There is a further, substantial estrangement from the Stoic position: incorporeals appear to be endowed with the ability to hold a moral value, which contradicts their ontological status as non-bodies. This is the case with “time”: Brev. 8.1: Re omnium pretiosissima luditur; fallit autem illos, quia res incorporalis est, quia sub oculos non venit ideoque vilissima aestimatur, immo paene nullum eius pretium est. Many trifle with the most precious of all things; they [scil. the non-philosophers] are misled by it, because it is an incorporeal, because it does not fall under the eye, and for this reason it is valued very little, actually next to nothing.

On the one hand, as we have said, the incorporeal “time” is correctly defined as what does not fall under the eye; and it is this invisibility that causes the stulti, the non-philosophers, to grant no value to time. In this respect we might say that the stulti, paradoxically, prove to be accomplished dialecticians! We should remember, in fact, that according to orthodox Stoicism only a body can be a good.48 Here, by contrast, Seneca calls time res pretiosissima, an assessment incompatible with its status as an incorporeal.49 46

On the subject of consensus and the “common opinions” (κοιναὶ ἔννοιαι), cf. Verbecke

1993. 47 Wildberger (2006: I 165–167) provides a detailed analysis of Seneca’s demonstration and its compatibility with Stoicism. 48 Sen. epist. 117.2: placet nostris quod bonum est corpus esse. 49 As rightly remarked by Wildberger 2006: I 115.

ontology and epistemology

229

We witness another dissimilarity of the same kind: in Epistle 117, as we have just seen, Seneca sharply maintains that sapere, contrary to what the Stoic dialecticians assert, is a good. Does he then impart a moral value to the incorporeals? Their ontological promotion as found in Epistle 58, where the incorporeals, though non-bodies, are nevertheless sharply distinguished from “non-being,” might lead us to believe so. We should not forget, however, what we read at benef. 6.1.2, where, by contrast, the beneficium’s incorporeal status imparts it unalterability through its very insubstantial nature: it is because it is an incorporeal that the beneficent action cannot be affected by anything happening after. I will draw two conclusions from this: Seneca is perfectly familiar with the status granted to incorporeals by Stoic dialectics, as shown by the technicality of his analyses; but, of this status, he retains only those aspects that suit his ethical purposes. The abstract character of the incorporeal may therefore, on the one hand, bestow a paradoxical security when benefits are at play, and, on the other, confine it to an unacceptable non-existence when it comes to “time” and “being wise.” Seneca sacrifices dialectical consistency, not because he is ignorant of, or disregards it, but because his scope and horizon are those of moral pedagogy, whose importance is far greater in his eyes. 1.5. The genus generale of bodies can be divided into successive species down to the individual (epist. 58.12). But a further type of division, or rather analysis, can be applied to bodies, by way of the four “categories.”50 These categories are not subdivisions, species by species, of a primal genus, but a catalog of the different metaphysical points of view under which a body may be considered: “Stoics present in fact a quadruple division into substrata (ὑποκείµενα), what is qualified (ποιά), what is disposed in a certain way (πως ἔχοντα), and what is disposed in a certain way in relation to something (πρός τί πως ἔχοντα)”;51 and each being falls in turn under all four categories.52 In Seneca these notions appear more or less explicitly in the course of several dialectical discussions. 50 The ancient testimonies themselves only refer to “genera,” but most modern interpreters prefer the term “category,” in order to avoid confusion with the genera of being (described by Seneca, as we saw, in Epistle 58). 51 SVF II 369, 371. Cf. Goldschmidt 1979: 13–25, Rist 1969: 152–172, Pasquino 1978, Graeser 1978, Long and Sedley 1987: chap. 28 and 29, Duhot 1991, Sonderegger 2000, Gourinat 2000: 129–136, Wildberger 2006: I: 86–91. 52 Plut. not. comm. 1083E. Duhot (1991) shows quite perspicuously that Stoic categories are not univocal, but rather establish the ontological level of a being by way of differentiating, according to the specific needs of the analysis at hand.

230

mireille armisen-marchetti

1.5.1. The Substratum The ὑποκείµενον, or “substratum” is referred to when a being is considered in relation to the passive matter it is made of, apart from the qualities that characterize it, and make it a “qualified being.” A problem is posed by the fact that Seneca gives no indication of adopting a specific Latin translation for the term ὑποκείµενον, so that when he uses words like substantia or materia, there is no way to determine whether his analysis is moving at the metaphysical level of the ὑποκείµενον or at the physical one of the οὐσία/ὕλη. In Epistle 113, however, the term “substance,” substantia, possibly refers to the ὑποκείµενον.53 The context is indeed about categories, as we shall presently see. 1.5.2. What Is Qualified According to a convincing hypothesis, the doctrine of categories was devised by Chrysippus in order to account for the ontological permanence of the individual through the changes affecting him, in the face of the Academics’ position on the issue;54 Chrysippus allegedly replied by introducing a difference between substance and the qualified being. Epistle 113, which investigates the notions of individual and identity, appears to echo this set of problems: Epist. 113.11: Omne animal donec moriatur id est quod coepit: homo donec moriatur homo est, equus equus, canis canis; transire in aliud non potest. All animate beings remain what they were from the beginning to the moment they die: a man is a man until he dies, a horse a horse, a dog a dog; they cannot change into something else.

We must assume that the Stoic analysis of the second category, the one referring to the “qualified individual” (ποιόν) lies behind this text. The idea that a being, as long as it exists, does not turn into something else rests upon the notion of “individual form” (ἀτοµωθὲν εἶδος), which matter receives from the active and divine principle pervading it. This form persists through all the changes affecting the substance—and only the substance—of the living being as long as it exists; it is what guarantees that the individual will hold fast and endure.55 53 Sen. epist. 113.4: singula animalia singulas habere debent substantias. Cf. also nat. 1.6.4 and 1.15.6. 54 Sedley 1989. Seneca repeatedly emphasizes the changes the human being goes through at the different ages and stages of his life—which, however, do not affect its ontological persistence: cf. epist. 58.22 f. (quoting Heraclitus’s image of the river), 104.12, 118.14, 121.16. 55 SVF II 395: “if it is true that there is, even in composite beings, the individual form (τὸ ἀτοµωθὲν εἶδος) in relation to which, according to the Stoic philosophers, it can be said that

ontology and epistemology

231

Seneca then tackles the correlated question of discernableness. Each individual is endowed with an irreducible originality: Epist. 113.15f.: Nullum interim animal alteri par est. Circumspice omnium corpora: nulli non et color proprius est et figura sua et magnitudo. […] Quae similia videntur, cum contuleris, diversa sunt. No two animate beings are alike. Inspect the body of each one: everyone has its own color, shape, and size. […] Things which appear to be alike, when you compare them, result to be different.56

The doctrinal background is disclosed by Plutarch, between the lines of a polemic he conducts against the Academics: “(the Stoics) loudly proclaim that (the Academics) confuse everything when they assert the impossibility of distinguishing things one from the other, by forcing one and the same quality (ἕνα ποιόν) to be in two different substances (ἐπὶ δυοῖν οὐσιῶν).”57 Forcing one quality to be in two substances amounts to jeopardizing the notion of “individual form,” which, as we have seen, is the hallmark of the individual. Seneca, however, does not engage in the discussion of these notions, and is content with referring to the differences between individuals as self-understood. This does not necessarily imply avoidance of technicalities: some of the reported arguments in support of discernability are indeed quite empirical.58 1.5.3. What Is Disposed in a Certain Way The third category describes the actual state of a being.59 Being disposed in a certain way refers to an individual who finds himself in a particular state, while his constituent elements remain unchanged: so, the fist is the hand disposed in a certain way.60 In Epistle 113 Seneca presents several occurrences of the quodam modo se habens (translation of πως ἔχον). The question addressed is: are virtues animate beings? Seneca reports the opinon of the nostri, also referred to as antiqui (epist. 113.1). This surely refers to the old Stoics, probably Chrysippus. Epist. 113.2: Animum constat animal esse. […] Virtus autem nihil aliud est quam animus quodam modo se habens; ergo animal est.

something is individually qualified (καθ’ ὅ ἰδίωc λέγεται ποιόν). This form, besides, arises and disappears all at one time, and remains the same throughout the whole life of the composite being, even if its several parts are born and die at different moments.” 56 Cf. SVF II 113. 57 Plut. not. comm. 36, 1077C = SVF II 112, 396. 58 So Cic. ac. pr. 2.56–58 resorts to the examples of twins and eggs. 59 SVF II 369, 390, 400. 60 S. Emp. Pyrrh. 2.81.

232

mireille armisen-marchetti It is certain that the soul is an animate being. […] Now, virtue is nothing but the soul disposed in a certain way; therefore it is an animate being.

Later in the letter similar statements are made about justice, then about courage.61 Seneca is following Stoic definitions, as confirmed by Sextus Epiricus, according to whom virtue is the guiding part of the soul, the ἡγεµονικόν, “in a certain disposition,” or “disposed in a certain way.”62 The Stoici quoted by Seneca avail themselves of this statement in order to assert that virtues are animate beings, and that therefore there are a number of animals within us—which Seneca rejects, because for him an animate being is an individual and there cannot be several individuals in relation to one substantia, a term that probably corresponds to ὑποκείµενον (epist. 113.4). Later, Seneca quotes a further Stoic statement, this time supporting his point: Epist. 113.24: Idem est animus et animus et iustus et prudens et fortis, ad singulas virtutes quodammodo se habens. The soul, and the just, provident, and courageous soul, disposed in a certain way in relation to each individual virtue, are one and the same thing.63

Resorting to the third category, then, allows us to account for the multiplicity of virtues without creating a number of new animate beings, i.e., new individuals. Virtue, inasmuch as it is animus quodammodo se habens, is a particular state of the soul, i.e., of a being, which is itself an individualized body and a qualified entity, a ποιόν. The analysis of virtue through the concept of quodammodo se habens attributes modifications to the soul without affecting its individuality and uniqueness. As for the expression ad singulas virtutes, it puts us on the track of the fourth category, which we shall now investigate. 1.5.4. What Is Disposed in a Certain Way in Relation to Something The Stoics describe in the following way the outward differentiating characteristics that pertain to a connection with other bodies and can cease without affecting the body concerned: “Beings disposed in a certain way in relation to something are like a man on the right, a father, and the like. […] What is disposed in a certain way in relation to something […] is dependent on a relation to something else.”64 Whoever is disposed in a certain way in relation to something (or somebody) will no longer be on the right, if his neighbor 61

Sen. epist. 113.7 and 11. SVF III 75 = S. Emp. adv. math. 11.22. 63 For the problem of the uniqueness and multiplicity of virtues, with Ariston’s and Chrysippus’s positions on the subject, cf. Long and Sedley 1987: chap. 61. 64 SVF II 403. 62

ontology and epistemology

233

changes his position, will no longer be a father if his son dies, but his own mode of being will not be affected. This idea appears in Epistle 121. Seneca has asked himself “whether all animate beings have a notion of their own constitution. We gather that they do, primarily from the fact that they move their limbs in an easy and suitable fashion […].”65 Constitutio corresponds to Chrysippus’s notion of σύστασις,66 and the only definition that we have of this notion is Seneca’s own: Epist. 121.10: Constitutio est principale animi quodammodo se habens erga corpus. The constitutio is the guiding part of the soul disposed in a certain way in relation to the body.

The principale animi, the hegemonic part of the soul, then, is characterized in its outward relation to another entity, the body. Seneca’s fictive interlocutor mocks with some justification the subtlety of this definition.67 Nonetheless, the advantage Seneca expects to draw from it is once more related to his aim to reconcile the permanence of the individual with the changes affecting his or her person. If we take the same individual at different ages, clearly, the baby, the child, the young man, and the old man will present considerable differences. How does the individual recognize himself, how is he aware of the permanence of his self all through these different stages? The fact is that the relations entertained by his ἡγεµονικόν with the changing elements of his physical person follow the fashion proper to relative beings. They may change without intrinsically affecting the ἡγεµονικόν, that is the self, or, at a later stage, the individual’s natural adaptation to itself, the conciliatio.68 2. Epistemology Seneca believes it possible to attain an apprehension, scientia, and a complete knowledge of all the beings defined by ontology. This knowledge is wisdom, sapientia, the one study that can veritably be called “liberal,” that is worthy of 65 Sen. epist. 121.5: […] an esset omnibus animalibus constitutionis suae sensus. Esse autem ex eo maxime apparet quod membra apte et expedite movent. At the beginning of the letter Seneca has disclosed that he is following Posidonius and Archedemus of Tarsus. 66 SVF III 178: “according to Chrysippus in the first book of “On Ends” the first characteristic distinctly pertaining to every living being is its own constitution and its awareness of it (τὴν αὐτοῦ σύστασιν καὶ τὴν ταύτης συνείδησιν).” 67 Sen. epist. 121.10: hoc tam perplexum et subtile et vobis quoque vix enarrabile […]. 68 Sen. epist. 121.15f.: unicuique aetati sua constitutio est, alia infanti, alia puero, alia seni: omnes ei constitutioni conciliantur in qua sunt. […] non enim puerum mihi aut iuvenem aut senem, sed me natura commendat.

234

mireille armisen-marchetti

the free man.69 The modes of this scientia are methodically analyzed in Epistle 89. Wisdom is defined as the “science of things divine and human,” divinorum et humanorum scientiam.70 Seneca mentions a further definition, the science of causes,71 but considers it superfluous, because, he says, causes are part of divine things. But how does one attain the complete knowledge that is wisdom? Through philosophy, considered—in accordance with etymology— as “love and taste for wisdom”:72 a definition that Seneca appears to take as his own, that, at any rate, he does not attribute to anybody else. By contrast, he attributes to different sources other definitions he quotes shortly after, which lay more stress on the ethical aspect, but do not seem to retain his attention.73 Philosophy, in turn, presents a marked unity, of a biological type, expressed through metaphors loaded with meaning. The philosophia is an “immense body,” in the likeness of the cosmos itself 74 (of which, as is well-known, the Stoics have a vitalist conception): an idea implying a dynamic unity and the interpenetration of the several parts, which entails the requirement to study philosophy not in linear sequence, but by the simultaneous exercise of its parts.75 In practice, however, subdivision cannot be helped, inasmuch as the human mind does not posses the ability to grasp it globally. But Seneca resolutely maintains76 that this stems from a merely practical need, not from any objective reality. Answering a question posed by Lucilius, he will then expound the divisions of philosophy, but will avoid crumbling it excessively—which would be not merely useless, it would impair our understanding of it.77 This rejection of excessive partition amounts to an implied criticism of certain exaggerated subdivisions current in Stoicism; Cleanthes, for example, distinguished six parts of philosophy.78 69 Sen. epist. 88.2. Cf. Stückelberger 1965, and, for the relation to Posidonius, Setaioli 1988: 316–322. 70 Sen. epist. 89.5; also 31.8, 68.2, 74.29, 88.33, 88.35, 90.3, 104.22, 110.8. It is a Stoic traditional definition: SVF II 35, 36. 71 Sen. epist. 89.5: sapientia est nosse divina et humana et horum causas. Cf. Cic. off. 2.5, Tusc. 4.57, 5.7. Cf. Zechel 1966: 41f. Possibly this definition goes back to Posidonius: cf. epist. 95.65, and the parallel with Philo, de congr. 79. 72 Sen. epist. 89.4: sapientiae amor et affectatio. 73 Sen. epist. 89.5: alii studium illam [scil. philosophiam] virtutis esse dixerunt, alii studium corrigendae mentis; a quibusdam dicta est adpetitio rectae rationis. Cf. Zechel 1966: 43f. 74 Sen. epist. 89.1: ingens corpus. The image comes from old Stoicism: SVF II 38. 75 SVF II 41. Cf. Hadot 1991. 76 Sen. epist. 89.1 f. 77 Sen. epist. 89.2 f. 78 SVF I 482.

ontology and epistemology

235

Epist. 89.9: Philosophiae tres partes esse dixerunt et maximi et plurimi auctores: moralem, naturalem, rationalem. According to the most authoritative and numerous sources, philosophy comprises three parts: ethics, physics, logic.79

This division is well attested in Stoic sources, which attribute it to Zeno but also point out that it predates Stoicism.80 Even though there is no chronological or hierarchical relation among the parts of philosophy, the order in which Seneca lists them here has few parallels81 and leads us to surmise that a hierarchical arrangement in decreasing order of dignity may be intended: at the end of Epistle 89 Seneca advises, in fact, refering everything to ethics, and in Epistle 88 he characterizes philosophy by the fact that, unlike the liberal arts, it teaches virtue. 82 On the other hand, we know that Seneca has a high appreciation of physics, and that, by contrast, he often criticizes the abuse of dialectics, i.e., logic. However, in Epistle 88.24, although this is a passage that—it must not be forgotten—is influenced by Posidonius, the order is physics, ethics, logic. Finally, at nat. 1 pr. 1, philosophy is denoted through two parts only, “the one concerning man and the one concerning the gods,” that is ethics and physics,83 the latter being conceived of as a theology: the fact is that Seneca does not take into account philosophy as a whole, but is content with comparing two of its parts, surely against the background of the definition of wisdom as the science of things divine and human. Ethics in turn admits of some divisions: Epist. 89.14: [Moralem partem] in tria rursus dividi placuit, ut prima esset inspectio suum cuique distribuens et aestimans quanto quidque dignum sit […] secunda de impetu, de actionibus tertia. It seemed convenient to divide ethics into three parts, the first concerned with evaluating and meting out what is due to each one as well as tax79

Cf. also Sen. epist. 88.24, nat. 1 pr. 1 (physics and ethics only). SVF II 35–38. Cic. fin. 4.4. For Stoic divisions of philosophy, cf. Pohlenz 1970: 32–36, Zechel 1966, Verbecke 1973, Hadot 1991, Muller 2006: 51–60, Wildberger 2006: I 133–137, II: 365–368. Zechel and Wildberger point out the marked similarity between Seneca and S. Emp. adv. math. 7.1–16, and assume the existence of a common source. 81 According to Diog. Laert. 7.41 Diogenes of Ptolemais also began with ethics. The most usual order, however, is physics, ethics, logic (SVF II 35, 37, 38, 39, 40); we also find logic, ethics, physics (SVF II 42, 44), and logic, physics, ethics (SVF II 43). 82 Sen. epist. 89.23: omnia ad mores et ad sedandam rabiem adfectuum referens. 83 The context proves, however, that Seneca has logic, too, in mind. The definition, atnat. 1 pr. 2, of ethics as what errores nostros discutit et lumen admovet quo discernantur ambigua vitae shows that Seneca envisages resorting to dialectics in order to tackle moral problems. Dialectics is in fact defined in Epistle 90.29 as the tool enabling us to establish truth and dispel “the ambiguities of life and language.” Cf. below. 80

236

mireille armisen-marchetti ing the value of everything, […] the second with impetus, the third with actions.84

Seneca returns to this triple division three times in Epistle 89.14f., and insists that there is no veritable moral life if any of these cognitions is missing. The order, then, is merely descriptive, with no hierarchical dignity or pedagogical chronology involved. Physics poses a more intricate problem. Its domain is divided into corporalia and incorporalia—which is tantamount to saying that it covers the area of the quod est—and each of these two parts, in turn, entails “degrees,” gradus.85 In fact, only the “degrees” of bodies are treated. The “domain of bodies,” corporum locus, is in turn divided into “things that act,” ea quae faciunt (i.e., the principles) and “things begotten from these,” quae ex his gignuntur (i.e., the elements). According to some authors, Seneca adds, the domain of the elementa is partitioned in materiam et causam omnia moventem et elementa, “into matter and cause imparting movement to all, and elements.” Materia and causa can be easily identified with the two principles—active and passive—which give rise to bodies. But what shall we make of elementa, which appears in turn as a genus and as a species of its own genus? In my opinion, what Seneca means is that the elementa can be scrutinized both from a metaphysical point of view, as resulting from the action of the supreme cause upon matter, and in and by themselves. 86 On the other hand, in Epistle 90.28f. we find a different partition of the domain of physics: theology (gods and daemons) comes first, then the initia rerum, i.e., the principles (aeternamque rationem toti inditam et vim omnium seminum singula proprie figurantem), and finally the soul. Last comes logic, which, in accordance with its etymology, is placed in relation with oratio, speech: Epist. 89.18: Superest ut rationalem partem philosophiae dividamus. Omnis oratio aut continua est aut inter respondentem et interrogantem discissa. Hanc διαλεκτικήν, illam ῥητορικὴν placuit vocari.

The same division, in a different order, appears in Diog. Laert. 7.84 (SVF III 1). Sen. epist. 89.16. For the quod est, cf. epist. 58.11 and 14. 86 One might be tempted to equate this division with the generic division in Diog. Laert. 7.132, which distinguishes the domains concerned with the cosmos, the elements, and etiology. This, however, is hardly possible: Diogenes’s partition applies to physics as a whole, not to the elements alone; besides, it is difficult to liken Seneca’s materia to the cosmos; finally, the causa omnia moven[s] is the supreme cause, not the study of the specific causes making up etiology, as it is in Diogenes. 85

ontology and epistemology

237

It remains for us to divide the rational part of philosophy. All speech is either uninterrupted or divided into answers and questions. It was decided to call the former rhetoric, the latter dialectics.

This division into rhetoric as uninterrupted speech and dialectics as an exchange of questions and answers, which is well-known from other sources,87 indicates that the distinction between rhetoric and dialectics is merely formal, the choice between the two being undoubtedly linked with practical conditions, inasmuch as rhetoric was in order in some situations and dialectics in other. This also entails rhetoric’s full pertinence to philosophy and, because of this, its status as a science. This conclusion, not drawn here by Seneca, appears in the Stoic sources: rhetoric is an ἐπιστήµη τοῦ λέγειν, which, just like dialectics, supplies truth.88 Epist. 89.17: ῾Ρητορικὴ verba curat et sensus et ordinem; διαλεκτικὴ in duas partes dividitur, in verba et significationes, id est in res quae dicuntur et vocabula quibus dicuntur. Rhetoric is concerned with words, sentences, and planning the discourse; dialectics is divided into two parts, words and meanings, that is, what is being said and the words by which it is said.

Sensus, in my opinion, means “sentence” in this passage: in this way Seneca’s list opens with the simple element, the word, and proceeding in order of growing complexity, closes with the organization of the speech itself, ordo being synonymous with dispositio.89 As far as the definition of dialectics is concerned, it coincides with Diogenes Laertius’s testimony at 7.62: “(Dialectics) is concerned, as Chrysippus says, with signifiers (σηµαίνοντα) and signifieds (σηµαινόµενα).” Starting out from this, there are numerous subdivisions, into which Seneca refuses to go, since, he says, they would require a whole volume.90 Another definition appears in Epistle 90: Epist. 90.29: [Sapientia] deinde a corporalibus se ad incorporalia transtulit veritatemque et argumenta eius excussit, post haec quemadmodum discernerentur vitae aut vocis ambigua; in utraque enim falsa veris inmixta sunt.

Diog. Laert. 7.41 f., S. Emp. adv. math. 2.6 f. Cf. SVF II 292–294. 89 I do not believe, contrary to Wildberger 2006: I: 139 (and in spite of Diog. Laert. 7.43: “there is also a division of rhetoric into invention, elocution, disposition, and performance”), that in this passage Seneca refers to elocutio, inventio, and dispositio: he would hardly mention elocutio before inventio. 90 A sample of these subdivisions appears in Diog. Laert. 7.43–45. Cf. Gourinat 2000: Part I: 19–107: “La notion de dialectique,” Hamacher 2006: 16–18. 88

238

mireille armisen-marchetti Then (wisdom) moved from corporeals to incorporeals and scrutinized truth and its foundations, after that how the ambiguities of life and language could be disentangled, for in either one the false is mixed with the true.

Dialectics, here, is connected with the notion of incorporeal,91 which is understandable inasmuch as it is concerned with the λεκτά, which are incorporeals. But the notion of ambiguity, which is also present, opens the way to a possible perversion of dialectics, when it falls into mean hands. A shrewd dialectician will then be able, by playing upon words, to muddle reality and distort morals, or simply take pleasure in intellectual subtleties that make us lose sight of the fundamentals,92 that is, the life choices induced by science and wisdom. Ontology and epistemology, as defined and studied here, can teach us something, beyond their specific domain, about Seneca’s philosophic attitude. Both evidence the solidity of his Stoic background. On the subject of ontology, founded for a considerable part on dialectics and its subtleties, Seneca displays undoubted competence. Although he does not expressly disclose the provenance of the notions he is handling, being content, most of the time, with hinting at the authors by a quidam or a nostri, it is not unduly difficult for us to determine their old or middle Stoic origin; he has no qualms, nonetheless, if he finds it useful for his parenetic intent, about wandering away from strictly Stoic definitions, as we have seen him doing in his treatment of incorporeals. Be that as it may, ontological notions are rarely investigated for their own sake; they are subordinate to other inquiries, or at most drawn forth as a reply to a specific question by Lucilius. By contrast, the question of the nature and the domain of knowledge is the object of a specific and comprehensive treatment, Epistle 89, which is partly complemented in outlying but adjacent texts: Epistles 88 and 90. Unlike ontology, then, this appears to Seneca to be a question worthy of full elucidation and an independent, if succinct, treatment—which seems very significant to me. What really matters, to Seneca, are not scholarly subtleties, not even Stoic ones. The true philosopher should never forget that the knowledge he seeks is one corpus, ingens though it be, and that its end is wisdom.

91 This definition is close to Posidonius’s ap. Diog. Laert. 7.62 (= F 188 K.): “dialectics is the science of the true, the false, and what is neither.” 92 Sen. epist. 48.4 f.: tu mihi verba distorques et syllabas digeris. Scilicet nisi interrogationes vaferrimas struxero et conclusione falsa a vero nascens mendacium adstrinxero, non potero a fugiendis petenda secernere. For dialectics as worthless play, cf., among others, epist. 45.5, 82.8–24 (cf. Hamacher 2006), 106.11, 108.23, 111.4, 113.26f., 117.19.

ETHICS I: PHILOSOPHY AS THERAPY, SELF-TRANSFORMATION, AND “LEBENSFORM”*

Aldo Setaioli

1. During the Hellenistic period an aspect of philosophy that was by no means new in Greek thought, namely the function of spiritual direction and therapy of the soul,1 became more clearly marked, as primacy was gradually awarded to ethics. This trend was obviously bound to find a favorable reception at Rome, where its most consistent expression is to be encountered precisely in Seneca’s work. A. Guillemin’s felicitous definition, “Sénèque directeur d’âmes,”2 is accepted today by nearly all scholars. His writings are tools of education3 and aim at the moral improvement of his own self as much as of his addressees, in order to attain happiness (vita beata). It must in fact be stressed from the very outset that Seneca considers himself to be in need of (self-)improvement no less than the people he addresses in his writings.4 It is in the unfaltering faithfulness to this ethical goal that his real consistency must be recognized, notwithstanding whatever doctrinal discrepancies may be detected in his work.5 We shall see that the latter are often made subservient to the ethical goal and in this way reabsorbed into our philosopher’s basically Stoic view. The first step of this therapeutic process must, of course, be negative. In Seneca’s view the therapist must first remove the attitudes that hinder his “healing” action and make it impossible to free souls from the “disease” *

Submitted for publication in 2007. Cf. Nussbaum 1994, a stimulating, if somewhat unhistoric book; also Voelke 1993. 2 Guillemin 1952, Guillemin 1953, Guillemin 1954. 3 Cf. von Albrecht 2004: 2: “Instrument philosophischer Erziehung und Selbsterziehung.” Von Albrecht is referring to the Epistulae morales, but the remark can be extended to the whole of Seneca’s philosophic work. The title of von Albrecht’s book felicitously summarizes this aspect of Seneca’s writing: the quest for transformation (and self-transformation) to be achieved by (mainly) verbal tools and leading to the good way of living. 4 Cf. epist. 8.2, 57.3, 75.15 f., dial. 7 (= vit. beat.).17.3 f., etc. 5 Cf., e.g., Setaioli 2000: 150 n. 205. 1

240

aldo setaioli

that affects them. Only after “conversion” can the positive improvements envisaged by Seneca’s program be expected. The medical metaphor is widespread in ancient philosophy6 and in Seneca it is absolutely pervasive.7 Mireille Armisen-Marchetti counts up to 248 references, between metaphors and similes, equating philosophy to medicine.8 So, the philosopher is primarily seen as a doctor of souls9 and the diseases that must be healed are passions (Greek path¯e). Cicero prefers rendering pathos with perturbatio rather than with morbus, although he feels this to be more literal;10 but he does employ the medical term aegritudo.11 Seneca distinguishes between adfectus and morbus,12 and although he employs the former in the sense of Greek pathos, he very often resorts to the image of disease in order to describe the condition of the soul fallen prey to passions.13 According to the Stoics, passion originates from a wrong judgment, 14 but it is not a merely theoretical mistake; in as much as it is a horm¯e pleonazousa it is strongly dynamic, it is a violent impulse insensitive to reason.15 Seneca fully accepts this conception.16 Two important consequences follow: 1) the first stage of the therapy will not be able to resort to completely rational arguments, which would have no effect on souls fallen prey to passion;17 and 2) in this stage the “doctor of souls” will be obliged to resort to an aggressive approach, if he expects his therapy, directed at those who are under the sway of the violence of passion, to be effective.18 Seneca’s “doctor” must be

Cf., e.g., Nussbaum 1994: 6 and passim. Cf. Steyns 1906: 51–70; Smith 1910: 39–46; Armisen-Marchetti 1989: 132–138, 317; Ficca 2001: 165–169. 8 Armisen-Marchetti 1989: 347. 9 The same applies to Epictetus and Marcus Aurelius, although each moves in his own way. Cf. Cooper 2006: 45. 10 Cic. Tusc. 3.7: ego poteram “morbos,” et id verbum esset e verbo, sed in consuetudinem nostram non caderet; 4.10: quae Graeci πάθη vocant, nobis perturbationes appellari magis placet quam morbos; fin. 3.35: perturbationes animorum […] quas Graeci path¯e appellant—poteram ego verbum ipsum interpretans “morbos” appellare, sed non conveniet ad omnia. 11 E.g., Cic. Tusc. 3.23. 12 Epist. 75.11, 83.10. See Pittet 1937: 75. 13 Cf. Borgo 1998: 13–16 (adfectus), 19 f. (aegritudo), 20–22 (aeger), 22 f. (aegroto), 134–136 (morbus). 14 Cf., e.g., SVF I 202, 207, III 384, 456, 459, 461. 15 Cf., e.g., SVF I 205, III 377, 378, 385, 412, 462, 479. For passion as something apeithes log¯ oi, cf. also SVF III 389, 394, 476. 16 E.g., epist. 85.6: si das aliquos adfectus sapienti, impar illis erit ratio et velut torrente quodam auferetur; 85.8: quantuscumque est [scil. adfectus], parere nescit, consilium non accipit. For this aspect see Wacht 1998: 515 f., 521 f. 17 Cf. esp. SVF III 389. 18 Cf. Husner 1924: 8 f. 7

ethics i: therapy, self-transformation, “lebensform”

241

ready to cut and burn, not restrict himself to bland cures;19 and his approach is linguistically matched by the other metaphor Seneca often employs in connection with his educational activity: the one referring to a “war,” or at any rate “fight,” against the passions.20 Not rarely the medical and the military images merge, especially in the “Consolations,”21 i.e., in the works belonging to a genre institutionally devoted to healing a wound of the soul. 22 We shall briefly investigate them before addressing the more personal sides of Seneca’s therapeutic program. 2. From our point of view the central problem posed by the consolatory genre lies in establishing the relationship between the two elements that converge in this type of writing: the literary and rhetorical aspect on the one hand, the philosophical on the other. The predominance of the former has been upheld by Kassel, whereas the opposite point of view has been defended by Johann.23 Although Posidonius included the consolatio in philosophical discourse, as reported by Seneca himself,24 there is no doubt that rhetoric plays a decisive role in this genre, as confirmed by the space allotted to it in rhetorical handbooks.25 Seneca’s thought may be “naturally consolatory,”26 but this does not imply that in his “Consolations” the rhetorical and literary element does not retain a crucial importance. In the consolatio addressed to his mother the purely literary aemulatio of the Ciceronian model is obvious;27 19

Cf., e.g., epist. 75.7. Cf. Wilson 1997: 62f., Ficca 2001: 169–180. Vivere […] militare est, says Seneca at epist. 96.5. See also Steyns 1906: 5–50, Smith 1910: 127–135, Armisen-Marchetti 1989: 94–97, Cervellera 1990. 21 Cf. Ficca 2001: 180–182. 22 Cf. Wilson 1997: 48: “the consolation is perhaps the paradigmatic instance in the therapeutic mode of philosophising.” 23 Kassel 1958, Johann 1968. I have expressed my opinion and discussed the copious literature in a series of papers: Setaioli 1997a (now collected and updated in Setaioli 2000: 275–323, 411), Setaioli 1999, Setaioli 2001a, Setaioli 2001b, Setaioli 2005a, Setaioli 2007c. 24 Epist. 95.65. Cf. Setaioli 1988: 336–349. 25 Cf. Ps. Dion. Hal. rhet. 6.4–6 (II, p. 281,1–283,19 U.–R.: within the treatment of the logos epitaphios); Menand. Rhet. III, p. 413,15–414,30 Sp.; cf. III, p. 421,14–422,4. 26 So Ficca 2001: 9: “Seneca, il cui pensiero è […] naturaliter consolatorio.” 27 Dial. 12 (= cons. Helv.).1.2: praeterea, cum omnia clarissimorum ingeniorum monimenta ad compescendos moderandosque luctus evolverem, non inveniebam exemplum eius qui consolatus suos esset cum ipse ab illis comploraretur. This passage closely recalls Cic. Att. 12.14.3: nihil enim de dolore minuendo scriptum ab ullo est quod ego non domi tuae legerim […] quin etiam feci, quod profecto ante me nemo, ut ipse me per litteras consolarer […] adfirmo tibi nullam consolationem 20

242

aldo setaioli

and in Epistle 99, a consolatio addressed to Marullus for the loss of a baby son, Seneca adopts the sch¯ema plagion, i.e., he purports to be scolding Marullus instead of consoling him, following the well-known rhetorical mode ostensibly pursuing a goal opposite to the one expected by the listener or the reader.28 This, of course, does not prevent him from employing the very same consolatory topoi that are found in the other writings belonging to the genre. This very text, however, exemplifies the functional use of rhetoric so characteristic of Seneca’s writing. Resorting to the sch¯ema plagion is justified at the end of the letter as the necessary means to attain Seneca’s real goal: not so much ephemeral and contingent consolation as the permanent strengthening of the soul in view of all possible misfortunes.29 In this way the Stoic “consolation” may be put on a par with the therapy against passions that appears in the rest of Seneca’s work: it aims not merely at curing a past wound, but at a lasting transformation of the soul. There are further aspects of Seneca’s “Consolations” that reappear in deeper and more developed modes in the rest of his work. In the “Consolation” addressed to his mother he is at the same time the “doctor” and the “patient,” just as he always portrays himself in his writings. Another consolatory text, Epistle 63 to Lucilius, opens with the avowal that the wise man’s apatheia cannot be expected from the common man, though this would indeed be best.30 This amounts to the frank admission that here, as in all of his work, Seneca’s therapeutic effort is not addressed to the sapiens—who does not need it; and at the end of the letter he counts himself among those who are a long way from the attainment of apatheia.31

esse talem. Cf. Cic. Att. 12.21.5. If Cicero is consoling himself for a misfortune that had befallen him rather than somebody else, Seneca surpasses him, since he is consoling somebody else for a misfortune by which he has been affected himself; that is why he can still boast, even after Cicero, that no “Consolation” is like the one he has written. Cf. Setaioli 2003: 63f. 28 Wilson 1997 misses this rhetorical device in Epistle 99. He even thinks (p. 66) that Marullus is a fictional character. But Seneca himself stresses the literary originality of his approach (non sum solitum morem secutus: epist. 99.1). This consists in applying this sch¯ema to the consolatio, where it was not common, surely not in the sch¯ema in and by itself. 29 Epist. 99.32: haec tibi scripsi, non tamquam expectaturus esses remedium a me tam serum […], sed ut castigarem exiguam illam moram qua a te recessisti, et in reliquum adhortarer contra fortunam tolleres animum eqs. In the same way, the Consolatio ad Helviam aims not merely to console his mother for his own personal exile, but to conquer the fear of exile—one of the most dreaded misfortunes (epist. 85.41, 91.8). 30 In practice, if not in theory, Seneca adopts Crantor’s position (Crantor F 3a Mette; cf. Setaioli 1999: 147). Cf. also epist. 99.15. 31 Epist. 63.14–16. Seneca, who is consoling Lucilius for the death of a friend, succumbed himself to grief at the death of Serenus. It must be noted, however, that from this experience Seneca has learned the necessity of the meditatio (numquam cogitaveram mori eum ante me

ethics i: therapy, self-transformation, “lebensform”

243

“Consolations” belonged to a well-established genre, although in Seneca they acquire some of the characters of his general therapeutic strategy. The latter, however, could not dispense with a theoretical foundation concerned with the possibility of ethical progress. According to the Stoics, every person, theoretically, is able to attain wisdom (sapientia). Man’s own nature impels him toward what is fitting for him, namely reason.32 Actually, though, this spur is hardly ever successful, due to the “perversion” worked upon man’s reason through the influence of what surrounds him, things and people alike.33 So, Seneca is convinced that each man has been equipped by nature with the tools to attain wisdom,34 though very few will actually reach that goal. The Stoic wise man is not a myth: though extremely rare, he has actually appeared on earth, and will appear again.35 Surely this will happen once in several centuries; therefore, in practice, much of the process of self-transformation promoted by Seneca will confine itself to a sort of second-degree ethics, which will not transcend the sphere of the kath¯ekonta, as defined by Panaetius.36 But Seneca never loses sight of perfect wisdom as an ideal.37 What he preaches is the sapientiae studium, the pursuit of wisdom. Happiness can be reached only through its attainment, but just striving toward it will make life tolerable.38 posse: epist. 63.14): a cornerstone of his therapeutic strategy (see below), which he amply employs in the “Consolations.” Cf. Armisen-Marchetti 1986: 188f. 32 The Stoics called this process oikei¯ osis (conciliatio): cf. SVF I 181, 197, II 724, III 178f., 229a, 492. Cf., e.g., Pembroke 1971. Most recently Gill 2006: 36–46. Seneca was familiar with this doctrine, which he discusses in Epistle 121 (cf. Setaioli 1988: 306). In this epistle he mainly dwells on the “conciliation” of living beings to their physical constitution (cf. also epist. 14.1, 116.3); but he also clearly envisages man’s “conciliation” to reason (epist. 121.3; cf. 104.23). See now Bees 2004: 46–74. 33 This is the Stoic theory of diastroph¯ e (perversio): cf. SVF III 228–236. Seneca was familiar with this doctrine: cf. epist. 94.52–58, with the commentary of Bellincioni 1979: 194–201. 34 Epist. 31.9: (natura) dedit tibi illa quae si non deserueris, par deo surges; 49.11: rationem […] imperfectam, sed quae perfici posset; 76.10: quid est in homine proprium? Ratio: haec recta et consummata felicitatem hominis implevit […] si hanc perfecit laudabilis est et finem naturae suae tetigit; 92.27: ratio vero dis hominibusque communis est: haec in illis consummata est, in nobis consummabilis; 92.30: capax est noster animus, perfertur illo si vitia non deprimant. For Seneca the possibility for man to attain wisdom is guaranteed by his divine and heavenly origin. Cf. Setaioli 2006–2007: 363 f. 35 Dial. 2 (= const.).7.1: non fingimus istud humani ingenii vanum decus nec ingentem imaginem falsae rei concipimus, sed qualem conformamus exhibuimus, exhibebimus, raro forsitan magnisque aetatium intervallis unum. Cf. 2.1. 36 Epist. 42.1: scis quem nunc virum bonum dicam? hunc secundae notae; nam ille alter fortasse tamquam phoenix semel anno quingentesimo nascitur. For the influence of Panaetius on Seneca, cf. Setaioli 2000: 130–139, 165–168, 180–182, 187–191. 37 For the importance of the figure of the sapiens in Seneca, see, e.g., Ganss 1952. 38 Epist. 16.1: liquere hoc tibi, Lucili, scio, neminem posse beate vivere sine sapientiae studio, et beatam vitam perfecta sapientia effici, ceterum tolerabilem etiam inchoata. Cf. Hengelbrock

244

aldo setaioli

So, although spiritual therapy does not by definition address the sapientes, the ideal of perfect wisdom will constantly be in the background. We possess no writing of Seneca’s addressed to someone who has not already been “converted” and persuaded to submit to his therapy and embark on the long journey to Stoic wisdom.39 Seneca does in fact maintain that it is the duty of the therapist to attempt treatment of even seemingly desperate cases before giving up;40 but generally speaking his work addresses people who, like himself, are trying to progress toward virtue and wisdom, i.e., Panaetius’s prokoptontes, or, as he says, proficientes. These may be at different levels or degrees of progress,41 but we must not think that below the sapiens there is only vacuum.42 3. Once the scope and purpose of Seneca’s therapy has been clarified, we are ready to investigate the instruments he uses to attain his goal. These are predominantly verbal. In the first stages he resorts to the rhetoric of the admonitio43 (which, as we saw, can be aggressive in tone), then to the technique of the meditatio, although the latter, as we shall see, is accompanied by practical, or partly practical, “exercises.” Though in both cases people still a long way from the goal are addressed, the meditatio, as well as the “exercises,” 2000: 103–111. Hengelbrock, however, is wrong when he claims that Seneca is inconsistent in placing the idea of perfect wisdom side by side with his promotion of moral progress. A similar mistake is found in Rist 1989: 2012: “at times he verges on optimistic heterodoxy, for the possibility of being a sage is less remote for the practical Seneca than for his Greek masters.” In reality, even Cleanthes and Chrysippus believed that virtue could be taught (SVF II 567, III 223). 39 We shall briefly mention later the alleged Epicurean leanings of Lucilius, at least at the beginning of the correspondence (cf. below, n. 48). What is important here is the pride Seneca takes in Lucilius’s moral progress under his direction (epist. 34.2). 40 Epist. 29.3: certum petat, eligat profecturos, ab iis quos desperavit recedat, non tamen cito relinquat et in ipsa desperatione extrema remedia temptet. Cf. 50.6. Seneca does admit, however, that in some cases no therapy has an effect: epist. 94.24, 31, clem. 1.2.2. Cf. dial. 8 (= de otio).3.3. 41 Cf. next note. At Epistle 25.1 Seneca says of two friends: alterius vitia emendanda, alterius frangenda sunt. But here he reaffirms that an effort must be made even in seemingly desperate cases: epist. 25.2: nec desperaveris etiam diutinos aegros posse sanari; 25.3: inpendam huic rei dies et utrum possit aliquid agi an non possit experiar. In some cases the proficiens may have already greatly advanced, like Serenus, the addressee of the De tranquillitate animi. 42 Epist. 75.8: “quid ergo? infra illum nulli gradus sunt? statim a sapientia praeceps est?” Non, ut existimo. Seneca goes on to establish a triple division of the proficientes (epist. 75.8–18). Cf. also epist. 52.3–7. 43 For Seneca’s admonitio and the theory of its style, see Setaioli 2000: 111–126, 141–155, where the main texts and the relevant bibliography are quoted and discussed.

ethics i: therapy, self-transformation, “lebensform”

245

can at least be recommended to people who have already decided to undertake their own reformation, whereas the rhetoric of admonishment— though necessary until the final stage is reached, i.e., until unemotional discourse addressing reason can be used—may also be directed to reluctant addressees who must still be “converted,” i.e., persuaded to embark on ethical improvement. The prose of Seneca’s admonitio will therefore resort to the resources of rhetoric and will not address the listener’s or reader’s reason, but rather play on their emotions. 44 In this connection Seneca admirably avails himself of his rhetorical background, creating his unmistakable style,45 but we shall see that the fruitful match with his philosophical standpoint allows him to proceed well beyond contemporary school rhetoric in a surprisingly “modern” direction. It might seem that in addressing the emotional rather than the rational sphere Seneca would run counter to his school’s positions, but this is definitely not the case. As observed above,46 according to the Stoics, passions, although really arising from a wrong judgment, are insensitive to reason. It is therefore futile to try correcting this judgment through rational arguments. Chrysippus proceeds even further: in the therapy of urgent cases the therapist must avail himself of any “medicine” to which the patient is liable to react, even if it does not correspond to truth, i.e., if it is a tenet of a philosophical school different from Stoicism.47 Incidentally, this is exactly what Seneca often does with Lucilius, when, somewhat surprisingly for a Stoic, he enlists Epicurus under the banner of his ethical admonitio (and meditatio).48

44 This corresponds to Panaetius’s theory of the sermo, as reported by Cic. off. 1.132–137, which covers both Seneca’s admonitio and his sermo (in Seneca’s terminology, the unemotional discourse to be used, as we shall see, at a later stage in the therapy). Although the admonitio may rise to a high pitch (e.g., epist. 60.1: queror, litigo, irascor; cf. epist. 25.1, 51.13), the therapist will stir the patient’s emotions but never lose rational control of himself (e.g., epist. 40.7). Cf. Cic. off. 1.136. The appeal to the addressee’s emotions is only the first step toward the final goal of restoring his reason. Cf. Setaioli 2000: 141–155. 45 We have already seen Seneca’s functional use of rhetoric in his “Consolations.” 46 Cf. n. 15. 47 SVF III 474. Significantly, Chrysippus expressed this idea while writing about the therapy of passions (en t¯oi peri path¯on therapeutik¯oi). He expressly mentions a Peripatetic as well as an Epicurean approach. These can be used to avoid jeopardizing the success of the therapy by losing time with doctrinal refutations. As far as Epicureanism is concerned, he says that the therapist must show the patient that even that doctrine rules out passion. 48 Some have argued that Lucilius’s original leanings were Epicurean (cf. Epicuri tui: epist. 23.9). See Setaioli 1988: 201 n. 866. More recently, Lucilius’s Epicureanism has been denied by Graver 1996: 27 (following Griffin 1976: 350–352), and asserted by Wacht 1998: 528. Chrysippus’s fragment seems to suggest that the idea cannot be lightly discarded.

246

aldo setaioli

Also, we must not forget that Seneca accepts Posidonius’s doctrine of the existence of an irrational element within the human soul.49 In spite of these antecedents in Greek Stoicism, some scholars tend to believe that Seneca sacrifices rational understanding in favor of a merely emotional appeal to make his addressees accept Stoicism.50 This is because the stage of the rhetorical admonitio—which, to be sure, is the most obvious aspect of Seneca’s work—is often mistaken for the whole of his therapeutic program, whereas it is only the first, propaedeutic stage.51 Seneca leaves no doubt about this.52 We shall soon see that the final stage addresses reason, and no longer the emotions. 4. Before we do that, however, we must look at the next therapeutic step suggested by Seneca: what he calls the meditatio.53 This resorts to techniques that are still verbal, but are matched by a series of practical or partly practical 49

Epist. 92.1. Posidonius is quoted at 92.10. Of course, the irrational element must obey the rational one. For our purpose it hardly makes any difference whether the beginning of the letter goes ultimately back to Posidonius, as I think it likely (see Setaioli 1988: 304f.; Setaioli 2000: 298f. n. 126, with the literature quoted and discussed) or his influence is limited to the words reported under his name at § 10, as others maintain. See lastly Setaioli 2007a: 689 n. 3. 50 This happens even in a fairly recent paper: Cooper 2006 (p. 47: “there is a danger— and I will argue that Seneca falls victim to it—that in relying so heavily on these rhetorical, emotion-evoking devices of the spiritual director, a Stoic writer will tend to forget or neglect the fact that the ultimate goal […] is to achieve a full philosophical understanding of the reasons why the truths of Stoicism really are true”; p. 55: “Seneca so completely cuts off the basis on which he is encouraging his addressee to live from the reasons provided by Stoic philosophical theory for living that way, that it becomes highly questionable whether they can be making real progress toward virtue and the fully happy life if they follow him”). Newman 1989 makes a similar mistake concerning the meditatio. See below, n. 66. But Seneca’s obvious annoyance with dialectical niceties does not imply that he recognizes no theoretical foundations to the ideas and behaviors he preaches. 51 Connected with this is the equivocation concerning the place of thepraecepta and the decreta (discussed in Epistles 94 and 95) in the therapeutic program. Some (e.g., Newman 1989: 1484) have maintained that learning general theory (decreta) precedes listening to specific directions (praecepta). Although Seneca is not clear on this point (epist. 95.38, 95.54 vs. epist. 95.64), however, we know that the praecepta generally preceded (cf. Dihle 1962: 92). The style of the admonitio is fit for the praecepta, whereas the decreta require a different type of discourse, the sermo (Setaioli 2000: 118). See below. 52 Epist. 33.6 f., 38.1, 94.43. 53 As already noted, this takes place at a stage later than “conversion”: people who practice meditatio have already agreed to embark on their ethical improvement.

ethics i: therapy, self-transformation, “lebensform”

247

exercises.54 The verbal meditatio can be considered a form of autosuggestion, the practical exercises a form of preparatory asceticism.55 It has become customary to speak of “spiritual exercises” in Seneca and in other ancient philosophers, but it can be dangerous to pair their therapeutic practices with those preached by St. Ignatius of Loyola.56 Whereas admonitio could address all passions, meditatio is basically designed to conquer fear (itself one of the Stoic main passions) by steeling Seneca and his addressees against the dread of future mishaps, such as poverty, sickness, exile, and the like.57 Just like admonitio, it does not yet address the rational sphere in as much as it does not try to correct wrong judgment and prove that all mishaps are really no evils, but only “indifferents” (adiaphora). This would be in line with Stoic theory, which at this stage Seneca’s addressees are not yet expected to have mastered. Actually, in some letters to Lucilius, we see Seneca seemingly adopt Epicurus’s point of view, refusing, like the Greek master does, to worry about what might indeed never happen.58 His more common attitude, however, is quite different: he advises that we must expect that all possible mishaps will indeed happen, in order to be prepared for them all.59 54

As remarked by P. Hadot (1981a: 14), this behavior is “l’œuvre non seulement de la pensée, mais de tout le psychisme de l’ individu.” Seneca’s term meditatio corresponds to the Greek melet¯e; not rarely it is joined by exercere / exerceri / exercitatio (cf. Greek ask¯esis). Cf. Bellincioni 1979: 184 f. 55 Such exercises are rehearsals and must not be confused with the ultimate standards of behavior, which can be acquired only at the end of the therapeutic process. Cf. below, n. 103. 56 This has been done by Rabbow 1954, who is rightly criticized by Newman 1989: 1476 n. 6. Seneca’s final goal was obviously different from Ignatius’s. For the practice, see also I. Hadot 1969. 57 See Armisen-Marchetti 1986: 186–188 and Wacht 1998: 526–528 for the philosophical antecedents of this practice. It had also been adopted by the Stoics (e.g.,SVF III 482, and see Wacht 1998: 528 n. 74). 58 Cf. Cic. Tusc. 3.32: [Epicurus censet] neque vetustate minui mala nec fieri praemeditata leviora, stultamque esse meditationem futuri mali aut fortasse ne futuri quidem: satis esse odiosum malum omne, cum venisset; qui autem semper cogitavisset accidere posse aliquid adversi, ei fieri illud sempiternum malum. Cf. Sen. epist. 13.4: illud tibi praecipio ne sis miser ante tempus, cum illa quae velut inminentia expavisti fortasse numquam ventura sint, certe non venerint; 74.33: quid autem dementius quam angi futuris nec se tormento reservare, sed arcessere sibi miserias et admovere? Here Seneca adopts one of Epicurus’s remedies (avocatio a cogitanda molestia: cf. Cic. Tusc. 3.33) and clearly says he is not following Stoicism (non loquor tecum Stoica lingua: epist. 13.4). In the “Consolations” he recommends Epicurus’s other remedy (revocatio ad contemplandas voluptates: Cic. ibid.):epist. 99.4f., 99.23, dial. 11 (= cons. Pol.).10.3. Cf. Armisen-Marchetti 1986: 188, 191–193, Wacht 1998: 526–529. We have already seen (supra, n. 47) that even Chrysippus admitted of doctrinal incongruity when urgent therapy was needed. A change of attitude appears at epist. 24.1 f. 59 Statements to this effect are almost countless in Seneca. See Armisen-Marchetti 1986,

248

aldo setaioli

Limits of space force us to make a mere mention of the greatest worry to be conquered: fear of death. This would require a paper, or rather a monograph, in its own right.60 Following in the wake of a long tradition, Seneca considers the meditatio mortis to be the most necessary of all meditationes,61 on the one hand because this is the one mishap which is sure to happen, on the other because it is the only one that cannot be enacted or rehearsed in advance. One noteworthy aspect of the verbal meditatio is, in Newman’s words, “the constant and rigorous application of particular phrases and images.”62 This will involve continuous repetition of the same ideas and encourage their ever-varying rhetorical cast63—a characteristic in perfect agreement with Seneca’s own way of writing. We can understand Fronto’s stern criticism,64 even though he clearly misses the connection between Seneca’s repetitious style and his ethical aims. It is perfectly clear that, like the admonitio, the verbal meditatio resorts amply to rhetorical devices addressing the emotions rather than reason.65 It is a sort of autosuggestion, which can be extremely useful from the therapeutic point of view, but can hardly be considered “the sole means for attaining the vita beata” or “the most important part of his (Seneca’s) teaching”;66 the appeal to the emotions is inherently propaedeutic, the final goal being in fact the restoration of reason. What makes the meditatio important is the

Newman 1989, Wacht 1998. As Armisen-Marchetti (1986: 191f.) shows, Seneca is not being inconsistent, but simply postponing the meditatio to the moment in which the proficiens is advanced enough to have gained rational control of the anticipation of future mishaps, rather than be overcome by it. 60 We shall only note that death (as well as exile: cf. epist. 85.41, 91.8) is already addressed in the “Consolations”; a further proof that these do not basically differ from the rest of Seneca’s writing. 61 E.g., epist. 70.18; and other countless passages in Seneca. 62 Newman 1989: 1475. 63 Cf. Newman 1989: 1480, Graver 1996: 130. Seneca makes this very clear, in practice as well as in theory. Cf., e.g., epist. 27.9: hoc saepe dicit Epicurus aliter atque aliter, sed numquam nimis dicitur quod numquam satis discitur; 94.26: quaecumque salutaria sunt saepe agitari debent, saepe versari, ut non tantum nota sint nobis sed etiam parata. Cf. Bellincioni 1979: 159 for similar passages. 64 Fronto ad M. Anton. De orationibus 4 (II, p. 104 Haines): primum illud in isto genere dicendi vitium turpissimum, quod eandem sententiam milliens alio atque alio amictu indutam referunt. 65 Cf. Newman 1989: 1475, 1478, 1488f., 1494. Newman, however, hardly distinguishes between admonitio and meditatio. A clear and convincing picture of the links connecting meditatio with rhetoric is sketched by Armisen-Marchetti 2004–2005. 66 Newman 1989: 1488. Cf. p. 1484 (“the heart and soul of the ethical life”), though there we find an important qualification (“the only means by which the parenetic part of philosophy can be effective”; the emphasis is mine).

ethics i: therapy, self-transformation, “lebensform”

249

fact that it aims not merely at providing a momentary soothing of fear, but at producing a permanent spiritual transformation,67 which will make it possible to advance to the next step of therapy. The practical or partly practical exercises that accompany the verbal meditatio, although also mainly designed to conquer fear, are wider in scope. They may address the opposite passion, too, namely desire; 68 they may also have other goals: for example, learning to concentrate in an unfavorable surrounding,69 or to alternate company and isolation in view of personal ethical advancement.70 Ultimately, they aim to achieve the permanent transformation of bona voluntas into bona mens and of impetus into habitus animi.71 Partly practical exercises include self-scrutiny and the morally profitable use of imagination. Seneca’s investigation of the self has been examined in depth by Alfonso Traina72 and in a more summary fashion by Michel Foucault.73 Here, we will only mention self-scrutiny as a “spiritual exercise.”74 Seneca both practices this on a daily basis75 and urges others to do so.76 Unlike other exercises, self-scrutiny looks back to the past, rather than forward to the future; but only by knowledge of the self will it be possible to become better77 and to act effectively to help others, now and in the future. As Traina has aptly pointed out, Seneca’s thought and language are in constant swing from the inner to the outer world, and vice versa. Self-scrutiny and self-transformation proceed at the same rate;78 actually the idea of self-transformation is most impressively formulated precisely in this connection: intellego, Lucili, non

As already seen in the “Consolations.” Cf. epist. 123.3: debemus exerceri ne haec timeamus, ne illa cupiamus. 69 Cf. the lively description in epist. 56. 70 E.g., dial. 9 (= tranq.).17.3; but more frequently Seneca urges isolation from the crowd, whose influence can easily nullify the progress of the proficiens. 71 Cf. epist. 16.1 and 6 respectively. 72 Traina 1974: 9–23. Among later studies we may mention Lotito 2001. 73 Foucault 1986: 53 f. 74 Cf. Edwards 1997. 75 Dial. 5 (= de ira 3).36, epist. 83.2. The practice originated in Pythagoreanism: cf. Cic. Cato 38; at least from a certain stage on, it was not meant to train memory (as maintained by Inwood 2005a: 343), but to improve the self: cf. carm. aur. 40–44, Hierocl. in carm. aur. 19, pp. 79–84 Koehler. 76 Epist. 16.2, 28.10, 68.6. 77 Epist. 28.8: “initium est salutis notitia peccati.” egregie mihi hoc dixisse videtur Epicurus; nam qui peccare se nescit corrigi non vult; deprehendas te oportet antequam emendes. Cf. Setaioli 1988: 220 f. for the peculiarly Senecan twist Epicurus’s sentence receives in the context. 78 Cf. Edwards 1997: 31. See Traina 1974: 41 for Seneca’s “linguaggio dell’interiorità” and “linguaggio della predicazione.” 68

250

aldo setaioli

emendari me tantum sed transfigurari.79 This verb, the Latin rendering of metasch¯ematizesthai,80 is followed by other expressions implying change and improvement,81 as well as by the ubiquitous medical metaphor.82 At this stage, however, the process is far from being concluded: Seneca has changed, but he is still far from the goal; wisdom has not yet been attained: he finds that there is still much in him that needs changing.83 Self-scrutiny may go beyond opening the proficiens’ eyes and actually initiate his reformation, but it can hardly bring it to conclusion. An exercise involving the use of imagination consists in fancying that our behavior is watched by some highly respected and ethically irreproachable figure.84 This obviously implies that the proficiens is still far from being able to stand on his own; he must be well on his way to spiritual sanity before he can dispense with such a tutor.85 Among the exercises involving practical behavior86 the most striking is the rehearsal of an undesirable situation, which those who had not assimilated Stoic theory feared and wrongly considered to be an evil, namely poverty. It was a form of asceticism87 which in Seneca’s times had lost much of its original meaning and had become a form of ostentatious snobbishness.88 79

Epist. 6.1. Cf. Bickel 1957; also the clarifications and remarks about the Greek word and Seneca’s use of transfigurare in Setaioli 1988: 283–285. 81 Epist. 6.1: hoc ipsum argumentum est in melius translati animi quod vitia sua quae adhuc ignorabat videt […] 2: cuperem itaque tecum communicare tam subitam mutationem mei. 82 Epist. 6.1: quibusdam aegris gratulatio fit cum ipsi aegros se esse senserunt. 83 Epist. 6.1: nec hoc promitto aut spero, nihil in me superesse quod mutandum sit. It may be interesting, from the linguistic point of view, to compare the present passage (intellego […] me […] transfigurari: epist. 6.1) with epist. 94.48: nondum sapiens est nisi in ea quae didicit animus eius transfiguratus est (reporting Ariston’s thought). Whereas transfigurari is an infectum, describing an event still in progress, transfiguratus est is a perfectum, indicating that the process is completed, i.e., the goal has been reached. 84 This is again an idea that Seneca borrows from Epicurus: epist. 11.8, 25.5f. Cf. Setaioli 1988: 195f.; for Seneca’s Roman adaptations cf. Setaioli 2003: 60f. See also epist. 104.21f. At epist. 32.1 Seneca suggests himself acting as Lucilius’s absent tutor (as Epicurus—and Zeno—had done before him). 85 Epist. 25.6: cum iam profeceris tantum ut sit tibi etiam tui reverentia, licebit dimittas paedagogum. 86 For other types of practical exercises see supra (notes 68–70). 87 Seneca had adopted several ascetic practices, including vegetarianism, in the enthusiasm of his youthful conversion to Pythagoreanism, but, as he tells us, had retained just a few of them: epist. 108.15f.; cf. 53.3, 83.5, 92.25. For the meaning and import ofask¯esis in the ancient world see the bibliography quoted by Allegri 2004: 13 n. 1. 88 Epist. 18.7: non est nunc quod existimes me dicere Timoneas cenas et pauperum cellas et quidquid aliud est per quod luxuria divitiarum taedio ludit; 100.6: desit sane varietas marmorum […] et pauperis cella et quidquid aliud luxuria […] miscet. 80

ethics i: therapy, self-transformation, “lebensform”

251

But Seneca is extremely serious about it. He advises exercises in voluntary poverty to Lucilius89 and practices them himself.90 Poverty is certainly not necessary to wisdom;91 what is important in this exercise is the shift from a form of defense against a dreaded mishap to the actual rehearsal of the self-sufficiency (autarkeia) of the wise man.92 But although this exercise, too, aims at ethical improvement, it is still a long way from the final goal. That poverty is no evil is not proved by theoretical reasoning exposing it as a mere “indifferent,” but only through practical rehearsal. “Is this, then, what I feared?”93 So Lucilius will say after submitting to this experience. And Seneca himself will learn through this very exercise that he still has a long way to go.94 In his own words, his impetus has not yet become a habitus animi.95 5. In the ambitious attempt to bring Seneca’s therapeutic project to its final goal, it will be necessary to resort once more to verbal tools; but it will be a type of discourse very different from the rhetoric appealing to the emotions, which Seneca promotes in theory and in practice for his ethical admonitio (and also for verbal meditatio). A clear theoretical sketch is provided by Seneca in Epistle 38. The admonitio needs oratorical forms (disputationes, contiones) in as much as it aims to convert, or at any rate to sway the addressee’s mind;96 but when it comes to actually instructing him in the theoretical foundations of wisdom, what is needed is the sermo,97 which, in Seneca’s own words, is concerned with res, not with verba;98 is addressed to reason; and dispenses 89

Epist. 18.5 f., 20.13. Epist. 87. For an in-depth analysis of this epistle, see Allegri 2004. 91 Seneca is a Stoic, not a Cynic: cf. Scarpat 1975: 92f., Goulet-Cazé 1986: 185f., Rist 1989: 1994. 92 Cf. Allegri 2004: 19, 25. For Seneca, e.g., epist. 20.8. The autarkeia, however, is still merely rehearsed, not yet theoretically grasped and acquired. 93 Epist. 18.5: hoc est quod timebatur? 94 Epist. 87.5: parum adhuc profeci: nondum audeo frugalitatem palam ferre; etiamnunc curo opiniones viatorum. See the analysis by Allegri 2004: 34–42. Just as Seneca addresses consolations to others but is unable to console himself (cf. supra, n. 31), so he advises Lucilius to undertake exercises in poverty by which he is himself exposed as still very far from perfection. Even the exercise of Epistle 56 ends in failure. 95 Cf. epist. 16.6 (supra, n. 71). 96 Such oratorical forms are indeed necessary ubi qui dubitat inpellendus est (epist. 38.1); cf. epist. 87.41. 97 The sermo is indeed necessary ubi […] non hoc agendum est, ut velit discere, sed ut discat (epist. 38.1). 98 Cf. epist. 40.14, 52.14, 59.4 f., 75.7, 100.10, 108.6 f., 115.1, dial. 9 (= tranq.).1.13. 90

252

aldo setaioli

with any appeal to the emotions.99 For Seneca, the philosopher must be able to command both types of discourse.100 Although most of Seneca’s work does indeed keep within the earlier therapeutic stages sketched above, and although whatever theoretical elements can be detected in it are anything but systematic, we cannot overlook the fact that he does envisage a further stage during which the philosophical basis of wisdom will be taught and learned. Although at times he tends to move even the decreta—the theoretical principles of ethics—into the sphere of the praecepta—the practical instructions101—it is nevertheless clear that the final goal of his therapy can be reached only when the patient has mastered and assimilated the philosophical principles. This has not happened in the previous stages, and for this reason the praecepta given at that time are often ineffective.102 Only when action is guided by theoretical knowledge can the level of ethical perfection marked by virtue be attained.103 The final stage of the studium sapientiae coincides with the acquisition of philosophical awareness. 6. Our survey of the stages and procedures in the therapy that aim to restore sanity and effect a (self-)reformation enabling the attainment of wisdom and virtue would not be complete without at least hinting at a very important 99

More on this will be found in Setaioli 2000: 111–217 (esp. 116–120). See also supra, n. 44. Cf. Setaioli 2000: 118 f. 101 Epist. 94.31: quid enim interest inter decreta philosophiae et praecepta nisi quod illa generalia praecepta sunt, haec specialia?; 95.34: in hac ergo morum perversitate desideratur solito vehementius aliquid quod mala inveterata discutiat: decretis agendum est ut evellatur penitus falsorum recepta persuasio. Note the “agonistic” ring of the latter passage, which is unexpected in reference to the decreta, allegedly suited to be expressed by the unemotional sermo rather than by the pugnacious admonitio. It is nevertheless apparent that in the last epistles problems of theory tend to become more and more important. Lucilius, by now, appears to be past the stage of admonitio and ready for proper philosophical instruction; to use the terms of Epistle 38.1, Seneca is no more addressing him ut velit discere, but rather ut discat. Cf. Setaioli, supra, pp. 191–200. 102 The clearest passage is perhaps epist. 95.37f.; cf. 94.6–8. The latter passage reports Ariston’s thought, which Seneca later rebuts, but only as far as the alleged uselessness of praecepta is concerned. Ariston’s central point—the need to correct the wrong opinion, which considers the adiaphora to be real goods or evils—is surely shared by Seneca. 103 Epist. 90.45: virtus non contingit animo nisi instituto et edocto et ad summum adsidua exercitatione perducto; 94.47: pars virtutis disciplina constat, pars exercitatione; et discas oportet et quod didicisti agendo confirmes. In these passages the exercitatio is not preparatory rehearsal but real practical application—though still tentative at first—of philosophical principles securely assimilated. Cf. Bellincioni 1979: 186f. In the second passage Seneca includes the praecepta in these principles. 100

ethics i: therapy, self-transformation, “lebensform”

253

instrument the proficiens can avail himself of, namely books and reading104 (writing being the reverse of the coin). The centuries that had elapsed since the famous myth of Theuth in Plato’s Phaedrus, disparaging writing in favor of lively word of mouth,105 had obviously not affected its appeal; an attitude wary of books and reading (as well as writing) is still apparent in Epictetus.106 Seneca, however, was not a freedman delivering public lectures in Greek, but a member of the Roman aristocracy addressing his social peers in Latin;107 so his teaching must be institutionally committed to the written word.108 As a possible rejoinder to Plato, Seneca stresses the latter’s superiority over oral instruction: writing (in the case in hand, letter writing) will not allow for offering advice for an immediate situation, but it is the proper vehicle for universal teaching, valid for everyone, including posterity109—which, as Seneca repeatedly says, is his ultimate addressee.110 As Lucilius progresses under Seneca’s guidance, he begins to help Seneca progress too.111 At the same rate, he is not merely the addressee of Seneca’s books:112 it is probably no coincidence that two consecutive epistles picture Seneca and Lucilius exchanging books of philosophy written by themselves.113 104

Graver 1996 devoted a brilliant dissertation to this topic, unfortunately taking no notice of my essay on Seneca’s theories of style and literary activity (Setaioli 1985; cf. below, n. 122). Cf. also Guglielmo 1997. 105 Plat. Phaedr. 274c–278b. 106 Epict. diss. 1.26.16, 4.4.2–18, 30, 33, 40–41. Cf. Graver 1996: 3f., 6, 50–56 (on p. 52 Crates of Mallos is mistakenly substituted for Crates of Thebes). Epictetus does admit of reading if it aims at ethical improvement, but appears to be much more narrow-minded than Seneca in this respect. See below. 107 The entire Epistle 40 is instructive of Seneca’s attitude toward philosophical lectures in Greek. Cf. Setaioli 1988: 14 f. 108 Epist. 33.9: quid est quare audiam quod legere possum? “Multum” inquit “viva vox facit.” Non quidem haec quae alienis verbis commodatur et actuari vice fungitur. This, of course, must be taken with a grain of salt: cf. epist. 6.5f. In Epistle 33, however, we witness a neat reversal of the positions of Epistle 6. As we shall see, Epistle 33 advocates the reading of philosophical works in their entirety—a far cry from Epistle 6.5: mittam itaque ipsos tibi libros, et ne multum operae inpendas dum passim profutura sectaris, inponam notas, ut ad ipsa quae probo et miror accedas. Epistle 33 obviously marks a more advanced stage. Cf. epist. 39.1. 109 Epist. 22.1 f. 110 Epist. 8.2, 21.5, 22.2, 64.7. Cf. 79.17. 111 Epist. 34.2. 112 As is the case with the preserved Epistulae morales, De providentia, and Naturales quaestiones. Cf. also epist. 106.1–3, 108.1. 113 In Epistle 45 Seneca sends Lucilius books, including some by himself (libros meos: 45.3); in Epistle 46 he receives a book by Lucilius. The content is surely philosophical (fecit aliquid et materia: 46.2; cf. epist. 75.3, dial. 9 [= tranq.].1.14).

254

aldo setaioli

Reading and writing not only promote the meditatio, but can be counted among the exercises helping us advance on the long path to wisdom.114 But reading has a place in more than one stage of Seneca’s therapy.115 One application appearing from the very beginning of the collection of the Epistulae morales easily fits into the frame of the meditatio. Epistle 2116 stresses the function of reading as nourishment for the soul, but also insists on narrowing down the range of reading material to a small number of trustworthy authors to be read over and over again,117 in order to glean some chosen thoughts to assimilate in the manner of the meditatio. At the end of the epistle Seneca offers Lucilius, as an example, a sentence by Epicurus, and will do the same in most of the following twenty-seven letters.118 This, however, is far from being Seneca’s final word on reading. 119 A more open position is already sketched in Epistle 33, shortly after discontinuing appending (mostly Epicurean) sentences to the letters. This epistle stresses the need to read texts (significantly, of Stoic philosophers) in their entirety, while at the same time urging originality within tradition. This idea is fully developed in Epistle 84,120 which takes the cue from the rhetorical theory of imitation121 to develop a strikingly “modern” conception of cultural training and education. 114

Cf., e.g., epist. 89.23: haec aliis dic, ut dum dicis audias ipse, scribe ut dum scribis legas. Interestingly, Lucilius, the recipient of Seneca’s admonitio, is in turn urged to address spoken and written admonitio to others as a way of assuring his own ethical progress. No wonder Seneca never leaves his writing tablets (epist. 87.3), and states (epist. 82.3) that otium sine litteris mors est et hominis vivi sepultura. See below, n. 132, for the topos of the mortua vita. 115 The extreme statement of Epistle 88.32 (quid est autem, quare existimem non futurum sapientem eum, qui litteras nescit, cum sapientia non sit in litteris?), which not surprisingly parallels a Cynic position (Antisthenes: Diog. Laert. 6.103; cf. Stückelberger 1965: 132), must be seen in the context of the letter. Cf. Setaioli 1988: 316–322. 116 See the analyses of von Albrecht 2004: 24–33 and Graver 1996: 125–131. 117 The dangers of uncontrolled reading are stressed elsewhere, too: e.g., epist. 45.1, 88.36–40, 89.18, dial. 9 (= tranq.).9.4–6, dial. 10 (= brev.).13.1–7, and epist. 106.11, the famous non vitae sed scholae discimus. Epist. 108.24–34 describes different approaches to reading. The only valid one is the “philosopher’s,” in as much as he reads for the sake of moral improvement. 118 Of course in most cases we are able to ascertain that Seneca found these sentences already collected in a gnomologion. Cf. Setaioli 1988: 182–223. 119 The transition from the position of Epistle 2 to the one developed in Epistle 84 is not abrupt, however. In both cases, for instance, the teachings drawn from reading are nourishment for the mind (2.2, 84.1), and must be “digested” (2.4, 84.6–7). 120 Continuity between Epistle 33 and Epistle 84 is emphasized at the level of linguistic expression, too: epist. 33.8: aliud tamen est meminisse, aliud scire eqs.; 84.7: concoquamus illa; alioqui in memoriam ibunt, non in ingenium. 121 Seneca was familiar with the rhetorical theory of imitatio / aemulatio, as shown by Epistle 79. Cf. Setaioli 2000: 199 f. He had even developed a compromise between aemulatio and the

ethics i: therapy, self-transformation, “lebensform”

255

Limits of space do not allow for further detail.122 It is perfectly clear, however, that at this stage reading is a means of self-transformation that transcends the preparatory process of meditatio and equips the proficiens with tools that will grant him intellectual independence and ethical autonomy, and will lead him to the final goal of his journey toward wisdom and virtue. 7. When the long process we have described has come to an end, the Stoic telos, the summum bonum, the restoration of harmony with Nature and thelogos, which is identical with happiness, wisdom, and virtue, has been reached. 123 That is why life must not be frittered away in useless activities,124 but totally and constantly devoted to self-improvement in view of this goal. Without the studium sapientiae life is an inextricable labyrinth.125 Only there can we find the North Star to direct us surely and safely toward the goal.126 And, as we have seen, life will become tolerable immediately after we have embarked on this journey.127 What is good is not living, but living well;128 this implies that once the ethical goal is attained, duration in time is irrelevant, and that death (suicide) must be accepted as a guarantee of freedom.129 sort of inferiority complex common at his time toward the great classics. Cf. Setaioli 2000: 201–205. 122 I must refer to Setaioli 2000: 206–215. We shall only point out that Seneca reverses the traditional metaphor describing model and imitator. The latter was no more than a blank waxed tablet receiving the model’s imprint (Isocr. adv. soph. 18, Dion. Hal. de imit. frg. 3, II, p. 200, 22f. U.–R., Dinarch. 8, I, p. 308, 10f., Theon prog. 2, p. 61, 30f. Sp.), but in Seneca he prints his personal seal on the material he receives from the models (epist. 84.8; cf. 115.1). And whereas for the rhetoricians reading is the nourishment of style (Theon prog. 2, p. 61, 28f.), for Seneca it feeds the ingenium (epist. 84.1). Also, he changes the meaning of the traditional simile equating “imitators” to honeybees being nearly the only ancient authority who emphasizes the bees’ active contribution in the production of honey (epist. 84.5). Cf. Setaioli 2000: 209 and n. 466. The results of my essay (Setaioli 2000: 111–217—updated on pp. 397–408—first published as Setaioli 1985, but submitted for publication in 1974) find confirmation in the summary, but pertinent remarks to be found in Foucault 1983: 11–13. 123 Cf., e.g., dial. 7 (= vit. beat.).8.2: idem est ergo beate vivere et secundum naturam. 124 Cf. epist. 32.2. 125 Epist. 44.7. 126 Epist. 71.2–4. 127 Epist. 16.1; cf. supra, n. 38. 128 Epist. 70.4: quae [scil. vita], ut scis, non semper retinenda est; non enim vivere bonum est, sed bene vivere; 101.15: quam bene vivas referre, non quam diu; saepe autem in hoc esse bene, ne diu. 129 Statements to both effects are countless in Seneca. For the idea of life’s completeness

256

aldo setaioli

To express the fullness of life identified with the attainment of wisdom and virtue, Seneca contrives an idiosyncratic twist of the pregnant meaning that vivere and vita received in common speech when they were used to express the enjoyment of life.130 Quite possibly he intentionally opposes his ethical and philosophical conception of a life fully lived to the materialistic one of contemporary pleasure seekers.131 For Seneca, a life not devoted to the studium sapientiae does not differ from death.132 Wisdom is an art:133 and although it is different from all others, and is actually the only real art,134 it rests on principles that can be learned.135 Surely, only very few will be able to learn the art of living;136 only the sapiens is the real “artist of life,” artifex vitae.137 We cannot flatter ourselves that we will be able to become like him; but we must live our whole life never losing sight of this ideal goal.

once the ethical goal is reached, cf., e.g., epist. 32.3, 40.10, 77.4, 93.2f., 101.8, etc. The passages on suicide are too numerous to be recorded here. 130 Cf. Setaioli 1988: 273–284. 131 Cf. epist. 123.10, where he parodies the meaning given to vivere and vita by Trimalchio and his peers. Cf. Setaioli 2004: 57–59. 132 Cf. epist. 55.3, 60.4, 77.18, 93.2–4, dial. 9 (= tranq.).5.5; cf. also epist. 82.3 (supra, n. 114). For the topos of the mortua vita, cf. Setaioli 2000: 314 n. 218. At dial. 8 (= de otio).7.1 Seneca tries to unify the three traditional lifestyles (bios phil¯edonos, philosophos, philotimos) under the common denominator of philosophical contemplatio. 133 Epist. 29.3: sapientia ars est; 90.44: ars est bonum fieri. 134 Cf. the whole of epist. 88. 135 Cf. Bellincioni 1979: 230 (on epist. 95.7: haec [sapientia] ars vitae est). 136 Cf. epist. 77.18: vivere vis: scis enim? 137 Dial. 7 (= vit. beat.).8.3, epist. 90.27, 95.7. Cf. Kuen 1994: 136.

ETHICS II: ACTION AND EMOTION

Margaret R. Graver Therapeutic writing cannot help but delve into theoretical psychology. Any effort to influence a reader’s behavioral dispositions will depend on certain working assumptions about how motivation works and how it interacts with belief. This is particularly true if a major concern of the therapeutic enterprise is to manage or eliminate the emotions of common experience. It was likely the practical exigencies of composing discourses to combat anger, grief, and fear that stirred Seneca’s interest in the psychological underpinnings of ethically significant action and emotion. That interest remained strong throughout his career as a prose writer and can be traced in explicit assertions that he makes in his own voice in a number of different works. These assertions are of two kinds: 1) descriptive claims about how we come to initiate different kinds of behavior, how our actions connect to our beliefs about what is best for us, and why our emotions often seem to override our better judgement; and 2) normative claims about behavior and inner experience within that idealized version of human existence which is the goal of personal development. These two philosophical agendas are linked by an integrated conception of human nature. Seneca assumes as a rule that a benevolent Nature has designed the human psyche to function in a rational way. While it is true that in our present state that rational nature is imperfect and subject to grave error, we are also capable of self-correction, and for that reason we can aspire to fulfill our human potential in lives of virtue and wisdom. In nearly every case Seneca indicates that his positions are not of his own devising but are those of the Stoic school, Stoici nostri or just nostri. These statements should be assessed with care, both to improve our knowledge of the history of the school and to understand Seneca’s own working methods. We need to be able to compare what we find in his works with the Stoic positions that were already in existence, ones that he might have learned from his teachers in philosophy or studied in treatises available at Rome. Only on the basis of such comparisons can we make any informed judgment about his relation to the ethical psychology of Chrysippus and other major Stoic philosophers—whether he is a deeply knowledgeable and orthodox exponent of early Stoic thought, an independent-minded innovator who

258

margaret r. graver

molds school doctrine to his own taste, or an eclectic who combines Stoic ideas with elements of other philosophical traditions. Unfortunately, the surviving evidence does not permit us to recover all of what was written and said by Stoic philosophers in the Hellenistic period. We do not have even one of the early treatises in anything like its original form, and we often cannot say with any certainty which author developed a particular line of argumentation and when. We do, however, have sufficient information from quoted fragments and from reliable doxographies to identify at least the major psychological assertions of the most influential Stoic founders. Also, we can draw upon the philosophical writings of Cicero for evidence of the reception of Greek ideas at Rome, whether or not we think that Cicero is likely to have been Seneca’s immediate source.1 That is what I propose to do here. I will first offer a brief summary account of Stoic ethical psychology in the period before Seneca, drawing on the types of material just named and considering first the springs of action generally and then the emotions and other affective responses.2 Once that account is in place, I will proceed to examine Seneca’s own handling of those same points, again beginning with the theory of non-emotive action and proceeding to his views on anger and other emotions. It is, I think, in observing the way he handles the existing doctrine, his characteristic emphases and manner of presentation, that we come closest to identifying a specifically Senecan set of views on these important topics. The Stoic Background: Thought, Belief, and Action A minimum requirement for a workable theory of action is that the animate being should have some way of registering facts or potential facts about the

1 For fuller information on these portions of Stoic doctrine and guidance through much of the surviving evidence, consult Inwood 1985, Long and Sedley 1987, Inwood and Donini 1999, Long 1999, Brennan 2003 and 2005. In what follows I cite source materials by their numbers in Long and Sedley (LS) wherever possible; the translations, however, are my own throughout. I also supply fuller details on most points covered in this chapter in Graver 2007, with additional pointers to the secondary literature. 2 Our word “emotion” is the nearest match in contemporary English usage for Seneca’s term adfectus (representing Greek πάθος). I do not use the older term “passion,” still favored by some interpreters, because that might be taken to imply that these Stoic theories were concerned only with the most intense and damaging emotions. In fact, they addressed all levels of emotional behavior, though as we shall see, other affective phenomena were also mentioned, both involuntary feelings and the “eupathic” responses of the wise.

ethics ii: action and emotion

259

environment, as that it is day or that danger may be approaching. The Stoics, like Aristotle, met this need by positing as an essential function of the psyche a capacity for “impression” (φαντασία).3 Since they conceived of the psyche as a material substance, an impression was for them a momentary modification in that mind-material, comparable to an imprint in wax. This material change corresponds in some way to the immaterial proposition one has in mind. For instance, one might, as part of perceiving some external object through the senses, form the impression, “That is a horse.”4 Other impressions arise from within, by combining and recombining items from the mind’s existing stock of concepts. Either way, the impression, by itself, is what one might call a mere thought. It has content of which one is aware, but no belief is formed unless one also comes to think of the relevant proposition as being either true or false. That further step was termed “assent” (συγκατάθεσις). Assent is conditioned on the character of the mind assenting: a mind characterized by tensile strength gives “strong” assent; weak minds give “weak” assent.5 But the likelihood and suitability of giving assent might also be expressed as a characteristic of the impression itself. Some impressions were said to be not only true but particularly clear or “graspable,” and these, when they gain assent, become the building blocks of knowledge in the person of perfect understanding.6 By contrast, a deranged or hallucinatory person assents to “empty” impressions.7 Much of the material on impressions is epistemological in interest, concerned with the nature of representation and the possibility of certain knowledge. But the basic mechanisms of impression and assent are also essential to the Stoic notion of what action is. In all animals, including humans, behavior is generated in response to impressions. The psychic event in which behavior is initiated is termed “impulse” (ὁρµή); it is this, rather than the resulting movement of the limbs, that is of interest in ethics and psychology. Naturally, the impressions to which we respond are ones we accept as true; even an animal must be assumed to believe certain things about its surroundings if it behaves accordingly.8 At a theoretical level the Stoics even said that an action just is a certain kind of assent. What initiates 3

Diog. Laert. 7.49–51 (LS 39A), Orig., De principiis 3.1.2 f. (LS 53A). Cic. ac. 2.21 (LS 39C). 5 Cic. ac. 1.41 f. (LS 41B); S. Emp. adv. math. 7.151–157 (LS 41C). 6 Cic. ac. 1.40–42 (LS 40B), 2.77 f. (LS 40D). 7 Chrysippus apud Aetius 4.12 (LS 39B). Compare S. Emp. adv. math. 7.243–249. 8 Chrysippus and Antipater apud Plut., De Stoicorum repugnantiis 1057a (LS 53S): “without assent is neither action nor impulse.” 4

260

margaret r. graver

behavior, according to one key source, is “an impulsory impression of what is then and there appropriate.”9 That is, an impulse can always be described as an assent to an impression with the content “it is appropriate for me to do X right now.” For instance, walking (as opposed to sleepwalking, or stumbling forward when pushed) is what happens when one decides, at some level, that walking is the thing to be doing just at that moment. What is it, then, that sets human action apart from the instinctive behaviors of animals? As rational beings, humans experience “rational” impressions, ones whose propositional content they can verbalize when needed. And clearly, they are more reflective than even the most intelligent animals in their manner of giving and withholding assent. Specifically, human beings were said by Stoics to possess “the concept of following,” i.e., of logical consequence.10 While animals can behave in complex ways and must be assumed to perform at least some inferences, only humans recognize the associative thought process itself. Because of this only humans are capable of evaluating and amending the thought processes that underlie their own behavior. It is for this reason that the actions of rational beings are said to be “up to us” while the behaviors of animals and young children are not. This account of action was formulated in such a way as to remain consistent with the school’s position on causal determinism. Each of our actions is determined by identifiable causes, some external to us and some internal: the external causes include impressions arising from our circumstances, while the internal causes are mental characteristics that predispose us to give or withhold assent in various cases.11 At no point did the Stoics see any need to posit a special faculty of will (i.e., an indeterminate freewill), to explain why we act in one way rather than another. For them, we can make choices based upon our character, and we can also make choices that shape that character and its ways of choosing in the future, all within a determinate world-order. In speaking of the human capacity for choice the Hellenistic Stoics seem occasionally to have used the term προαίρεσις (“volition”), for which the best Latin equivalent is voluntas.12 But their use of the term did not express any commitment to libertarianism of the sort that makes some mental events strictly independent of antecedent causation.

Stob. 2.7.9 (86W; LS 53Q); cf. 2.7.9b (LS 33I). S. Emp. adv. math. 8.275 (LS 53T). 11 Cic. fat. 39–43 (LS 62C). For discussion, see Bobzien 1998a. 12 Evidence for προαίρεσις in the early period is reviewed in Voelke 1973: 142f., Dobbin 1991, and Graver 2003; summary in Graver 2007: 233 n. 12. 10

ethics ii: action and emotion

261

The Stoic Background: Emotion and Eupathic Response We come now to the Stoics’ extraordinarily thoughtful analysis of the ordinary emotions.13 In identifying the causes of occurrent emotions (e.g., an angry outburst in relation to a specific event, as opposed to a persistent state of susceptibility to anger), the Stoics referred to the same explanation they gave for actions generally, saying in effect that emotions are actions and that we are responsible for them in the same way as we are responsible for any action. They did not disregard those features that set emotions as a class apart from non-emotive actions like walking across a room, but insisted that those features—in brief, the way emotions feel and their tendency to override our better judgment—can be explained without appeal to extra motivation centers or psychic functions.14 Thus an emotion was, since Zeno, defined as an “excessive impulse.” It is, in the first instance, an impulse to “contract,” “uplift,” “extend,” or otherwise alter the mind-material itself; this alteration is what we feel in the chest region when we are excited or upset. But it may at the same time be an impulse to observable action, as when we strike someone in anger. Like every impulse, an emotion is also an assent to a specifiable propositional content, and that content has a characteristic structure. This is most clearly stated in accounts of the four genus emotions: fear, desire, delight, and distress. Distress, for instance, is caused by a belief that an evil is present, combined with a belief that mental suffering is the appropriate response to such an evil.15 In assenting to the conjunct of these we accept something in the form of an impulsory impression; e.g., that “it is now appropriate for me to suffer”; that is, to contract the psyche. Some treatises must have gone into considerable detail spelling out both phenomenological and intentional accounts of numerous species emotions coming under each of the four genera; these are reflected in the extant doxographical accounts. 13 Source texts are collected in Graver 2002a: 203–223, and Long and Sedley 1987: ch. 65. Recent studies include Tielemann 2003, Nussbaum 2004, Gill 2005, Price 2005, Graver 2007. 14 The Stoics’ entire approach to psychic activity presupposes a single command center that directs the animate being’s response to stimuli. This command center or ἡγεµονικόν is just the physical center of the indwelling πνεῦµα that endows each of us with all our properties, physical and structural properties and basic life-functions as well as our mental functions. One might speak of “parts” of the psyche when distinguishing among the different activities and functions of this important stretch of πνεῦµα, but one would not mean by this what Plato (arguably) and later Platonists (certainly) meant by “parts,” i.e., multiple and competing centers of motivation. See further, notes 37–39 below. 15 Stob. 2.7.10b (90W), Cic. Tusc. 3.25, 61, 79.

262

margaret r. graver

The inclusion of the term “excessive” in the definition requires additional explanation. Chrysippus is known to have devoted extra attention to this point, seeking to clarify how it is that an emotion can occur “through impulse”; i.e., as a species of action, and yet we also feel carried away by it.16 He used the analogy of a person who tries to stop running. Because running is an inherently vigorous activity, one cannot stop just in an instant; there will always be some lag between deciding to stop and being able to stop. During that moment one feels “carried away” by forward momentum, yet we do not say that running is something that occurs in us against our will. The loss of control in emotion is like this. Once the emotional response is initiated, there is a period of time in which we cannot stop ourselves from feeling, and it is still the case that in the moment when it begins, that response occurs through a decision that we make. Consequently, it may be of some use for an advisor to try to influence that decision before the fact by arguing against the evaluative and other beliefs that dispose us in that direction. But it will not be very useful to make those same arguments while the emotion is actually going on, during the period of “inflammation” or emotional flooding. One should still make the attempt, but some arguments will certainly be ineffective.17 Emotions of the ordinary kind were classified as an activity of vice, on the grounds that they could not occur without assent to evaluative propositions that are in reality false. For instance, one does not grieve unless one believes that bereavement is a bad thing: in Stoic ethics, only one’s own moral failings are evil, so that evaluative belief is just a mistake. It follows that the elimination of false belief that comes with the accession of wisdom would entail the elimination of emotions as we know them. The wise person exhibits ἀπάθεια not by suppressing emotions but because in the condition of perfect understanding the judgments that give rise to our ordinary emotions no longer seem correct. The Stoics did not, however, deny that the capacity for affective response is a natural characteristic of the human species. Indeed, they strongly asserted this, claiming that the ideal human condition would include perfected forms of affective response called εὐπάθειαι (“good emotions”).18 These, too, would be the responses of rational beings, experienced through impression and assent. Among the three eupathic genera “joy” involves awareness that a good 16

Galen, De placitis Hippocratis et Platonis 4.2.8–18 (LS 65J), 4.6.24–46 (SVF 3.475). Chrysippus apud Orig., Contra Celsum 8.51 (SVF 3.474). 18 Cic. Tusc. 4.12–15, Diog. Laert. 7.116. Especially helpful discussions of the underlying issue may be found in Striker 1991 and Irwin 1998. 17

ethics ii: action and emotion

263

is present, “wishing” (βούλησις) that a good is in prospect, and “caution” that an evil is in prospect. The goods and evils recognized by the wise person are not externals, however, but features of one’s own character or conduct; for instance, performing a generous action might provide an opportunity for joy. That the εὐπάθειαι have this structure is made clear by the absence of any eupathic genus for present evils: the wise have no occasion for such a response, since by definition no bad activity or trait can be present in them. Conversely, they have abundant occasion for joy, since they are continually exercising the virtues. They can rejoice, too, in the good activities of their friends, regarding these as integral to themselves through a sense of community, and can wish for those goods to continue.19 Further, they can experience erotic love toward young persons who exhibit potential for virtue.20 Both emotions and eupathic responses could also be contrasted with quasi-affective phenomena that do not occur through impulse. Chrysippus was interested in involuntary tears and laughter, which he refers to impressions and to “the beginnings of the circumstances bringing about the movement.”21 And it is clear from several sources that nonprejudicial terms like “biting” were used by Hellenistic Stoics for low-level affective events that do not have the moral significance of emotions proper. Cicero writes, reporting Stoic doctrine, that once the “entirely voluntary belief” that grieving is appropriate is removed, distress itself is eliminated; nonetheless, the mind will still experience “a bite and a small contraction from time to time,” and this is “natural” even in the wise.22 His inclusion of the word “natural” is of interest; it suggests again the view that the capacity to experience certain feelings is accepted by Stoics as part of human nature. Also of interest is his explicit reference to “voluntary belief” in those psychic events that are properly termed emotions. This expression has a close parallel in a Stoicizing passage in Origen, where the Greek term is προαιρετικός.23 It is reasonable to suppose that this language reflects the usage of one or more of the Hellenistic Stoics, designating actual emotions as “voluntary” as one way

Diog. Laert. 7.124 (LS 67P), Stob. 2.7.11i (101 f. W), 2.7.11k (106W), 2.7.11 m (108W), Cic.fin.

3.70. 20

Stob. 2.7.5b9 (65W), Diog. Laert. 7.113, 130. Galen, De placitis Hippocratis et Platonis 4.7.16 f. (LS 65O). Further to the history of the concept, see Graver 1999. 22 Cic. Tusc. 3.83. For the terminology compare Philo, Quaestiones in Genesim 2.57, Plut., De virtute morali 449a. ∆ῆξις was used by Zeno and Chrysippus in the same way as συστολή and ἔπαρσις: Galen, De placitis Hippocratis et Platonis 4.3.2 (LS 65K). 23 Orig. Comm. in Ps. 4.5. For voluntarius in connection with emotion note also Cic. ac. 1.38 (reporting Zeno). 21

264

margaret r. graver

of distinguishing them from unassented quasi-affective reactions, sometimes called προπάθειαι or “pre-emotions.” To say that an emotion, or any action, occurs through assent or voluntarily is to say that its principal cause is some feature of one’s own moral character. But where does character itself come from? Stoic explanations for emotive traits of character like misogyny or irascibility, and more generally for longstanding beliefs about the goodness or badness of certain external objects and the appropriateness of certain responses, made reference to a number of factors. They spoke of cultural influences transmitted by parents and teachers, and Chrysippus also spoke, more obscurely, of the “persuasiveness” of certain impressions.24 But there are also formative influences that come from ourselves. By repeatedly desiring money, for instance, one can cause greed to become part of one’s character, and conversely by exercising one’s rational capacities for self-assessment and correction of view one can prevent such traits from forming.25 The Theory of Action in Seneca Seneca’s concern with impressions, for which his usual term is species, is primarily as they figure in the account of action.26 He does not, for instance, report the Stoic definition of an impression as a mental alteration that “reveals both itself and what made it,” and neither does he enter into the (to us) fascinating Hellenistic debates concerning the mind’s powers of representation. At most he shows, in passing remarks. that he has a basic familiarity with the role of impressions in epistemology. Thus, in Epistle 71.24 he mentions the old example of straight objects that, when seen underwater, give the impression of being curved or bent; the fault, he says, is not in the objects themselves but in us. Another passage mentions the “empty” impressions experienced by the insane: “they are instigated by the impression of some circumstance that the afflicted mind is unable to refute as empty” (dial. 9 [= tranq.].12.5). Even here, though, his emphasis falls on the role of the impression as a stimulus to action more than on the epistemological distinctions that had interested Chrysippus.27 24 Diog. Laert. 7.89; Galen, De placitis Hippocratis et Platonis 5.5.14 (LS 65M). See further Graver 2007: 149–171. Seneca echoes the first part of the explanation in epist. 115.11f. 25 Cic. Tusc. 4.24 f.; compare Epikt. diatr. 2.18.8–10. 26 It should be noted that species in Seneca sometimes means “false seeming” rather than “impression.” 27 Supra, note 7.

ethics ii: action and emotion

265

Again, Seneca is not much inclined to explore his school’s position on impression as a physical event, viz., that it is an alteration in the psyche’s directive faculty (ἡγεµονικόν). That he is nonetheless aware of this aspect of the teaching on impressions appears from some remarks he makes about animals early in De ira. Animals are not capable of anger, he argues, because anger and other emotions come about through certain mental operations of which animals are not capable. The reason they are not capable lies in the nature of animal impressions, which is dependent in turn on the character of an animal’s directive faculty. Just as they have a voice, but an inarticulate voice, muddled and unable to form words; just as they have a tongue, but a fettered tongue that is not free for various movements, so also their directive faculty itself (ipsum principale) is not in the least fine-textured or exact. Thus it receives impressions of things by which it is stimulated to impulses, but those impressions are murky and confused. Hence their attacks and agitations are vehement, but are not cases of fear, worry, sadness, and anger; rather they are just emotion-analogues (his quaedam similia). That is why they fall off quickly and change into their opposites: they go from fierce rage or terror to being fed, from bellowing and running madly about to immediate rest and sleep.28

For any creature, it seems, the quality of its impressions can be inferred from observations of its linguistic capabilities and of its behavior generally. In nonhuman animals, the inability to use language and the changeable behavior Seneca describes give evidence of “murky and confused” impressions; in humans, language use and more persistent behaviors evince a capacity for clearer, more precise impressions. In both cases, theory traces the nature of the impressions back to physical characteristics of the mind. Seneca is also well acquainted with the Stoic analysis of action as originating in assent to a hormetic or action-inducing impression “of what is then and there appropriate.” He spells out the relevant doctrine with particular clarity in Epistle 113.18: Every rational animal does nothing unless (1) it has been stimulated by an impression of some fact, (2) it has then entertained an impulse, and (3) assent has confirmed this impulse. Let me explain what assent is. “It is fitting for me to walk”: I walk only when I have said this to myself and ratified this, my judgment. “It is fitting for me to sit”: then only do I sit.

Here, adsensio is his equivalent for the Greek συγκατάθεσις, and impetus is equivalent to ὁρµή. But before one can actually have an impulse one must, as 28 Dial. 3 (= de ira 1).3.7 f.; cf. epist. 121.12 f. For emotion analogues in animals compare Cic. Tusc. 4.31.

266

margaret r. graver

Seneca puts it, “entertain” that impulse (impetum capere).29 After receiving some sense-impression (e.g., the availability of a comfortable chair), the mind generates a thought in propositional form as to what action might be taken; only if this is ratified does the response occur. To “entertain an impulse” is thus to have in mind a proposition like “It is fitting for me to sit.” This is exactly what the Stoic source in Stobaeus calls an “impulsory impression.”30 Seneca does not attempt to replicate that exact phrase, preferring to remain within a more natural Latin idiom, but he both knows and uses the doctrine it expresses. Further, Seneca states quite clearly the theoretical basis of the Stoic assertion that virtue and knowledge are one; or, putting it another way, that an agent whose rationality has been perfected would act properly on every occasion. Again the key conception is that of a hormetic impression, this time perhaps a “graspable” hormetic impression. In the person of perfect understanding, an impression leading to action is marked by a special clarity: Virtue itself is located in our better part, namely the rational part. What is this virtue? True and unshakeable judgment, for from this come the impulses of the mind, and by it every impression that stimulates impulse is rendered crystal clear (redigetur ad liquidum).31

Knowledge, which for Seneca as for earlier Stoics consists not in any one item of true and justified belief but in an overall condition of harmony within the belief set, guarantees the propriety of further assent, including any assent to impulsory impressions. Thus, the wise are inerrant in behavior as well as in their theoretical judgments. From their point of view it is immediately apparent what the facts of the situation are, and they act accordingly. Assent is all that is needed to make one responsible for action. Seneca does not employ the notion of an uncaused “act of volition” any more than Chrysippus does. Many have however been struck by passages such as the 29 The expression has caused some confusion for interpreters who assume that impetum capere must be equivalent to Greek ὁρµᾶν, e.g., Zöller 2003: 149, Stevens 2000, Rist 1989, Ioppolo 1987: 460, Inwood 1985: 179f., 282. For the Latin usage see OLD s.v. capio 16, and compare epist. 78.2: Saepe impetum cepi abrumpendae vitae: patris me indulgentissimi senectus retinuit. See further, note 41 below. 30 Supra, note 9. 31 Epist. 71.32. For liquidus in connection with the mental experience of the wise compare clem. 2.6: Adice, quod sapiens et providet et in expedito consilium habet; numquam autem liquidum sincerumque ex turbido venit. As Wildberger 2006: 78 observes, Seneca does not use conprehendere or conprehensio in the sense of the Greek καταλαµβάνω / κατάληψις.

ethics ii: action and emotion

267

following, in which willingness (velle or voluntas) emerges as a key ingredient in moral progress.32 Let us press on; let us persevere. Greater challenges lie ahead than those we have overcome. But most of progress consists in being willing to make progress. This I recognize in myself: I am willing—with my entire mind I am willing. (epist. 71.36) Whatever can make you good is within you. What do you need to be good? Willingness. (epist. 80.4)

The “willingness” of which Seneca speaks in these and related passages need not be interpreted as some kind of mysterious faculty for generating uncaused mental events. His point is rather that the attitude one adopts in the present toward moral progress can be expected to make a real difference in one’s future behavior. In effect, one can act now to shape those habits of mind that will determine one’s actions in the future, for instance by carefully observing the habits of some person one admires. There is a discernible continuity between this notion of voluntas and earlier Stoic conceptions of choice or, more specifically, “choice before choice” (προαίρεσις). A similar notion is prominent in the writings of the Greek Stoic Epictetus in the late first century ad.33 The Emotions in Seneca Like other Stoics, Seneca, when speaking in a theoretical vein, treats the emotions of anger, fear, grief, and delight as a special case of action, consisting, like every action, in assent to a hormetic impression. Unlike Cicero, he does not lay out in any systematic manner the elaborate classification system of emotions by genus and species, with definitions for each; at most, he quotes a standard definition here and there as the occasion arises.34 His primary interest is in the psychological basis of emotion and in the possibility that a determined progressor might alter his or her emotional dispositions for the

32 The passages quoted here are among those treated in Inwood 2005a: 135–141; see also Donini 1982: 202–204 and, among older treatments, Voelke 1973 and Pohlenz 1948–1949. The case for Senecan innovation has recently been revived in Zöller 2003; but cf. Smith 2004. Voluntas in Seneca may also mean “intention,” for instance, in Epistles 35.4 and 36.5 and frequently in De Beneficiis with reference to the intent of a person doing a favor. 33 See supra, note 12, with Stob. 2.7.9 (87W). Both Kahn (1988: 255) and Inwood (2005a: 21f.) are struck by the similarity and are inclined to conclude from it, incorrectly in my view, that Seneca is the innovator and has directly influenced Epictetus. 34 Notably in dial. 3 (= de ira 1).2.3 (anger), clem. 2.5 (pity).

268

margaret r. graver

better. He is particularly concerned that his readers understand the difference between involuntary feelings and full-scale responses that involve assent; the latter, but not the former, are subject to amelioration. In the same vein, he is careful to point out that actual emotional responses may be overwhelming in their intensity even though voluntarily initiated. For him as for his Stoic predecessors, the tendency of emotions to run away with us is an important reason to avoid having them in the first place. Hence the first book of De ira goes to great lengths to distinguish Seneca’s own Stoic position from the rival view that he attributes to Aristotle and the Peripatetics, that emotions can be useful and should be moderated rather than eliminated.35 For him, limiting or controlling one’s emotions is not a workable strategy: once one commits oneself to the impulse, there is no additional mental capacity that would enable one to control it. With some things, the beginnings are in our power, but after that they carry us on by their own force, not allowing a return. Bodies allowed to fall from a height have no control of themselves: they cannot resist or delay their downward course, for the irrevocable fall has cut off all deliberation, all repentance; they cannot help but arrive where they are going, though they could have avoided going there at all. Even so the mind, once it propels itself into anger, love, and other emotions, is not permitted to check its impulse. Its own weight and the downward tendency of its faults must carry it to the bottom. It is best to reject the initial stimulus to anger right away, to resist its very seeds and to make every effort not to fall into anger. For if it once begins to carry us off, the return to health is difficult: there is no reasoning once emotion has been let in, once it has been granted some prerogative through our willingness (voluntate nostra).36

Responsibility for whatever is done in anger or another emotion is firmly assigned to the reasoning mind itself. We should not make the mistake of thinking that there is some other motive force within a person, some emotive part of us that wrests control of our motivations away from the faculty of judgment. Seneca means to resist any such splitting of the psyche into opposing camps: for him, it is the mind’s own mistakes that carry it away. The mind is not off by itself, observing the emotions from outside, so as not to allow them to proceed further than they should; rather it is itself changed into the emotion. For that reason it cannot regather its useful and salutary force 35 Dial. 3 (= de ira 1).7 f.; see also 2.35. Setaioli (1988: 141–152) discusses Seneca’s sources in this portion of De ira; see also Fillion-Lahille 1984. 36 Dial. 3 (= de ira 1).7.4–8.1. The imagery should be compared with Chrysippus’s runner analogy, for which see note 16 supra. In Epistle 40.7, Seneca uses a downhill-runner analogy that matches even more closely with that of Chrysippus, but applies it to the orator who lets his eloquence run away with him.

ethics ii: action and emotion

269

after it has been betrayed and weakened. As I said, it does not have its own separate and disjoined location; no, emotion and reason are alterations of the mind for the better and for the worse. (dial. 3 [= de ira 1].8.2f.)

This is very much the unitary psychology of the Stoics. Indeed, his remark about the mind being changed into the emotion closely resembles an account of the same issue that is cited at length by Plutarch with attribution to “Zeno, Chrysippus, and other Stoics.”37 The essential moral insight of the passage is not, however, exclusive to Stoicism, but could be embraced as well by Plato, and indeed by Plato’s later adherents, doctrinaire though they were about psychic tripartition.38 It is simply that rather than allowing emotion to run our lives, we should act in accordance with thoughtful consideration of our long-run best interests. As long as he can maintain this objective, Seneca is not overly concerned about psychic monism for its own sake, but allows himself to speak in ways that suggest bipartition or even tripartition where rhetorical considerations make this advantageous. Even in the above material from De ira he does not hesitate to deploy metaphors of struggle and combat that might seem to imply some kind of division between reasoning and emotive faculties; for instance, he speaks of emotions as an “enemy” to be barricaded outside one’s mental city. Similar figurative language may be found at many places in his work, not only in the tragedies, where strict philosophical propriety is hardly to be expected, but in the prose writings as well. A passage in Epistle 92 reveals how far he was willing to go in this direction. Addressing an interlocutor who favors the idea of competing motivation-centers, Seneca writes in a way that even Plutarch might approve, mentioning a “spirited, ambitious, unruly part located in the affections” and also a “base, idling part devoted to pleasures.”39 37 Virt. Mor. 441cd (LS 61B): “They hold that the emotional [part or power] is not distinguished from the rational by some difference in its nature, but that it is the same part of the mind—I mean that which they call the intellect or directive faculty. During emotions and [other] changes in accordance with a condition or state, this directive faculty is turned and changed throughout its whole, becoming vice and virtue. And it has nothing irrational in itself, but is called “irrational” when it is carried away by the excessiveness of the impulse toward some ill-suited object contrary to reason’s choosing. For emotion, they say, is wicked and uncontrolled reason, which gains additional vehemence and strength through a bad and erroneous judgment.” 38 For the development of psychic monism as an issue in the ancient debate, see Gill 2005. 39 The passage is well discussed in Inwood (2005a: 38–41), who also supplies additional examples of seemingly dualistic figurative language. Unlike Inwood, I am disinclined to think that Seneca posits this psychic division merely for the sake of argument; he is simply unconcerned about that particular debate at this point.

270

margaret r. graver

Yet he does not mean to depart in any serious way from the Chrysippan camp, for shortly thereafter, in Epistle 116, we find him speaking again in terms that resemble the first book of De ira.40 Meanwhile, Seneca does not deny that some of the feelings we experience in the presence of emotive stimuli are indeed involuntary. He makes this important concession at the beginning of De ira Book 2, at which point he also provides a more systematic account of the causes of anger. True anger, he says, is generated only when the mind assents to the impression of injury; it may therefore be distinguished from “that impulse which is stirred involuntarily” and which “follows immediately upon the impression itself.”41 Anger characteristically involves a linking of at least two ideas, as that one ought not to have been wronged and that one ought to take revenge.42 A reaction that does not involve assent does not have this complexity. Numerous examples are given, among them blushing at bad language, excitement while watching a fight, stage fright, and, interestingly, responses to literature and the visual arts. All such he calls “beginnings preliminary to emotion” and says again that they are involuntary, “movements of minds, which do not will to be moved.” 43 Thus, if one perceives oneself as wronged and entertains a desire for revenge, but immediately settles down, one has not experienced anger but only a preliminary to anger (dial. 4 [=de ira 2].3.4 f.). Seneca did not originate the discussion of involuntary feelings within Stoicism: as we have seen, the same point is made in Cicero’s report of the Stoic theory, and there are traces of it elsewhere.44 But he explains the 40 Wacht (1998) rightly emphasizes the continuity between Seneca’s position in De ira 1 and that of Epistles 85 and 116. 41 Dial. 4 (= de ira 2).1.4. The impulse that is stirred involuntarily is an impetus in some lesser sense than the impetus that is genuine anger for which we are responsible (est enim impetus; numquam autem impetus sine adsensu mentis est: de ira 2.3.4). The passage should be compared with Epistle 113.18 cited supra; here, however, we have not only the mind entertaining an impulsory impression that a certain response may be appropriate, but also an observable alteration in the body (such as an increased heart rate) that could reasonably be called a response in its own right and yet stops short of being a volitional response. 42 Compare note 15 supra, and see further Vogt 2006. 43 Dial. 4 (= de ira 2).2.5. The expressions voluntarius and voluntate et iudicio are used repeatedly in this passage in a manner strikingly similar to what we find in Cic. Tusc. 3.64, 3.66, and 3.83. Compare the material cited supra in notes 12 and 17. 44 It is particularly striking that he favors the term “biting” (morsus) for involuntary feelings of mental pain, sometimes pairing it with “contraction,” as does Cicero in Tusc. 3.83. Examples include dial. 6 (= cons. Marc.).7.1, 1 (= prov.).4.1, 9 (= tranq.).1.9, epist. 99.14f. We should not, however, assume that Seneca derives this terminology from Cicero himself, since Philo of Alexandria makes the same pairing in Quaestiones in Genesim 2.57. It must derive from the earlier Stoic writings in Greek.

ethics ii: action and emotion

271

received doctrine at much greater length than any other author, and with greater richness of examples, some clearly of his own devising. Indeed, his treatment of the theme is so elaborate as to create certain difficulties of interpretation. The most significant concerns the role of the body in affective response. Seneca says both here and in his later treatment of the theme in the Moral Epistles that the involuntary feelings are movements “of the body” rather than “of the mind.”45 In view of this one might be tempted to assume that the distinction being drawn is that between a strictly mental event, i.e., the assigning of a truth-value to certain propositions, and the corporeal realization of that event. But this cannot be right, for he also calls the involuntary feelings “movements of the mind,” and some of his examples clearly require full conceptualization of the stimulus and even linguistic processing. In calling such events responses “of the body,” Seneca is merely saying that they occur without exercise of the rational mind’s most characteristic function, that of assent. One portion of the De ira discussion that has often been found puzzling is section 2.4, which speaks not of two mental events, pre-emotion and emotion, but of a sequence of three: Let me tell you how the emotions begin, or grow, or get carried away. The first movement is non-volitional, a kind of preparation for emotion, a warning, as it were. The second is volitional but not contumacious, like this, “It is appropriate for me to take revenge, since I have been injured,” or “It is appropriate for this person to be punished, since he has committed a crime.” The third movement is already beyond control. It wants to take revenge not if it is appropriate, but no matter what; it has overthrown reason.

If, as is usually assumed, this says that there is a second movement between the pre-emotion and anger itself, that would indeed be a major alteration in Stoic psychology and would present a glaring inconsistency with the Stoic position Seneca defends in Book 1.46 However, a better interpretation is 45 Dial. 4 (= de ira 2).4.2, 4.3.3, epist. 11.1, 71.29. Epist. 71.27 explicitly associates the involuntary feelings with the irrational part of the psyche, mentioning physical pain as belonging to the same category of response (Inwood 2005a: 41). For clarification concerning the Stoics’ version of body-soul dualism, see Long 1996a: 224–249; pace Fillion-Lahille 1984, Rist 1989, and Setaioli 2000: 141, one need not think specifically of Posidonius. 46 A reading along these lines is made practically inevitable by the usual assumption of a sharp break of sense after paragraph 2.4. See esp. Sorabji 2000, who contends for Senecan innovation (esp. 61–63); also Donini 1995 and Vogt 2006: 69f.; among older treatments Holler 1934 and Fillion-Lahille 1984: 163f. I respond to Sorabji at greater length in Graver 2002b. Inwood (2005a: 61–63), who also stops at the end of 2.4, attempts to preserve Seneca’s orthodoxy by making the second and third movement different aspects of the same event. In fact, there is a

272

margaret r. graver

available. The second movement may be anger itself, which though powerful is not as refractory as what anger often leads to, namely actual loss of rationality and the behavior of the insane. For Seneca goes on to say that there is a further state of mind in which people “rage around at random and delight in human blood,” not because they believe they have received an injury, but for pleasure. Examples include Phalaris, the tyrant who tortured men for amusement, and Hannibal, who looked at a ditch filled with blood and exclaimed “O beauteous sight!” This state of mind is not anger: Seneca’s term for it is feritas, “brutishness.”47 But it has its origins in anger: when anger is exercised and satiated too often, it “casts out every human contract from the mind” and passes into a new state (dial. 4 [= de ira 2].5.2f.). In contrast to this bloodthirsty condition ordinary anger, for all its dangers, appears “not contumacious.” A single episode of anger may therefore be a step along the path to the disintegration of our rational condition. But giving free rein to one’s emotions may have other consequences, which, though less extreme, are also undesirable for oneself. Seneca resembles Cicero in naming certain emotive traits of character as evils for the self. These he calls morbi, “diseases,” and, again following earlier Stoic theory, traces their causation to repeated episodes of the relevant emotions without effort at amendment.48 Such conditions differ from the insanity of De ira 2.5 in that they are rational states, consisting in specific beliefs, deeply ingrained errors about the value of external objects. As such, they are still within the reach of moral suasion, and indeed ridding us of all such vicious dispositions is the chief aim of ethical therapy. Affective Responses in the Wise Despite his lively sense of the dangers of emotional experience, Seneca does not by any means believe that the ideal condition for human beings is one devoid of all affective response. In several of his works he speaks emphatically of the feelings of joy experienced by the Stoic sage. This joy is phenomenologically similar to the ordinary person’s delight in the birth of a continuous line of thought from 2.1 through the end of 2.5. The interpretation given here is defended in more detail in Graver 2007: 120–132. 47 Seneca’s treatment of the subject has some points in common with discussions of θηριότης in Aristotle (eth. Nic. 7.1, 7.5) and Theophrastus (apud Simpl., In Aristotelis categorias 8.235). 48 Epist. 75.11 f.; cf. note 25 supra.

ethics ii: action and emotion

273

child or in winning an election: Seneca speaks of “exhilaration” (hilaritas) and an “uplift of the mind” (elatio animi). True joy, however, is constant and reliable, because its object is not some chance event but goods that come from within, that are under one’s own control and have inherent stability. When one has such a foundation, he cannot but be visited, like it or not, by constant exhilaration, by gladness that is deep and comes from the depths. For he is rejoicing in what is his own; he desires nothing more than what he has at home.49

This is the Stoic εὐπάθεια, although Seneca does not use that or any other class term and does not list the three eupathic responses as found in Cicero and other standard sources.50 His emphasis on joy in particular bears comparison with the development of this Stoic theme in Philo of Alexandria, the Jewish exegete.51 While Seneca is familiar with the Stoic claim that the wise person will experience erotic love, this idea is not particularly important to him.52 He is, however, deeply interested in the Stoic notion of wise friendship, which he explores especially in the ninth Moral Epistle. The wise cherish their friends because they value the opportunity to exercise the virtues for their benefit, for instance by sitting at the bedside of one who is ill. Their affection runs deep: losing a friend is like having a hand cut off or both eyes blinded. Nonetheless, the wise person is self-sufficient (se contentus) and impassive (impatiens) in the sense of the Stoic ἀπάθεια. He will not be distressed in times of bereavement and will continue happy, taking satisfaction still in his diminished existence and confident in his ability to make new friends. He indeed feels the loss, Seneca says, but conquers it.53 49 Dial. 7 (= vit. beat.).4.2–5; compare dial. 2 (= const.).9.4, 4 (= de ira 2).6.2, epist. 23.3f., 27.2 f., 59.1 f. 50 Seneca may in fact have been unaware of the Stoic classification of eupathic responses into three genera. In Epistle 85.26 he uses the word “caution” (cautio) for an attitude of the wise person as contrasted with the ordinary person’s fear, but seems to mean only that the wise person can act to avoid danger. This is non-emotive action, called by Stoics “selection” (ἐκλογή), rather than εὐπάθεια. 51 Philo, Quaestiones in Genesim 4.15 f., 19, 101; see Graver 1999: 312–318. 52 Amor in Epistles 81.12 and 116.5 seems to refer specifically to erotic love. Erotic love in the non-wise condition is regularly classed by Seneca as an activity of vice, in parallel with anger and fear; an interesting example is dial. 4 (= de ira 2).15.3. His affection for Paulina (indulgendum est honestis adfectibus: epist. 104.2–3) falls into the category of tender concern for family members generally and should perhaps be treated as φιλία rather than ἐρώς. 53 Epist. 9.3–5; see further Graver 2007: 183f. On grief in others of the Moral Epistles see also Wilson 1997.

274

margaret r. graver

It is in fact quite important to Seneca’s conception of the moral ideal to admit that even a perfected mind would still feel something in circumstances of the kind most of us would find sad or frightening or delightful. This is a theme he sounds over and over again throughout his works. Although the wise person does not believe that such things as death or the loss of a family member are evils, and so does not experience distress or fear itself, he may still tremble or grow pale or change expression when faced with such things.54 When bereaved, he may shed tears either voluntarily, out of joy in remembering the friend’s conversation, or involuntarily.55 If he should meet with external goods, he may experience a thrill, though in comparison with his joy in virtue this will be only a “tiny, meaningless, short-lived movement of the body.”56 He can also blush.57 In conjunction with this theme Seneca regularly adds that such feelings are “natural” or are “commanded by nature.” It is his intention, he explains, to show that the wise person “does not stray outside the natural order” but is a human being like other human beings.58 Wisdom has not altered the fundamental psychological characteristics of human nature. One retains the capacity for emotions even where that capacity is no longer exercised. Conclusion Scholarly accounts of Seneca’s position relative to the ethical psychology of the older Stoa have sometimes presented him either as an innovator who developed his own significantly changed version of Stoicism or as an eclectic who combined elements of Stoicism with ideas from other philosophical traditions, perhaps under Platonizing influence from Posidonius.59 My review of the evidence yields quite a different understanding of Seneca’s overall 54 Dial. 4 (= de ira 2).2.2, epist. 57.3, 71.29, 74.30f.; further examples below and in note 45 supra. Compare de ira 1.16.7, on “suspicions and shadows of the emotions,” mentioning the Stoic founder Zeno. 55 Epist. 99.18f. The voluntary tears would seem to be tears of joy, a eupathic response. This is the only text known to me in which a eupathic response gives rise to weeping. 56 Dial. 7 (= vit. beat.).4.2.4. 57 Epist. 11.2. 58 Epist. 71.27–29; cf. 116.3: Quis negat omnis adfectus a quodam quasi naturali fluere principio? Curam nobis nostri natura mandavit, sed huic ubi nimium indulseris, vitium est. Seneca associates an unnatural flatness of affect with the Cynic school, for instance in dial. 10 (= brev.).14.2: hominis naturam cum Stoicis vincere, cum Cynicis excedere. It may be the Cynic impassivity, rather than the Stoic, that he has in mind indial. 6 (= cons. Marc.).4.1, 12 (= cons. Helv.) and 16.1, 11 (= cons. Pol.).18.5 f. (cf. Pohlenz 1948–1949: 308). 59 For examples see supra, notes 29, 32, 33, 45, 46.

ethics ii: action and emotion

275

project. Neither in psychology nor in ethics did he make it his business to alter any essential commitments of the school to which he adhered.60 Although he claimed the right to abandon the official line, in fact he was well satisfied with the existing Stoic theory of action and the emotions, and devoted his efforts rather to inventing novel rhetorical strategies for putting the most edifying elements of that theory across to his readers. Central for him is the notion that human beings are rational creatures and that rationality is perfectible: our actions are driven by belief, not by forces beyond our control, and beliefs can be corrected. Yet it is no contradiction to say that we are also emotional creatures, equipped by nature to respond affectively to what we perceive as good or evil. The goal of moral progress is not to try to eliminate all affective response, but to understand at a deep level what things really are good or bad for us. If we fail to do this, we set ourselves up for responses that can easily run away with us, with many dangerous consequences. But if we should succeed, we would not therefore become completely unresponsive to the kinds of objects that stir our emotions in our present flawed condition. We would still feel the same kinds of sensations we have now, but in slight and short-lived versions, while in response to genuine goods, the goods of virtuous action and of loving relationships, we would have new, clear, and strong feelings, and above all feelings of joy.61

60 In metaphysics his attitude was different, but metaphysical speculation was rampant in his day; see Sedley 2005a. His views on death and the afterlife seem also to have been heterodox; see Rist 1989 and Smith, infra, pp. 343–361. 61 I would like to thank R. Scott Smith, who read an earlier draft of this chapter and made several very helpful suggestions.

ETHICS III: FREE WILL AND AUTONOMY*

Aldo Setaioli

1. In a famous passage of the Naturales quaestiones, whose purpose is apparently to lay emphasis on the fact that human action (in the case at hand, vows and expiation ceremonies aimed at averting future misfortunes predicted by divination) does not lose its meaning in the deterministic world sketched by Stoic philosophy, Seneca makes a promise that, though related to the point being treated, appears to be wider in scope: “when the time to treat this topic comes, I shall explain how, though fate exists, something remains nevertheless within man’s discretion”1—in hominis arbitrio. This promise, however, is nowhere kept by Seneca, at least in his surviving work. All the reader finds is another reference to the same problem, in an epistle where Seneca, once more, avoids tackling it directly: “this is not the time to embark on a discussion about what falls under our prerogative (quid sit iuris nostri) if providence is in command, or if we are bound and dragged by the chain of fate, or if unpredictable chance is all-powerful.”2 All that both passages offer are mere, if tantalizing, hints; their very linguistic cast, however, is anything but devoid of interest. Although in both cases Seneca resorts to juridical metaphors,3 he is well aware that the *

Submitted for publication in 2007. Sen. nat. 2.38.3: cum de ista re agetur, dicam quemadmodum manente fato aliquid sit in hominis arbitrio. 2 Sen. epist.16.6: sed non est nunc in hanc disputationem transeundum, quid sit iuris nostri, si providentia in imperio est aut si fatorum series inligatos trahit aut si repentina ac subita dominantur. These two passages have often been discussed (Riesco Terrero 1966: 61f. only refers to nat. 2.38.3). Suffice it to refer to Armisen-Marchetti 2000: 210f., Wildberger 2006: I 320f. Mazzoli (1977) sees in these and other passages the announcement of a new work, the Exhortationes. According to him, this work contained “la più compiuta risposta al problema quemadmodum manente fato aliquid sit in hominis arbitrio” (Mazzoli 1977: 31), which he sees embodied in frg. 24 Haase = F 89 Vottero. Cf. also Vottero 1998: 61f. 3 This is true also as far as the terminology employed by the Greek Stoics is concerned. Their definition of ἐλευθερία as ἐξουσία αὐτοπραγίας (SVF III 355) resorts to juridical language (cf. Wildberger 2006: I 340), as does the further definition of ἐξουσία as νοµίµη ἐπιτροπή (SVF III 544). Both definitions correspond to Seneca’s iuris nostri, rather than to arbitrium. 1

278

aldo setaioli

philosophical problem of freedom and autonomy has nothing to do with freedom in the legal sense of the word.4 The terms he employs in the two passages, however, appear to imply two quite different standpoints. The word arbitrium conveys the idea of the power of free choice or decision,5 whereas ius signifies what one is entitled to—normally by law, rules, or regulations.6 This is borne out by Seneca’s usage in several passages,7 including some in which both terms appear together with the clear semantic distinction we have just sketched,8 and a famous text where liberum arbitrium is allotted to clementia, which is considered to be free from the constraints of law.9 Before we take a closer look at these two different standpoints, we should also emphasize the fact that, in the second passage, the whole context stresses the importance of philosophy as a means of moral progress and self-transformation,10 to which one can turn regardless of whether one believes in fate, providence, or mere chance. The first two are compatible with the orthodox Stoic conception, which makes provision for both εἱµαρµένη and πρόνοια; mere chance, however, can hardly be reconciled with it.11 What Seneca is doing here is referring to the words of the inter4

Seneca makes this quite clear: nat. 3 pr. 16: non e iure Quiritium liberum sed e iure naturae (cf. Wildberger 2006: II 930 n. 1513); epist. 80.5: in tabellas vanum coicitur nomen libertatis, quam nec qui emerunt habent nec qui vendiderunt. 5 Bobzien (1998a: 232) translates in hominis arbitrio as “in a person’s power.” Gourinat’s translation (2005: 236) is more accurate: “something is left to the decision of men.” Cf. also Wildberger 2006: I 341: “in der Entscheidungsmacht des Menschen.” 6 Cf., e.g., TLL s.v. arbitrium (II, 412, 67 f.), OLD s.v. ius (10). 7 E.g., benef. 2.18.7, dial. 10 (= brev.).15.3, epist. 70.19, 104.8; in all these passages arbitrium is expressly connected with eligere. At epist. 120.11 the life of the wise man is described as arbitrii sui tota. 8 E.g., dial. 11 (= cons. Pol.).9.3: sui iuris et arbitrii (Polybius’s brother follows his own rules and can do whatever he wants); and the opposition iudex/arbiter at benef. 3.7.5: illum [= iudicem] formula includit et certos, quos non excedat, terminos ponit, huius [= arbitri] libera et nullis adstricta vinculis religio. Even at nat. 2.38.3 there is an opposition ius/arbitrium: something is still within man’s discretion, even though the opponents of Stoicism maintain that the prerogative of acting has been totally entrusted to fate: omne ius agendi traditum. Sen. Med. 137f.: alieni arbitrii / iurisque factus is rightly translated by Chaumartin 1996a: 161 “soumis à la volonté, au pouvoir d’autrui.” For Seneca’s use of sui iuris and sui arbitrii, cf. Kurth 1994: 112. 9 Sen. clem. 2.7.3: clementia liberum arbitrium habet, non sub formula, sed ex aequo et bono iudicat. See on this passage Bellincioni 1984a (= Bellincioni 1986: 113–125), Malaspina 2001a: 409f., also for the numerous occurrences of the formula liberum arbitrium in Livy and elsewhere. Cf. also Inwood 2005a: 207, who rightly renders Seneca’sliberum arbitrium as “freedom of decision.” 10 Cf. Setaioli, supra, pp. 239–256. 11 See below for what Seneca has to say on fortuna.

ethics iii: free will and autonomy

279

locutor, who had previously equated Stoicism’s fate and providence with Epicureanism’s chance, as both depriving human efforts at self-improvement of all meaning.12 Two important consequences follow: On the one hand, what is at issue is not primarily the problem of the freedom of choice, but rather the preservation of the meaning of human effort—which confirms the difference from the passage of the Naturales quaestiones, though there, too, as we have remarked at the outset, the whole context aims to save the significance of human action. Secondly, what Seneca means to stress is the importance of philosophy as a tool for self-transformation and improvement, regardless of the initial convictions of the proficiens. Such was the main purpose of his own philosophical writing, and Lucilius himself may have had Epicurean leanings.13 Seneca, then, had to take the freedom to embark on such a program of self-transformation more or less for granted, regardless of whatever theoretical stipulations purely doctrinal premises might entail. On the other hand, as we shall see, these were compatible with a level of autonomy that makes man morally responsible, as Seneca himself makes clear: though his nature impels man to perfect his innate reason, he acquires merits and deserves praise, if he strives to attain this end.14 We may conclude that, even as he declines to tackle the problem, Seneca provides us with important clues, which must be kept in mind as we try to sketch a fuller picture of his attitude toward these difficult questions from hints and statements scattered throughout his work.15 It has been rightly remarked that “the reconciliation of fate and moral responsibility was the dominant and characteristic problem of Stoic moral philosophy.”16 This, of course, only implies that all that the early Stoics endeavored to establish was that man possesses a certain autonomy, in that he is himself the efficient cause of what fate has established for him. His responsibility stems from this, not from a power of deciding or doing 12 Sen. epist. 16.4: dicet aliquis, “quid mihi prodest philosophia, si fatum est? quid prodest, si deus rector est? quid prodest si casus imperat? […]” 5: sive nos inexorabili lege fata constringunt, sive arbiter deus universi cuncta disposuit, sive casus res humanas sine ordine impellit et iactat, philosophia nos tueri debet […]. For §6, cf. supra, n. 2. Although chance (τύχη) was considered by the Stoics to be the aspect of fate that cannot be grasped by the human mind (cf. below, on Seneca’s fortuna), it seems hardly likely that Seneca had the Stoic conception in mind in this reference to casus, as maintained by Wildberger 2006: I 337f., following Hachmann 1995: 140f. 13 Cf. Setaioli, supra, p. 245. 14 Cf. Sen. benef. 6.21.3, epist. 76.10, quoted and discussed below, notes 73f. 15 As we shall see, Wildberger’s attempt (2006: I 342–348) to elicit a fuller standpoint from Seneca’s treatment of the Etrusca disciplina in the second book of the Naturales quaestiones (in particular from nat. 2.38.4 and 2.46) is hardly successful. See below, notes 71, 117. 16 Inwood 1985: 66.

280

aldo setaioli

otherwise.17 Accordingly, the problem of free will cannot even be posed for the Stoics, if this formula only refers to the problems arising from the idea of the freedom to decide, will, or do otherwise as related to determinism.18 As Susanne Bobzien remarks,19 determinism is firmly grounded in Stoic cosmology, although it was bound to have a great impact on Stoic ethics; but, as Brad Inwood observed,20 while dealing with human action the Stoic philosophers had to assume—or rather take for granted—that their determinism could be reconciled with ethics and moral evaluation without disrupting the coherence of their system. And, as Anthony Long puts it, “it would be surprising if philosophers whose main concern was ethical, robbed the human mind of libera voluntas.”21 While investigating Seneca’s views on free will and autonomy, the way they fit in the whole picture of Stoic thinking must never be lost sight of.22 2. We may start with Seneca’s ideas on freedom, libertas. We know that Cleanthes wrote a περὶ ἐλευθερίας23 and that, for the Greek Stoics, ἐλευθερία was one of the τελικὰ ἀγαθά.24 However, as Bobzien maintains,25 ἐλευθερία is never connected in the early Stoics with free choice or the power to do otherwise. She also contends that the connection between ἐλευθερία and what depends on us26 was not made before Epictetus,27 who allegedly 17

Cf. Bobzien 1998b: 135–137. As Bobzien (1998b: 136) does. She distinguishes three types of indeterminist freedom: freedom to do otherwise, freedom of decision, freedom of will (pp. 133f., 136). Only the first two types appear in Bobzien 1998a: 277. We will not connect the expressions “free will” and “freedom of will” with the problem of the existence of will as an autonomous part or faculty of the soul. 19 Bobzien 1998a: 16. Cf. also Long and Sedley 1987: I 342 f., 392–394, etc. 20 Inwood 1985: 99. 21 Long 1971: 173. For the function of will in this connection in Seneca, see below, § 5. 22 According to Bobzien (1998a: 12f.), in Seneca “there is too much, too unorthodox, too difficult, too unstructured material” to include him in a study of Stoic determinism. The Stoic framework, however, must always be kept in mind in order to assess Seneca’s pronouncements. 23 SVF I 481. 24 SVF III 107. 25 Bobzien 1998a: 338 f. 26 (Τὸ) ἐφ᾽ἡµῖν in the standard formulation, which, however, does not seem to be attested for Chrysippus (Bobzien 1998a: 332 n. 3), unless SVF II 984 and 1007 are considered direct quotations from him, as they are printed by von Arnim. In SVF II 998 the expression τὸ πὰρ’ ἡµᾶς is used. According to Epictetus what was in our power was only the “use of presentations” (φαντασίαι). 27 Bobzien 1998a: 341–345; cf. Inwood 2005a: 302 f. 18

ethics iii: free will and autonomy

281

conceived of freedom as the knowledge of what is in our power and the will to pursue only this.28 But, as Jula Wildberger has pointed out,29 Epictetus had not really been the first to do so. Seneca himself was fully familiar with the doctrine of “reservation” (ὑπεξαίρεσις, exceptio):30 the wise man will never be disappointed, because he always makes provision for an unforeseen intervention of fate; but what really makes the wise man successful—and free—is the absolute value accorded to moral action per se, regardless of the outcome.31 Seneca never tires of stressing this idea,32 and in this he, of course, agrees with all the Stoics.33 This is what they meant when they stated that only the wise man is free.34 Though the term libertas carries several quite different overtones in Seneca,35 he also uses it to express the wise man’s freedom, in keeping with this Stoic idea. In an interesting passage of the De brevitate vitae he opposes the “half-freedom” Cicero speaks of in one of his letters to the “total and solid freedom” of the wise man.36 This freedom is first described 28

Bobzien 1998a: 342. Wildberger 2006: II 927 f. n. 1509. She quotes Sen. epist. 51.8f. Even clearer is SVF III 356 (= Dion Chrys. 14.16), where freedom is defined as the science of what is permitted and what is forbidden. Dio was an older contemporary of Epictetus’s. 30 Discussed by Inwood 1985: 119–126, 165–175, Bobzien 1998a: 333, Reydams-Schils 2005: 28f., 63f., Sharples 2005: 205f., Wildberger 2006: I 350. For Seneca, cf. benef. 4.34: sapiens […] ad omnia cum exceptione venit: “si nihil interciderit, quod inpediat.” Ideo omnia illi succedere dicimus eqs.; dial. 9 (= tranq.).13.3: hoc est quare sapienti nihil contra opinionem dicamus accidere[…]; in primis autem cogitavit aliquid posse propositis suis resistere. The often repeated (e.g., dial. 9 [= tranq.].13.4, dial. 6 [= cons. Marc.].9.2, epist. 76.34, 91.3) and seemingly hackneyed idea that a foreseen misfortune hurts less must be seen in this light. For early Stoicism, cf. SVF III 564. 31 Cf., e.g., Sharples 2005: 204 f. 32 E.g., Sen. benef. 4.1.3, 4.12.4, 4.21.6, 4.22.2, epist. 14.16, 76.28f., 93.9f. (for the latter passage, see Setaioli 2000: 318 f.). 33 For the common Stoic idea of the honestum per se expetendum and praiseworthy even if not praised by anyone, cf. Setaioli 2008. For Epictetus, e.g., diatr. 4.4.44, 4.8.1–4 (only the correctness of the δόγµατα makes an action good). 34 Cf., e.g., SVF I 218, 222, III 355, 362, 363, 364, 544, 593. 35 Inwood (2005a: 302f.) distinguishes four types of Senecan freedom: 1) freedom guaranteed by suicide; 2) immunity from offences and injuries; 3) freedom stemming from the acceptance of fate; and 4) freedom from passions and emotions. All these types of freedom certainly exist, and others could be added: e.g., freedom stemming from αὐτάρκεια, which makes the wise man unassailable by fortune and anything external; also freedom stemming from self-possession (see below, note 184). More to the point is perhaps a distinction between a negative conception of freedom (which Inwood’s second and fourth type would fall under; cf. Traina 1987: 77) and a positive one. We cannot include the important theme of freedom guaranteed by suicide in our treatment, but it can be pointed out that this is one of the cases in which man’s freedom of choice is taken for granted. 36 Sen. dial. 10 (= brev.).5.3: semiliberum se dixit Cicero: at mehercules numquam sapiens in tam humile nomen procedet, numquam semiliber erit, integrae semper libertatis et solidae, 29

282

aldo setaioli

negatively: the wise man is solutus, i.e., “unbound,” free of constrictions;37 immediately after he is called “autonomous” (sui iuris), an expression that reminds us of the passage from Epistle 16 we have quoted and discussed at the beginning. So, the wise man’s freedom is not an arbitrium, a power of free choice or decision, but rather an autonomy granted within a normative framework.38 To be free man must surely put himself above the allurements of passion and the attacks of fortune, but most of all he has to accept wholeheartedly what fate (which in Stoicism is tantamount to God or providence) has decided for him. Only when he wants what God wants39 shall he be free. Seneca says this much very clearly: “freedom is obedience to God.”40 We know that this conception of freedom, which was endorsed by the most orthodox Stoics, was later labeled a ἡµιδουλεία, a “half-slavery.”41 If we could assume that this quip was already current in Seneca’s time, it would indeed be tempting to suppose that when he says that the wise man will never be “half-free”42 he meant to reject this sneer at the Stoic idea of freedom.43 The ontological side of this idea (accepting and actively endorsing the divine plan) perfectly corresponds to the ethical side (ridding oneself of whatever impedes man’s rational nature intended by god):44 both are linked through the need, proclaimed by Seneca, to put ourselves at the service of philosophy, in order to attain real freedom.45 We have mentioned the doctrine of ὑπεξαίρεσις/exceptio; but this is only valid in relation to man’s ignorance of God’s plan. Exceptio is indeed one way to prepare and fortify oneself against the unforeseen; another are the practical or partly practical “exercises” recommended by Seneca.46 But man must always be prepared to accept as his own will whatever fate

solutus et sui iuris et altior ceteris. On this subject, see Wildberger 2006: I 348 f.; and for the connection with Cicero, Setaioli 2003: 58–60. 37 Cf. supra, note 35. 38 See supra, on ius and arbitrium. 39 E.g., Sen. epist. 66.39, 71.16, 74.20, 96.2; cf. Wildberger 2006: I 272–274. 40 Sen. dial. 7 (= vit. beat.).15.7: in regno nati sumus: deo parere libertas est. 41 By the Cynic Oenomaus, polemicizing with Chrysippus (SVF II 978). 42 Cf. supra, note 36. 43 On the other hand, Seneca may also have reacted to this sneer by a polemical exaggeration when he equates real freedom with a very special kind of slavery: dial. 7 (= vit. beat.).15.7 (supra, note 40), epist. 8.7 (below, note 45). 44 Cf. Baldarotta 1994: 26. 45 Sen. epist. 8.7: philosophiae servias oportet, ut tibi contingat vera libertas. Philosophy was, in fact, the science of things godly and human: epist. 89.5. 46 Cf. Setaioli, supra, pp. 239–256.

ethics iii: free will and autonomy

283

has in store for him. In this, again, Seneca agrees with all the Stoics.47 Chrysippus had expressed this idea very clearly: “if I knew that fate has decided for me to be ill, I would set my course towards this by my own accord.”48 Seneca expresses this same idea in the words he puts in the mouth of his friend, the Cynic philosopher Demetrius, whose only complaint was that the gods had not made known to him their will, which he would have fulfilled by his own accord, no matter how hard or disagreeable.49 According to the Stoics this is the best of all possible worlds, being ruled by deified Reason;50 therefore what has been established from the beginning cannot be changed.51 Being in agreement with God’s plan, therefore, is to be at one with Reason and Nature,52 and consequently with ourselves, since our own nature is rational.53 If we do not conform to God and nature, not only shall we be frustrated, but we shall renounce our rational nature. This is expressed in the clearest terms in four famous verses by Cleanthes, which Seneca translated into Latin54 with the addition of a fifth line.55 47

Bobzien (1998a: 347 f., 349 f.) is quite wrong when she says that the verses of Cleanthes (SVF I 527) we shall discuss below assume that one must know what fate has decided in order to conform to it. Stoics were not above resorting to divination or astrology (naturally integrated in their system: see, e.g., Sen. nat. 2.32, 2.45, and cf., e.g., Magris 1990: 65), but the wise man does not need to know in advance the details of God’s plan to accept it wholeheartedly. 48 SVF III 191. Cf. Epict. diatr. 2.10.5 f., 3.5.9 f. 49 Sen. dial. 1 (= prov.).5.5. Cf also nat. 3 pr. 12: quid est praecipuum? posse laeto animo adversa tolerare, quicquid acciderit sic ferre quasi volueris tibi accidere; debuisses enim velle si scisses omnia ex decreto dei fieri. 50 Partial losses or damages must be seen in the context of the well-being of the whole: see below, note 176. 51 Sen. nat. 1 pr. 3: necesse est eadem placere ei cui nisi optima placere non possunt; cf. 2.36. 52 It is of course unhistoric to consider this a rationalization: cf. Inwood 2005a: 249–270, and my review, Setaioli 2007a: 694. 53 This is what the Stoics meant with their formula ὁµολογουµένως (τῇ φύσει) ζῆν. 54 SVF I 527, Sen. epist. 107.11. Cf. Setaioli 1988: 70–82, Bobzien 1998a: 346–356, Sharples 2005, Wildberger 2006: I 294–299. 55 In Seneca’s translation we find details that are missing in Cleanthes’s verses but are very closely paralleled in Simplicius’s commentary on the chapter of Epictetus’s Encheiridion quoting them: 1) the fool’s groaning as he is forced to follow fate; 2) the perfect syntactical correspondence in the expression of this idea: future indicative accompanied by present participle(s) (comitabor gemens ~ οἰµώζων καὶ στένων ἀκολουθήσω); 3) the fifth verse (ducunt volentem fata, nolentem trahunt ~ ἀνάγκη […] πάντα ἄγουσα, ἑκόντα τε καὶ ἄκοντα, with the typically Senecan polarizing and antithetical reduplication of the verb; though Long (1971: 195 n. 15) sees a Chrysippean distinction between “fate” and “necessity,” Seneca clearly equated the latter with fate: nat. 2.36., dial. 1 (= prov.).5.8, cf. Rist 1969: 126). Unless we are ready to attribute all of this to mere chance, we will have to admit that Seneca read Cleanthes’s verses together with a commentary that was later used by Simplicius (Bobzien [1998a: 356] cavalierly dismisses

284

aldo setaioli

In Seneca’s translation, as everywhere in his writing, style plays an important role not only in the expression, but in the very shaping of his thinking, as has been repeatedly and incontrovertibly shown by Alfonso Traina. Susanne Bobzien thinks that “whether or not Cleanthes thought that it was fated whether someone wants or does not want to follow […], we do not know.”56 I am not completely certain about this; what is for sure, however, is that Seneca’s translation shows that for him the choice between following or not, between being good or bad, is in our power: malusque patiar facere quod licuit bono. His antithetical reduplication of πείσοµαι, which he surely read in Cleanthes’s fourth verse,57 through the future patiar and the infinitive facere governed by Seneca’s own and very meaningful impersonal licuit, introduces in Cleanthes’s text the idea that taking the better course falls within man’s prerogative, at least if he is wise and endowed with his own ius, which enables him not to become “bad”—a neat example of the inseparability of form and content in Seneca’s writings, where the former, far from being a “bellelettristic adornment”58 of the latter, is itself flesh and blood of his thought.

the survival of such commentaries already in the late second or early third century ad, but she is unaware of the correspondences between Seneca and Simplicius). The mixing of poetical texts with details taken from commentaries was current in Latin translations, as repeatedly pointed out by my teacher, Alessandro Ronconi. According to Bobzien (1998a: 346 n. 57, 348 n. 53), the fifth verse only shows how Seneca understood Cleanthes; Sharples (2005: 214) offers mere hypotheses about the fifth verse based on his failure to realize that the idea of “dragging” does not need to be explained through the influence of a different text, being probably the result of the antithetical reduplication of a verb (cf. ἄγουσα in Simplicius), so typical of Senecan style. Seneca surely read Cleanthes’s fourth verse not as we find it in Epictetus (κακὸς γενόµενος οὐδὲν ἧττον ἕψοµαι), but in the form transmitted by Vettius Valens (κακὸς γενόµενος αὐτὸ τοῦτο πείσοµαι), as shown by the second of his two antithetical reduplications in this line (κακός ~ malus/bono; πείσοµαι ~ patiar/facere […] licuit). As Wildberger (2006: II 892) points out, Bobzien (1998a: 349) clearly misunderstands this verse; she takes κακὸς γενόµενος to be the reason for not wishing to follow fate, whereas it is obviously the result or consequence (the meaning is: “if I do not wish to follow, I will become bad and [be forced to] follow nevertheless”). The same mistake is found in Long and Sedley 1987: I 386 (“but if I become bad and am unwilling, I shall follow none the less”). 56 Bobzien 1998a: 349 (cf. already Long and Sedley 1987: I 392). Bobzien’s distinction (pp. 347f.) between factual and normative universal law is groundless: for both Cleanthes and Seneca man has the duty to accept God’s plan willingly; but fate will have its course, whether we accept it or not; so this law is both normative and factual. Cf. also Sen. nat. 3 pr. 12 (supra, note 49). 57 Cf. supra, note 55. How this πείσοµαι, obviously the future of πάσχω, can be taken to be the future of πείθοµαι (as done by Andreoni Fontecedro 1986–1987: 370 n. 16, Andreoni Fontecedro 1993, Andreoni Fontecedro 2000: 186) is beyond my comprehension. See Setaioli 2000: 233–242, Setaioli 2002. 58 So Inwood 2005a: 164.

ethics iii: free will and autonomy

285

We should always keep this in mind, when we read the countless passages59 in which Seneca preaches the acceptance of fate as the achievement of the only real freedom. The verb here used by Seneca (licuit) does not permit us to equate this with a totally emancipated arbitrium; nevertheless, in his translation of Cleanthes’s lines what is at stake is a choice; what we should probably understand is that—whether we choose to avail ourselves of it or not—we have received the power to make the right choice through God’s benevolent plan.60 3. Seneca, it appears, has made a really bold claim: we have been shaped (one might say “determined”) by the cosmos’s supreme power, whose kin and part we indeed are;61 however, it seems that what has been given us is not merely the autonomy to act in a morally responsible way, but the actual capability of choosing and deciding. We should check how this claim is borne out in Seneca’s writings and compare his views with those of the Greek Stoics on freedom and autonomy. What Chrysippus was apparently preoccupied with was not so much asserting “free will,” in the sense of freedom of decision or of doing otherwise, but rather reconciling determinism with moral responsibility—i.e., an attempt to substantiate a “compatibilist” view leaving room for both the latter and the former. Fate acts both from outside and from inside us; our own individual nature and character is the immediate (“perfect”) cause for the way we react to an external stimulus, which is only a remote and “auxiliary” cause.62 Space does not allow us to go into closer detail, but we should at least refer the reader to two important testimonies on Chrysippus’s view, as reported by two Latin authors: Cicero and Gellius.63 It appears that not only the external stimulus but also our individual natures are foreordained by fate.64 There is no doubt that here the concept of moral responsibility is based 59 Space prevents us from dwelling on them here. However, I would like to point out at least an important passage that Andreoni Fontecedro (1992: 163) calls “un manifesto dell’ideologia del fato stoico”: Sen. dial. 1 (= prov.).5.7 f. 60 Cf. Sen. epist. 31.9: tutum iter est, iucundum est, ad quod natura te instruxit. Dedit tibi illa quae si non deserueris, par deo surges. 61 Cf. Setaioli 2006–2007: 350 f. 62 Cf., e.g., Bobzien 1998a: 234–329, Sharples 2005: 201 f. 63 Cic. fat. 39–44 (= SVF II 974) and Gell. 7.2 (= SVF II 1000). 64 Gell. 7.2.9: idque ipsum (the way people with different natures react) ut ea ratione fiat, naturalis illa et necessaria rerum consequentia efficit, quae fatum vocatur. Cf. Bobzien 1998a: 253.

286

aldo setaioli

on the autonomy of the agent—in that he is himself the cause of his action, which is in accord with his nature and character—not on his indeterminist freedom of choice and decision.65 Chrysippus, however, maintained that granting or withholding assent to the external stimulus is in our power.66 This is not the place to stress the weakness of this position.67 We should rather try not to miss the full implication of our testimonies. When Gourinat68 translates Gellius’s words ingenia tamen ipsa mentium nostrarum proinde sunt fato obnoxia ut proprietas eorum est et ipsa qualitas69 as “the nature of our minds is subject to fate in the same way as their own properties and their quality,” he clearly misunderstands Gellius’s meaning. What Gellius is actually saying is that our different natures are influenced by fate according to their individual properties and quality. It should not escape our attention that shortly after Gellius says that it is not fatale, but quasi fatale that men of evil nature should not be free from fault.70 What is meant is probably that the way we act only indirectly depends from (external) fate, the main factor being our own (internal, though itself fated) nature. There is no trace in Seneca of the image of the cone and the cylinder with which Chrysippus elucidated his doctrine of the external stimulus (illustrated through the impulse that objects having this form receive from an exterior agent) and the personal reaction (in Chrysippus’s image the way such objects move according to their shape). The doctrine itself, however, was not unknown to him. It is hardly recognizable in the passage of the Naturales quaestiones, where Wildberger strives to track it down;71 but it is clearly implied in other parts of his work. In a very important passage of the De beneficiis Seneca states that the good man cannot behave any different

65 As Bobzien (1998a: 255) puts it, “according to Chrysippus, someone can rightly say ‘I was fated to do it,’ but could not say ‘It was fate which did that, hence it was not me who did that.’” 66 Cic. fat. 41, 43, Gell. 7.2.11; cf. SVF II 981, 988, etc. 67 Cf., e.g., Bobzien 1998a: 298f., Inwood 1985: 70; also Inwood 1985: 99, quoted supra, note 20. 68 Gourinat 2005b: 233. 69 Gell. 7.2.7. Long and Sedley (1987: I 388) translate it correctly. 70 Gell. 7.2.10: est enim genere ipso quasi fatale et consequens, ut mala ingenia peccatis et erroribus non vacent. 71 Wildberger (2006: I 346–348) tries to find this doctrine at nat. 2.46: [Iuppiter] singulis non admovet manum: vim et causam omnibus dedit. This reference seems striking at first sight; but the words singulis and omnibus are not connected with moral agents to which Jupiter has given an external cause or impulse (causam) and an individual nature (vim) by which they move in a certain way. Instead, they refer to events (divinatory signs) preordained by Jupiter, to which he has given a meaning (vim) and a purpose (causam).

ethics iii: free will and autonomy

287

from the way his own nature makes him behave; 72 but he goes immediately on to say that this nullifies neither his will nor his merits; 73 we have already seen that although reason is the intrinsic characteristic of man, if he develops his rationality through a correct “conciliation” to his own nature, he deserves praise.74 This praise is indeed no less deserved for the fact that the moral agent acts according to his own nature; Seneca reminds us that though God has sown “divine seeds” in all men, the outcome will be very different: in the “good” they will bear divine fruits, in the “bad” they will die to be replaced by chaff and stubble;75 both the “good” and the “bad” react according to their nature, but never for a moment does Seneca doubt that both fully deserve this moral assessment. In all these texts there seems to be no further individual characterization beyond the good or wise man on one side and the crowd of the stulti or mali on the other. But we should not forget that Seneca harbored a lively interest for individual character and personality, following in the wake of Panaetius.76 Seneca, then, agrees with the Greek Stoics in holding that the “bad” are bound to act badly and the “good” to accomplish good deeds;77 but there is a further point, which he takes up and develops. According to Alexander of Aphrodisias,78 the main reason for Chrysippus to deny that autonomy consists in being able to choose between opposites was that, if we assume this, all moral action would be abolished, because, though vice and virtue depend on us (in the way sketched supra), there is really no such choice: bad men can only act badly and good men can only act well.79 What would be lost in that case is praise and blame (which are naturally inherent to virtue and vice), and also suasion and dissuasion. 72 Sen. benef. 6.21.2: vir bonus non potest non facere quae facit; non enim erit bonus, nisi fecerit. Cf. Inwood 1985: 110. 73 Sen. benef. 6.21.3: si necesse est illi velle ob hoc, quia nihil habet melius, quod velit, ipse se cogit; ita, quod tamquam coacto non deberem, tamquam cogenti debeo. For God, cf. frg. 122 Haase = F 84 Vottero (below, note 80). 74 Sen. epist. 76.10: quid est in homine proprium? ratio: haec recta et consummata felicitatem hominis implevit. ergo si omnis res, cum bonum suum perfecit laudabilis est et ad finem naturae suae pervenit, homini autem bonum ratio est, si hanc perfecit laudabilis est et finem naturae suae tetigit. 75 Sen. epist. 73.16: semina in corporibus humanis divina dispersa sunt, quae si bonus cultor excipit, similia origini prodeunt et paria iis ex quibus orta sunt surgunt: si malus, non aliter quam humus sterilis ac palustris necat ac deinde creat purgamenta pro frugibus. 76 Cf. Setaioli 2000: 111–217, 397–408. For Chrysippus, cf. Cic. fat. 8 (= SVF II 951). 77 Cf. SVF III 110. 78 Alex. Aphr. fat. 21 (= SVF II 984). 79 Cf. Long 1971: 184.

288

aldo setaioli

We have seen how the Stoics—and Seneca—saved moral responsibility: though we act according to our nature, we are ourselves the cause of our action: the good man is not forced to act well by an exterior force, but, to borrow Seneca’s words, ipse se cogit.80 But, as we just said, this fragment mentions not merely praise and blame, but also “suasion” and “dissuasion”—προτροπαί and ἀποτροπαί. In the testimony by Gellius that we have already referred to, Chrysippus is reported as stating that if somebody’s bad nature has not been aided by education it will have no way to avoid faults and misdeeds.81 This implies that the Stoics laid great emphasis on education, in relation to the formation and transformation of one’s character.82 As we have remarked at the beginning, the Stoics needed to take man’s capability of moral improvement more or less for granted.83 They never doubted that virtue was teachable.84 As far as Seneca is concerned, the main goal of his philosophical writing was the moral improvement of both himself and his readers.85 He gives an answer to the two questions that, according to Bobzien, remain unanswered in the scanty remains of the Greek Stoics: “whether everyone was considered equally teachable and capable of change, and whether the teachability remains the same at any stage of a person’s life.”86 According to Seneca, philosophical therapy must be attempted in all cases,87 and teaching is not the same for everyone and at all stages, though suasion and dissuasion are indeed an indispensable tool.88

80 Sen. benef. 6.21.3 (supra, note 73). Cf. frg. 122 Haase = F 84 Vottero: “ergo” inquit “deum non laudabimus, cui naturalis est virtus?” […] Immo laudabimus; quamvis enim naturalis illi sit, sibi illam dedit, quoniam deus ipse natura est. 81 Gell. 7.2.8: si vero sunt [scil. ingenia] aspera et inscita et rudia nullisque bonarum artium adminiculis fulta, etiamsi parvo sive nullo fatalis incommodi conflictu urgeantur, sua tamen scaevitate et voluntario impetu in assidua delicta et in errores se ruunt. 82 Cf. SVF III 225, 366. 83 Bobzien (1998a: 290–301) provides a clear picture of this need as well as of the related philosophical problems. 84 Cf., e.g., SVF I 567, III 223. 85 Cf. Setaioli, supra, pp. 239–256. 86 Bobzien 1998a: 295. 87 Sen. epist. 25.2, 29.3, 50.6 (although there are cases in which it proves unsuccessful: epist. 94.24 and 31, clem. 1.2.2, dial. 8 [= de otio].3.3). Strategies will vary according to the stages of corruption: cf., e.g., epist. 25.1, dial. 4 (= de ira 2).18.2. Cf. Setaioli, supra, pp. 239–256. 88 The emotional admonitio must precede the institutio proper, though the former remains necessary all along. Cf. Setaioli, supra, pp. 239–256.

ethics iii: free will and autonomy

289

4. We have seen that, although we act according to our individual nature, we are morally responsible for our actions. These are the result of accepting the stimulus of an external presentation, i.e., of our assent to it.89 But in a determinist system even our assent is determined by fate,90 and, as a consequence, so are our actions. This seriously jeopardizes the usefulness and the very meaning of human action and/or willingness to act, if the latter is itself inescapably inscribed in a circle determined not by us as agents, but by fate. It is not surprising, therefore, that the so-called “Idle” or “Lazy Argument” (ἀργὸς λόγος) was brought to bear against this type of determinism, possibly by the Megarics.91 It is reported by Origen92 and Cicero,93 in approximately these terms: “if it is fated for you to recover from disease, you will, whether or not you call for a doctor; if you are not fated to recover, you will not, whether or not you call for a doctor; it is therefore useless to call for a doctor.” Chrysippus’s rebuttal is recorded by Cicero,94 from whom we learn that the Greek master had devised the doctrine of the confatalia, according to which our actions are indeed included in fate but are posited as necessary, “co-fated” conditions for the fulfillment of its decrees. In this case we have fata copulata, inasmuch as an event is coupled with a necessary condition: calling for a doctor is as fatally preordained as recovering from the disease. 95 If no co-fated condition is included in a fatal decree, then we have what Cicero calls fata simplicia.96 The Stoic rebuttal of the “Idle Argument” is found in two more Latin texts: a long scholium by the Servius Danielinus to Virgil’s Aeneid and a work by Seneca.97 Both 89

Cf., e.g., Inwood 1985: 81, and see supra, note 66. Cf., e.g., SVF II 993, 998. Incidentally, as Gourinat (2005b: 234 n. 93) points out, this would make moral improvement philosophically problematic. But the Stoics—and certainly not Seneca—do not seem to have been troubled by the contradictoriness of these two stands. One could observe that with Seneca (and Epictetus) the stress is clearly shifted from the physical and ontological domain to ethics, i.e., from the problem of how assent is produced to the moral worth of the correct attitude and intention, regardless of how it comes about. 91 Cf. Bobzien 1998a: 180: “there is some reason to think that it may have originated among the Megarics and logicians such as Diodorus Cronus.” 92 Orig. contr. Cels. 2.20 (= SVF II 957): ὁ ἀργὸς καλούµενος λόγος. 93 Cic. fat. 28 f. (ignava ratio). 94 Cic. fat. 30 (= SVF II 956). 95 Confatalis corresponds to συγ(καθ)ειµαρµένος: cf. SVF II 998 and the texts listed by Bobzien 1998a: 212 n. 93. 96 Cf. also SVF II 998: ἁπλῶς καθείµαρτο. 97 Serv. Dan. ad Aen. 4.696, Sen. nat. 2.37f. For both texts, see Setaioli 2004–2005: 13–18. For Seneca, see also Armisen-Marchetti 2000: 207–211, Gourinat 2005b, Setaioli 2006–2007: 360f., 90

290

aldo setaioli

conflate fate and conditional prophecy,98 but both present clear parallels with Cicero. The Servius Danielinus uses a terminology that is different from Cicero’s, but clearly equivalent to it: his fata denuntiativa and condicionalia obviously correspond to Cicero’s fata simplicia and copulata,99 and in his commentary there are striking similarities with Seneca’s treatment of the Etruscan lore concerning lightning and thunderbolts, in which his mention and discussion of the “Idle Argument” is included.100 If, as I think it likely,101 this exegesis, which is applied by the Latin scholiast to the death of Dido, was originally devised in connection with Homer’s verses about Achilles’s choice of life102—which is referred to in the text of the Servius Danielinus—we must conclude that the Homeric scholiast linked the doctrine of the confatalia with freedom of choice and/or decision, rather than with the significance of human action. Unlike Cicero and the Servius Danielinus, Seneca does not employ strictly technical terms,103 but he describes the co-fated conditions as clearly as they do with the verb adfatum est, which has been wrongly corrected by some editors.104 The same idea is also expressed through another word, which has sparked much debate: suspensa. Seneca says that some events have been left “hanging” by the gods in such a way that they may take a turn for the

Wildberger 2006: I 328–336. Bobzien 1998a: 180–183 provides a clear picture of the debate, but her treatment of Seneca is marred by serious misunderstandings. See below, note 113. 98 Seneca does so more clearly than the scholium, although Bobzien (1998a: 181 n. 6) refers only to the latter in this connection. Cf. Sen. nat. 2.38.2, where several predictions are made to be dependent on determinate conditions. 99 There is no reason to deny that the scholium reflects Stoic doctrine, as Hine (1981: 369–371) does. Cf. Setaioli 2004–2005: 16 f. for a detailed refutation. 100 Serv. Dan. ad Aen. 8.524 is almost identical to Sen. nat. 2.39.1 (classification of thunderbolts). Hine (1981: 379 f.) and Parroni (2002: 521) think that the scholiast drew upon Seneca, which I believe to be very unlikely. Cf. Setaioli 2004–2005: 18 and n. 336. 101 Cf. Setaioli 2004–2005: 16 f. 102 Hom. Il. 9.410–416. Eustathius’s commentary on these verses (II, 746, 12–19 van der Valk), though employing no technical terminology, does indeed explain them as a conditional prophecy, also referring to the gods’ warning to Aegisthus (Hom. Od. 1.35–43) and to Odysseus’s companions eating the Sun’s cattle and thus forfeiting their return to Ithaca (Hom. Od. 1.6–9) as further instances. 103 This can be observed very often in Seneca: cf. Armisen-Marchetti 1996a. 104 Sen. nat. 2.38.2: in illo fati ordine quo patrimonium illi grande promittitur, hoc quoque protinus adfatum est, ut etiam naviget. Hine (1981: 100 and 377) prints adfatum between two cruces, as he also does in his Teubner edition of 1996. Parroni (2002: 136) prefers Russell’s conjecture adiectum. Adfatum is rightly preserved by Marino 1996a and Vottero 1989. The passive usage of the verb receives support from Cic. fat. 30: si ita fatum sit “nascetur Oedipus Laio” eqs., and offers no real linguistic problem: cf. Setaioli 2004–2005: 16 n. 327. It is actually a very close rendering of the Greek συνείµαρται.

ethics iii: free will and autonomy

291

good if expiation rites are performed—a condition itself included in fate.105 Though many scholars believe that suspensa here means “undetermined” or “undecided,”106 it is clear that what Seneca means is “depending (upon)” or “attached (to)”; in other words, he is referring to Cicero’s fata copulata,107 in which the event is inextricably connected with a co-fated condition. This is supported both by what immediately follows—ipsum quoque in fato est— and by Seneca’s linguistic usage.108 Seneca has the interlocutor formulate the “Idle Argument” in a peculiar way (by resorting to generic propositions about the future),109 immediately proceeds to declare it false for the fata copulata,110 and goes on by making the interlocutor retort that the co-fated condition does not change the determinist picture, since it is also included in fate111—an objection that was natural enough for anyone who rejected determinism.112 Seneca must

105

Sen. nat. 2.37.2: quaedam a dis immortalibus ita suspensa relicta sunt ut in bonum vertant si admotae dis preces fuerint, si vota suscepta: ita non est hoc contra fatum, sed ipsum quoque in fato est. 106 So Hine 1981: 369, who, as a consequence, must deny that this position is genuinely Stoic. This is the current interpretation: cf., e.g., Riesco Terrero 1966: 64, Oltramare 1973: I 87, Vottero 1989: 341, Marino 1996a: 115, Parroni 2002: 137. Wildberger 2006: I 332 translates suspensa as “in der Schwebe,” but must admit that it really means “von dem Verhalten der Menschen abhängig.” 107 This is implicitly admitted by Wildberger 2006: I 334f.; more explicitly by Wildberger 2006: II 932 n. 1522. 108 Armisen-Marchetti (2000: 210 n. 52) quotes epist. 58.8 and 98.1. Frg. 26 Haase = F 61 Vottero is added in Setaioli 2006–2007: 361 n. 227; but many more texts could be quoted: dial. 7 (= vit. beat.).15.3, dial. 10 (= brev.).2.1, epist. 78.13, 101.9, and, most of all, nat. 2.45.2: hic [scil. Iuppiter] est, ex quo suspensa sunt omnia, causa causarum, to be compared with benef. 4.7.2: ille [scil. deus] est prima omnium causa, ex qua ceterae pendent. The expression in suspenso can mean “deferred” or “postponed” (dial. 4 [= de ira 2].22.4); for the rest, when suspensus is used absolutely it is usually referred to a psychological state of suspense (e.g., dial. 9 [= tranq.].2.10, dial. 10 [= brev.].12.5, etc.); when it governs a complement, as in the passages quoted above, or a consecutive clause, like in our text, it always means “depending (upon)” or “attached (to).” 109 Sen. nat. 2.37.3: “aut futurum” inquit “est aut non: si futurum est fiet, etiamsi vota non suscipis; si non est futurum, etiamsi susceperis vota, non fiet.” However, the reference to the doctor, further down (nat. 2.38.4), is a sure sign that Seneca also had the traditional form of the argument in mind. 110 It is clear from Cic. fat. 30 that for Chrysippus the “Idle Argument” was valid only for the fata simplicia. 111 Sen. nat. 2.38.1: “hoc quoque” inquit “ipsum necesse est fato comprensum sit, ut aut suscipias vota aut non.” 112 Cic. fat. 31 may not be Carneades’s reply to Chrysippus’s refutation of the “Idle Argument,” as maintained by Bobzien 1998a: 231 n. 126; surely, however, these words by Seneca’s interlocutor, though formally reminiscent of Chrysippus’s refutation, are really the traditional objection to any determinism, as we find it expressed in Diogenianus ap. Eus. Pr. Ev. 4.3.11 =

292

aldo setaioli

admit that the confatalia are indeed included in fate—he could not contradict Chrysippus—and finds a way out by resorting to tautology (“what is bound to happen will happen”) and postponing the discussion about what is left in hominis arbitrio.113 This term114 possibly shows that Seneca, like the Greek source of the Servius Danielinus, connected the doctrine of the confatalia with freedom of choice and decision, rather than (or at least besides) the relevance and moral import of human action—even though, as we have remarked at the beginning, he preferred to postpone this thorny discussion, and never kept his promise to tackle the problem. It may indeed be granted that in and of itself the doctrine of co-fatedness implies the autonomy of the agent in the same sense as Chrysippus’s image of the cone and the cylinder: although the co-fated conditions are themselves established by fate, they are nevertheless effected by and through us.115 It is hardly possible, however, to elicit this from Seneca’s statement that the haruspex and the doctor are ministers of fate,116 as Wildberger strives to do.117 In this text the stress lies on someone else’s activity, rather than on that of the person involved. Chrysippus118 emphasizes the agent’s effort, e.g., in calling for a doctor, but in Seneca he passively receives his healing through the physician’s agency. Seneca does stress the need for an active attitude, but it implies the general acceptance of and cooperation with fate’s plan, in the sense illustrated above, rather than the individual effort directed at acting autonomously for one’s own ends. If fate has established that someone will SVF II 939 (specifically in relation with precautions suggested by divination, as in Seneca): if we admit determinism, all is ruled by fate and nothing is in our power; cf. Sen. nat. 2.38.3: ista nobis opponi solent, ut probetur nihil voluntati nostrae relictum eqs. 113 Bobzien is totally mistaken in distinguishing the lines respectively spoken by Seneca (or the Stoic spokesman) and the interlocutor in nat. 2.37.3–2.38.3. The interlocutor’s lines are marked by inquit, according to Seneca’s usage; all the rest is spoken by Seneca (or by the Stoic spokesman). Bobzien (1998a: 204) believes the first words of nat. 2.38.1 (supra, note 111) to be spoken by Seneca and to be the restatement of Chrysippus’s refutation of the “Idle Argument” (she translates inquit as “he—i.e., Chrysippus—says”), whereas they are really the standard objection to any determinism (cf. preceding note) spoken by the interlocutor. Besides, Bobzien 1998a: 231 and n. 127 assigns the last words of nat. 2.38.2 to the interlocutor, whereas in reality they are spoken by Seneca and continue the series of examples introduced in the preceding lines. 114 Cf. supra, notes 5–9. 115 As is clear from SVF II 998 (Diogenianus). Cf. Gourinat 2005b: 234. 116 Sen. nat. 2.38.4: hoc prodest [scil. haruspex], quod fati minister est; sic cum sanitas debeatur fato debetur et medico, quia ad nos beneficium fati per huius manus venit. 117 Wildberger 2006: I 342. 118 Consistently in Cicero’s (SVF II 956), Origenes’s (SVF II 957), and Diogenianus’s (SVF II 998) reports.

ethics iii: free will and autonomy

293

become eloquent provided he studies literature, we must teach him: 119 here again the fulfillment of the co-fated condition comes through the agency of others, rather than of the party directly involved. Seneca says not that this person must learn, but that he must be taught.120 5. Chrysippus based his idea that all events are determined by fate not merely on the principle of bivalence (every proposition, including those about the future, is either true or false), which falls within the domain of logic,121 but also on causality, which is part of physics122 and is actually the indispensable foundation of the validity of the principle of bivalence.123 Fate is what is true from eternity, but it is also a series of causes, as Seneca too repeatedly remarks.124 It is a series of events preordained from eternity, unfolding for us in temporal succession—like the uncoiling of a rope, according to the image preserved by Cicero,125 or, in a formulation possibly going back to Posidonius, the flow from eternity of an eternal truth.126 Seneca’s conception is hardly different: cursum irrevocabilem ingressa ex destinato fluunt ;127 and, as Wildberger observes,128 Seneca too assumes that what is fated is true from 119

Sen. nat. 2.38.2: fatum est ut hic disertus sit, sed si litteras didicerit; at eodem fato continetur ut litteras discat: ideo docendus est. 120 Wildberger 2006: I 337 does not see this nuance. At II 922 n. 1493 she proposes correcting expiabit (varia lectio: expiat) at nat. 2.38.3: at hoc quoque in fato est, ut expiet: ideo expiabit to expiet (jussive subjunctive, following upon the preceding expiet), which would stress the agent’s direct responsibility for action, against the general posture of the context, as illustrated here. 121 As already propounded by Aristotle, De Interpretatione 9, 18a28–19b4, who, however, exempted from this principle the propositions about the future, which are not yet either true or false. For Epicurus’s position, cf. Cic. fat. 37f. An overview of the ancient debate can be found in Rist 1969: 112–120. 122 Cf. especially Cic. fat. 20 f. See the masterly treatment by Sedley 2005b; cf. also Bobzien 1998a: 59–86, Wildberger 2006: I 322–327. 123 As Inwood (1985: 216) rightly points out, in Stoicism “logic, physics, and ethics were all woven together into a virtually seamless fabric.” 124 E.g., Sen. benef. 4.7.2: quom fatum nihil aliud sit quam series implexa causarum; epist. 19.6: dicimus seriem esse causarum ex quibus nectitur fatum. 125 Cic. div. 1.127: non enim illa quae futura sunt subito exsistunt, sed est quasi rudentis explicatio sic traductio temporis nihil novi efficientis et primum quidque replicantis. 126 Cic. div. 1.125: ea est ex omni aeternitate fluens veritas sempiterna. These words are included in Posid. F 377 Theiler, whereas they are not attributed to Posidonius by EdelsteinKidd. Cf. Cic. nat. 1.40, 1.55, 3.14. 127 Sen. nat. 2.35.2. Cf. Magris 1990: 74. 128 Wildberger 2006: I 323.

294

aldo setaioli

eternity when he remarks that a true prediction cannot be made invalid by a subsequent one based on a new divinatory sign, because “nothing is truer than truth.”129 Possibly, he conflates logic and physics, too, when he states that God “wrote” fate (notice the plural, fata) once and for all,130 which must be taken to mean that God established (“formulated in writing,” as it were) certain propositions that are for ever true; but we must not forget that God also established the causes: he is the causa causarum.131 But how do logic and physics relate to ethics in this connection for Seneca? As we have seen, the Stoics maintained that granting or withholding assent to external presentations is in our power.132 After Seneca, Epictetus connects the capability of “using the presentations”133 with the power of “seeking and declining,”134 which, according to him, is a gift bestowed by God upon man. As we shall see, Seneca, like Epictetus, considers the human will to be essentially free.135 The opponents of Stoicism had accused Chrysippus of nullifying man’s will,136 although terms related to will often appear in the relevant fragments and testimonies.137 Space does not permit us to embark on a detailed analysis of these, but we must say something at least about Seneca’s conception of will. In the Stoic system, will (βούλησις) is one of the εὐπάθειαι typical of the sage.138 But, according to Pohlenz,139 Seneca opens a breach in Stoic intellectualism by laying stress on the Roman concept of voluntas. His position has recently been greatly developed by Zöller, who grounds every man’s freedom to shape his own moral character in the alleged dichotomy inherent both in the cosmos and the soul.140 Although the limits of his

129 Sen. nat. 2.34.2: vero verius nihil est: si aves futura cecinerunt, non potest hoc auspicium fulmine irritum fieri, aut non futura cecinerunt. 130 Sen. dial. 1 (= prov.).5.8: ille omnium conditor et rector scripsit quidem fata sed sequitur; semper paret, semel iussit (see below, note 166, for the reduction of God to temporal contingency in this text); and, of course, the very etymology offatum encouraged a Roman Stoic to conceive of fate as a series of uttered propositions. Cf. Pötscher 1978: 418 and n. 103. 131 Cf. Sen. nat. 2.45.2 (supra, note 108), also note 124. 132 Cf. SVF II 974–1007. 133 Epict. diatr. 1.1.12: τὴν δύναµιν […] τὴν χρηστικὴν ταῖς φαντασίαις. 134 Epict. diatr. 1.1.12: ὀρεκτικήν τε καὶ ἐκκλιτικήν. 135 Cf., e.g, Epict. diatr. 1.17.28: ἐὰν θέλῃς, ἐλεύθερος εἶ, closely paralleled by Sen. epist. 80.4: quid opus est ut bonus sis? velle!; cf. epist. 31.5: quid votis opus est? fac te ipse felicem. 136 Cf. Cic. fat. 9 (= SVF II 951), Gell. 7.2.5 (= SVF II 1000), Sen. nat. 2.38.3. 137 E.g., SVF II 998 (Diogenianus: βούλεσθαι is referred to the effort of the agent in the doctrine of co-fatedness, as voluntas in Sen. nat. 2.38.3), Gell. 7.2.8: voluntario impetu, 7.2.11: voluntas. 138 Cf. SVF III 431, 432, 435 (= Sen. epist. 59.2), 437, 438, 517 (= Sen. epist. 95.7). 139 Pohlenz 1967: II, 89 f. 140 Zöller 2003: 231f. Zöller devotes a whole chapter to Seneca’s alleged psychological

ethics iii: free will and autonomy

295

work have been well illustrated by Wildberger 141 and his positions are often problematic, we may agree with his statement142 that in Seneca character shaping and moral improvement are clearly connected with (good) will. It is hardly important, for our purpose, to determine whether will is or is not an autonomous mental faculty in Seneca;143 we should rather observe that for him it is hardly a εὐπάθεια of the accomplished sage,144 but rather a natural endowment that must be trained and improved. We must be aware of the fact that in Seneca terms like velle, voluntas, voluntarius do not always convey the same meaning.145 What is important here is the will to improve oneself morally, whose freedom is taken for granted by Seneca. At the lowest moral level, this will may be lacking,146 but under normal circumstances it does not need to be learned;147 what can and should be learned is the correct use of will.148 Will can lack direction,149 but if properly channeled it becomes essential to moral improvement.150 At this stage it is already bona voluntas, which can be a mere psychological state not yet

dualism (pp. 140–153). For monism vs. dualism in Stoic psychology, see Inwood 1985: 33– 37, 131–143, 155–165, 176–180 (about Seneca). Cf. also Inwood 2005a: 23–64 and my review: Setaioli 2007a: 689f. Zöller relies on Sen. epist. 71.27, 92.8, dial. 3 (= de ira 1).8.1–3 to uphold Seneca’s psychological dualism. Although I do believe that in Epistle 92 there are elements of dualism (cf. Setaioli, supra, p. 246), it should be noticed that in Epistle 92.3 the concept of voluntas is connected with the rational part of the soul. 141 Wildberger 2006: II 925 n. 1498. 142 Zöller 2003: 146. 143 Cf. Inwood 2005a: 132–156 and Setaioli 2007a: 691f. For a discussion of the rich bibliography, see Zöller 2003: 19–46; also Wildberger 2006: II 924 f. n. 1497. 144 Even though Seneca envisages this, too: epist. 116.1: cum tibi cupere interdixero, velle permittam. 145 Cf. Voelke 1973: 162f. Inwood (2005a: 132–156) detaches the investigation of Seneca’s concept of will from his use of velle and related words. We must indeed be wary of Seneca’s use of terminology. At dial. 3 (= de ira 1).20.6 he criticizes Livy for writing vir ingenii magni magis quam boni, on the grounds that goodness and greatness cannot be separated. But at dial. 6 (= cons. Marc.).16.3 he himself says of the Gracchi: etiam qui bonos viros negaverit magnos fatebitur. 146 Sen. epist. 116.8: nolle est in causa, non posse praetenditur. Clearly, the lack of good will is seen as a free choice. With a different nuance, cf. also Seneca’s translation of Cleanthes’s verses, epist. 107.11 (fac nolle: v. 3). Will can also lead to the worst form of slavery (epist. 47.17) and impel us to passion (e.g., dial. 3 [= de ira 1].8.1). 147 Sen. epist. 81.13: nemo referre gratiam scit nisi sapiens. stultus quoque, utcumque scit et quemadmodum potest, referat; scientia illi potius quam voluntas desit: velle non discitur. The fool’s will is an impulse (caused by a ὁρµητικὴ φαντασία), which is there, but must be perfected, as we shall presently see. 148 I.e., to which φαντασίαι we should grant our assent. Cf. Wildberger 2006: I 339f. 149 Sen. epist. 21.1: quid velis nescis, melius probas honesta quam sequeris; cf. 37.5, 52.1. 150 Sen. epist. 71.36: magna pars est profectus velle proficere. Cf. epist. 80.4.

296

aldo setaioli

having attained bona mens,151 but is the necessary condition for attaining it152 and is still found in those who have already reached that goal.153 The next stage is the condition of the man who has greatly progressed on the path of moral improvement, whose will is already fairly unswerving, though not yet perfectly consistent like the sage’s.154 The final stage is the recta voluntas, the upright will, which proceeds from an upright mind and unfolds in upright action.155 This goal can be reached only because the will to progress morally is free and virtue is teachable, as Seneca repeats in this very context.156 At this point will becomes unalterable: firma voluntas.157 Discussing this latter passage we have seen that this immutability does not destroy the wise man’s will158—it actually makes it perfect.159 Perfect will is the one that always wills the same, and it is tantamount to wisdom; 160 such is the will not only of the wise man, but also of the gods.161 We may conclude that our earlier remarks were right: the belief in both freedom of will and the meaning of goal-directed activity are necessary assumptions for Seneca in order to carry out his mission as a self-appointed spiritual director aiming at the moral improvement of his addressees as well as of his own self.162 He could never have accepted the idea that moral teaching and advice—suasion and dissuasion163—are useless. But Seneca’s ethics does not bring about any real breach in the Stoic system. When the final stage of the training of will is attained, the wise man’s will

151 Sen. epist. 72. 9: habet aliquis bonam voluntatem, habet profectum, sed cui multum desit a summo. 152 Sen. epist. 16.1: perseverandum est et adsiduo studio robur addendum, donec bona mens sit quod bona voluntas est. Baldarotta (1994: 28) misunderstands this passage: “occorre un assiduo impegno ed esercizio perché il pensiero retto e buono (bona mens) si tramuti in buona volontà.” 153 Cf. Sen. benef. 4.21.6, 5.3.2. 154 Sen. epist. 35.4: mutatio voluntatis indicat animum natare […] non vagatur quod fixum atque fundatum est: istud sapienti contingit, aliquatenus et proficienti provectoque. 155 Sen. epist. 95.57: actio recta non erit nisi recta fuerit voluntas; ab hac enim est actio. rursus voluntas non erit recta nisi animus rectus fuerit; ab hoc enim est voluntas. In Stoic terms, the recta voluntas proceeds from the ὀρθὸς λόγοc and unfolds in the κατόρθωµα. 156 Sen. epist. 95.56: virtus et aliorum scientia est et sui; discendum de ipsa est ut ipsa discatur. 157 Sen. benef. 6.21.2, a passage we have already discussed. 158 Cf. supra, notes 72 f. 159 Sen. benef. 6.21.4. Cf. Zöller 2003: 220 f. 160 Sen. epist. 20.5: quid est sapientia? semper idem velle et idem nolle. Atepist. 35.4 and 52.1 consistency of will is indicated as a goal. 161 Sen. benef. 6.23.1, and the whole context. Cf. nat. 1 pr. 3; for dial. 1 (= prov.).5.8 see below, note 166. 162 Cf. Setaioli, supra, pp. 239–256 and supra in this chapter. Scholars who deny that such was the purpose of Seneca’s writing (e.g., Inwood 2005a: 144) miss the basic point of his philosophy. 163 The προτροπαί and ἀποτροπαί of SVF II 984. Cf. supra, §3.

ethics iii: free will and autonomy

297

is not directed to his own, private improvement any more: it has become one with the will of God on a cosmic scale; both are immutable, because both will the best. Man’s freedom has become the same as God’s freedom,164 who cannot but follow what he, as perfect Reason, has established from all eternity. Assimilating oneself to God is indeed the final goal of human life,165 and, according to Seneca, the promise of philosophy.166 Man’s acceptance of God’s will as his own includes putting up with the dark side of fate, which cannot be grasped by the human mind: τύχη or fortuna.167 Theoretically, Seneca himself equates it with fatum and natura as one of the names or aspects of God.168 In practice, however, fortuna and related terms are employed by Seneca in reference to events that are not merely unexpected, but also unreliable and often disagreeable, although the wise man considers them to be “indifferent,” literally “making no difference” at the moral level, which is the only one that matters. Seneca’s attitude toward fortuna goes beyond καρτερία, the brave bearing of misfortune, which was one of the subordinate forms of ἀνδρεία, fortitude,169 and is very often defiant. If, in a letter, he quotes Virgil to express the manly acceptance of misfortune,170 in another he corrects him: if the gods have decided contrary to human expectations, theirs is the better decision.171 As a consequence, fortuna is often 164 Sen. nat. 1 pr. 3: necesse est eadem placere ei cui nisi optima placere non possunt, nec ob hoc minus est liber ac potens: ipse est enim necessitas sua; benef. 6.23.1f. We have seen that, just like God, the wise man too establishes his own necessity: benef. 6.21.3 (supra, note 73). 165 Cf. Setaioli 2006–2007: 363, Wildberger 2006: I 271 f. 166 Sen. epist. 48.11: hoc enim est quod mihi philosophia promittit, ut parem deo faciat. Man is equipped to reach this goal: epist. 31.9. It must be said, however, that not only does Seneca raise man to the level of God, he also occasionally lowers God to the level of man. At dial. 1 (= prov.).5.8 he writes: ille omnium conditor et rector scripsit quidem fata, sed sequitur; semper paret, semel iussit. As Canfora (1999: 24) observes, God is thus subjected to the same temporal contingency as man. Wildberger 2006: I 50 (cf. II, 556f. n. 314) tries to explain Seneca’s formulation. Cf. also benef. 4.32.1, 6.23.3–5. Elsewhere, Seneca unequivocally follows the orthodox conception of God in terms of eternity, not of temporal succession: e.g., nat. 2.36 illius [= dei] divinitati omne praesens sit; cf. Cic. fat. 1.128: sunt enim omnia, sed tempore absunt. 167 For the τύχη as ἄδηλος αἰτία cf. SVF II 965–973. For the Stoics, of course, mere chance is unthinkable: cf., e.g., Wildberger 2006: I 47 f. 168 Sen. benef. 4.8.3: sic nunc naturam voca, fatum, fortunam: omnia eiusdem dei nomina sunt varie utentis sua potestate. Rozelaar (1976: 457) considers this passage to indicate that there is a difference between fatum and fortuna. 169 Cf. SVF III 264, 265, 269, 275. Seneca, of course, has a notion of this too: cf., e.g., epist. 65.5, dial. 1 (= prov.).5.7, etc. Baldarotta (1994: 24) considers the term fortuna to be a vox media in Seneca. As far as καρτερία is concerned, the figure of the wise man unconquered by fortune had been introduced long before Seneca: SVF I 449 (Persaeus of Citium): ὁ σοφὸς ὑπὸ τῆς τύχης ἀήττητός ἐστι καὶ ἀδούλωτος κτλ. 170 Sen. epist.76.33–35, quoting Verg. Aen. 6.103–105. 171 Sen. epist. 98.5, where Virgil’s dis aliter visum—Aen. 2.428—is corrected to di melius.

298

aldo setaioli

personified as the enemy with whom the wise man must wage an unrelenting fight.172 Freedom results from accepting all of fate as god’s will and our own; we should bravely and freely offer ourselves to fortune as we do to fate.173 But perhaps Seneca’s agonistic attitude conceals more than a difference between the two—fate and fortuna—merely due to the limitations of man’s understanding. Chrysippus had taught that good cannot exist without evil, but for him the latter’s function is mainly to make the former stand out more clearly.174 In his famous hymn to Zeus, Cleanthes had claimed that the misdeeds of evil men cannot be laid at God’s door.175 In some fragments of Chrysippus’s, however, the presence of evil is explained in different ways, several of which reappear in Seneca. What appears to be bad may really be good in the general economy of the cosmos; 176 or it may be a test for man; 177 and some fragments hint at some limitation of God’s power.178 This reappears several times in Seneca: evil and imperfections are due to God’s want of effectiveness. He may be simply following Chrysippus,179 but the occurrence of the same theme in two letters clearly influenced by Platonism180 leads us to suspect that non-Stoic ideas may also be at play in this conception. Investigating this aspect exceeds the limits of this chapter, but what is clearly implied is that at times fortuna may be an intrinsically irrational element, rather than one whose rationality cannot be grasped by the human mind. If this is so, then man’s triumph over fortuna is not merely the guarantee of

172 The masterly treatment by Traina 1976: 9–13 (introd.) may be mentioned in this connection. Cf. also Rozelaar 1976: 169f., Busch 1961. The texts that might be quoted are countless. 173 The two following texts, occurring one shortly after the other in the same work should be placed side by side: Sen. dial. 1 (= prov.).4.12: praebendi fortunae sumus; 5.8: quid est boni viri? praebere se fato. Cf. Niem 2002: 190, 203. 174 SVF II 1169, 1170, 1181. Cf. Gourinat 2005b: 240. 175 SVF II 537, p. 238, 13 von Arnim; but God can turn even these misdeeds to the good. Cf. SVF III 1184. 176 SVF II 1171, 1176, 1181, 1184. Cf. Sen. epist.74.20: sciatque illa ipsa quibus laedi videtur ad conservationem universi pertinere; 95.50: [deos] qui humani generis tutelam gerunt interdum curiosi singulorum; dial. 1 (= prov.).3.1: universis, quorum maior dis cura quam singulorum est. 177 SVF II 1173, 1181. Cf. Seneca’s whole De providentia. 178 SVF II 1178, 1182, 1183; cf. 1180. 179 As assumed by Niem 2002: 86–88, in relation to Sen. dial. 1 (= prov.).5.9: non potest artifex mutare materiam. At dial. 1 (= prov.).6.6 god himself says: non poteram vos istis [the misfortunes] subducere. Epict. diatr. 1.1.12 has God say the same: οὐκ ἐδυνάµην. At nat. 1 pr. 16 Seneca envisages the possibility that matter may prevent God from creating perfect beings (cf. SVF II 1170). 180 Cf. Sen. epist. 58.27f., 65.10 (a peculiar translation of Plat. Tim. 29de). See Setaioli 2006– 2007: 344–346. Cf. also Dragona-Monachou 1994: 4439, 4442.

ethics iii: free will and autonomy

299

his freedom181—just as his defeat entails the most grievous servitude182—but also his contribution to the perfection of this world by fully restoring the impaired rationality of the divine plan.183 His freedom will then not be limited to himself—being his own, the undisputed master of himself184—but it will expand to a truly cosmic scale. 6. Stoicism allowed the split between interiority and exteriority, between man and the transcendent, to be overcome. Man can, of course, demarcate a space of his own within whose boundaries he will be totally free—Seneca is in many ways the philosopher of the inner self; but if we attune our own λόγος to the λόγος of the cosmos, of which ours is part, not only shall we accept the divine plan as our own will, but we shall fulfill our natural goal as rational beings and be at one with God and as free as he is. We shall then experience the exhilarating feeling of being “swept with the cosmos”—cum universo rapi.185 Our life will then be in agreement with the Stoic ideal: “living consistently,” our own will being in perfect and constant unison with God’s. Being good is possible only if our will and action are unalterably directed toward the good.186 When this goal is reached, the opposition between freedom and determinism becomes totally meaningless—and Seneca never doubted that it was up to us to attain it.

181

Cf., e.g., Sen. epist. 51.8 f. On this passage, see Lefèvre 1983: 66–68. Cf. Sen. dial. 7 (= vit. beat.).15.3. 183 Cf. Setaioli 2006–2007: 365. For a fuller treatment, cf. Setaioli, infra, pp. 384–388, 397–401. 184 Cf., e.g., epist. 75.18. See the masterful work by Traina 1987. If one is not in possession of oneself, one will again fall under a grievous servitude (Sen. nat. 3 pr. 16f.), just as if one is under the sway of fortuna (cf. supra, note 182). There is no need to assume that expressions like suum esse must correspond to the concept of ἐφ’ ἡµῖν, as Wildberger (2006: I 341) does. 185 Sen. dial. 1 (= prov.).5.8. 186 Cf., e.g., Sen. epist. 20.5, 34.4, 35.4, 95.58, 120.10 f. 182

ETHICS IV: WISDOM AND VIRTUE*

Jula Wildberger In Seneca’s prose writings the concepts “wisdom,” “virtue,” and related ideas are negotiated in a complex system of many coordinates. There are, for example, the debates within the inner circle of philosophy: Seneca and other Stoics defend their views against Epicureans, Peripatetics, Cynics, Academics, Sceptics, and Platonists—or against their fellow Stoics with differing positions.1 If we regard Stoicism as a practical enterprise, there is the tension between the ideals of a philosophical theory and the real life of ordinary people. This tension pervades Seneca’s œuvre and is one of the driving forces behind its dialogical, epistolary, and hortatory form and its focus on friendship, the self, and exemplarity.2 When we look at Seneca as a Roman philosopher, we observe a tension between Stoic conceptions of complete agreement and absolute human perfection, on the one hand, and, on the other hand, the Roman discourse of manliness, embodied in the hallowed moral examples of old.3 Yet we also encounter new ideas about a man’s excellence that had been developing since the late Republic. During the Neronian period, in particular, male members of the Roman elite found ample encouragement to strive for aesthetic, cultural, or intellectual refinement.4 Seneca opposes tendencies to seek status and * This article was submitted for publication in December 2008. When the author received the proofs in April 2013, only a partial and fragmentary bibliographical update was possible. 1 Quite a number of scholars see elements of psychological dualism and Platonism in Seneca’s philosophy (discussions in Inwood 1993, Gauly 2004, Bonazzi and Helmig 2007, Harte et al. 2010, and in chapters “Physics I” (Smith, pp. 343–361) and “Physics II” (Gauly, pp. 363–378) of this volume). The outline given here implies that Seneca is not a dualist. On wisdom and virtue in Seneca, see also: Ganss 1952, Cancik and Cancik-Lindemaier 1991, Wyszomirski 1993, Torre 1995, Classen 2000, Isnardi Parente 2000, Cambiano 2001, Russell 2004, Jedan 2009. I have not read Fortner 2002. 2 See, e.g., Cancik 1967, Hadot 1969, Inwood 1995 and 1999, Hengelbrock 2000, Setaioli, supra, pp. 239–256. I fully agree with Setaioli, who insists that Seneca does not break with orthodox Stoic conceptions, and hope to show more clearly below why the Stoic conception of the good does not admit of an eclectic compromise. All the same, there is a tension between ideal and practice—if only the tension that the ideal may never be fulfilled (Sen. epist. 42.1, Brouwer 2002). 3 On moral exempla in Seneca, see, in particular epist. 24 and 120 and Mayer 1991. I use the word “manliness” and not “masculinity” to capture both connotations of Latin virtus: that one is a real man and that one is brave and manly. 4 See, e.g., Cain 1993, Edwards 1993, Krostenko 2001.

302

jula wildberger

advancement through conspicuous consumption and pastimes that were associated with femininity and softness.5 However, he, too, promotes an innovative model of the active political life a Stoic should lead. Paradoxically, Seneca’s Stoic life unfolds in private retreat, in exchanges with friends and readers, outside the traditional spheres for practicing Roman virtues.6 It is obvious that not all of these coordinates can be explored within a short chapter. Therefore, I intend to focus on two aspects that I believe to be best suited for fleshing out the peculiarities of Seneca’s approach. First, I will try to outline his theoretical stance, both against the backdrop of the Stoic discourse in general and with regard to his engagement with Epicureanism. Second, I will point out how Seneca uses traditional ideals of manliness and also raise the question to what extent this may be more than just a veneer to attract a compatriot audience.7 1. The Physics of Virtue For a Stoic, speaking about wisdom and virtue means speaking about happiness (εὐδαιµονία), which in Seneca’s and Cicero’s translation becomes a happy life (vita beata).8 In this context, Greek Stoics distinguish between the highest good (σκοπός) and the end (τέλος) “for the sake of which everything is done, but which is not itself done for the sake of anything.” The end is “to obtain happiness, which is the same as being happy.”9 It is an incorporeal predicate (κατηγόρηµα), an effect caused by the highest good, which is a three-dimensional body. Seneca is well aware of this difference (e.g., epist. 117.1–17)10 but prefers not to take explicit account of it. He uses the expression “the highest good” 5

See, e.g., nat. 7.32 and dial. 10 (= brev.).12. See, e.g., Williams 2003, Gauly 2004. 7 Ganss (1952) and Wyszomirski (1993) offer comprehensive overviews that can be used as supplements to the selective presentation of key elements given in this chapter. 8 Görler 1996: 163. The change was due to aesthetic reasons (Cic. nat. 1.95, Quint. inst. 8.3.32), but it reflects well the particular Stoic quality of happiness. It is the good life and thus an activity (see, e.g., Sen. epist. 67.7), not just a desirable emotion or a state like Epicurean freedom from pain. Although the Stoics too ascribed a set of special “affective responses” to their sage (see M. Graver, supra, pp. 257–275), they regarded these not as essential features of happiness, but only as a concomitant positive effect of it (compare epist. 59.16). 9 Ar. Did. in Stob. 2, p. 77 f. Wachsmuth, trans. Long and Sedley 1987 (63A1). See also p. 76, lines 16–23. 10 In this letter, Seneca presents a complex analysis of the difference between sapere and sapientia. Detailed discussions are to be found in Cooper 2004: 324–332, Wildberger 2006: 161–178, Inwood 2007a: 288–301; see also Wyszomirski 1991. 6

ethics iv: wisdom and virtue

303

(summum bonum) indiscriminately for both the highest good and end,11 as in the following passage: […] that there is completed virtue (perfecta virtus), an evenness and continuity of a life that is in harmony with itself throughout everything, something which cannot happen unless one attains knowledge about things (rerum scientia) and a systematic method (ars) by means of which [such] knowledge about human and divine things can be acquired. This is the highest good.12

After describing the Stoic end, Seneca refers back to what he callssummum bonum with the ambiguous demonstrative pronoun “this” (hoc). The reader cannot decide whether the pronoun’s referents are the complete subordinate clauses that contain the descriptions of the highest good or only the subjects of the clauses, i.e., whether the thing called summum bonum consists of incorporeal predicates such as “having completed virtue” and “attaining knowledge” or of bodies, to which a Stoic refers when using nouns like “virtue” or “knowledge.” As so often, Seneca’s terminological looseness exploits a peculiarity of the Stoic system with surprising precision. The pronoun “this” can very well refer to both highest good and end because the one cannot occur without the other. This is a consequence of two distinctive features of the Stoic theory of causation. First, the Stoics distinguish a particular class of incorporeal things, the sayables (λεκτά), that breach the gulf between language and the physical world. Predicates are the most basic form of sayable. If they actually belong to a body, e.g., the predicate “being wise,” which belongs to a person who actually is wise, they are at the same time effects.13 Second, orthodox Stoics defended a rigorously reductionist theory of causation, acknowledging only one type of cause: the efficient cause that has its effect through physical contact, by touching the body on which it has its effect.14 This means that effects can only obtain as long as a cause for them is physically present. Accordingly, the reason why someone has a particular good is the physical presence of that good itself. The predicates “having virtue” or “being virtuous”

11 For the technical sense of summum bonum, see, e.g., epist. 9.1, 21.4, 66.45, and dial. 7 (= vit. beat.) passim. The Ciceronian term finis is rarely used in this sense, and usually with a clarifying attribute: epist. 71.4 and 95.45: finis summi boni; epist. 78.25: fines bonorum ac malorum; 76.10: naturae suae; epist. 124.23: suus finis; epist. 52.4 (not quite in a technical sense); epist. 93.8: maximus finis (playing with the frequent use of finis in the sense of death). 12 Sen. epist. 31.8. Unless indicated otherwise, translations are my own. 13 Ar. Did. 18 in Stob. 1, p. 138 f. Wachsmuth; see also Mansfeld 2001a and 2001b. 14 Seneca discusses causation in Epistle 65, where he defends Stoic reductionism (see also the discussion by Gauly, infra, pp. 363–378). On the necessity of physical contact see Sen. epist. 106.8, 117.7, 117.10, Wildberger 2006: 13 f.

304

jula wildberger

are caused by a virtue that is physically present within the person in question. A wise person is wise because a body, a wisdom, is active in that person as the cause of effects such as “is wise” or “acts wisely.”15 While, for the reasons indicated, the distinction between corporeal highest good and incorporeal end might not be of prime importance for understanding Stoic ethics, it is vital to keep in mind that a Stoic, when talking about virtue, wisdom, or the good life, does not discuss just a set of abstract rules or an ideal image but a certain class of three-dimensional bodies. Virtue is a qualitative state, and, for a Stoic, a qualitative state is the same as a body in a qualitative state. In the case of virtue, this body is the mind. When discussing Stoic virtue, one is discussing tokens of perfectly developed rational minds. […] virtue is nothing else but the mind in a certain state.16 Virtue is an unintensifiable disposition in agreement (διάθεσις ὁµολογουµένη). […] And happiness lies in virtue, inasmuch as virtue is the soul so made as to produce the agreement of one’s whole life.17

In the second of the two definitions quoted above, the genus of virtue is given as “unintensifiable disposition” (διάθεσις). Unintensifiable dispositions are a particular class of qualitative state that “do not admit intensification or relaxation” and cannot become more or less what they are. The example given in our source18 is the disposition “straightness” of a stick. If you bend a stick, it becomes curved, not less straight: it loses its disposition “straightness” (εὐθύτης) altogether. Playing with the polysemy of the Latin word rectus, which means “straight” (εὐθύς) as well as “right” (ὀρθός), Seneca uses the same example to illustrate this central feature that distinguishes the Stoic conception of virtue from those of other schools, namely that, as an unintensifiable disposition, virtue is something absolute, either completely and perfectly present or not present at all.

Ar. Did. 18 in Stob. 1, p. 138, lines 16–22 Wachsmuth (report of Zeno’s views). Sen. epist. 113.2: virtus autem nihil aliud est quam animus quodammodo se habens. A Greek version of the same idea (S. Emp. adv. math. 11.23) indicates that animus must be the leading part of the soul in this case. The leading part of rational beings is called “mind” (διάνοια). In Epistle 113.2 Seneca reports other Stoics’ arguments that virtue, being a human (or divine) mind, must be an animal. Seneca challenges this view in this letter. Yet, in Epistle 113.7, while playing the role of an interlocutor to his Stoic opponent, he accepts the quoted statement as true and identifies “the mind disposed in a certain manner” with “a qualitative state (habitus) of the mind and a certain power (vis).” (For Seneca’s solution to the question see Wildberger 2006: 90 f., Inwood 2007a: 286.) The mind itself is defined as spiritus quodammodo se habens in epist. 50.6. See further Smith, infra, pp. 343–361. 17 Diog. Laert. 7.89, trans. Gerson and Inwood (II-94), slightly altered. 18 Simpl. in Cat. p. 237 f. Kalbfleisch. The quotation is from p. 237, lines 30f. 16

ethics iv: wisdom and virtue

305

[…] that only this is a good: what is honorable; that this thing cannot be slackened or intensified, no more than you would bend the ruler (regula) generally used to test whether something is straight (rectum). Whatever you change in it is a violation of its straightness. The same we will therefore say about virtue: virtue, too, is straight/right (recta) and does not admit bending. (epist. 71.19f.)

Taken for itself, the image of the ruler is misleading. One might mistake straightness for stiffness and immobility. Yet the Stoic conception of an unintensifiable disposition does not exclude qualitative change. Having described the kind of mind that is a virtue (epist. 66.6), Seneca distinguishes between the different qualities and movements of such a mind and its absolute excellence and goodness, any change of which would destroy the virtuous disposition of that mind. On the other hand, there are many types of virtue, which unfold themselves according to the various situations in life and different actions. Virtue itself becomes neither smaller nor greater. For the highest good cannot decrease, and virtue cannot go back [scil. to a previous state of imperfection]; instead it transforms itself into another quality all the time, adapting itself to the shape of actions which it is about to perform. […] Therefore, the power and greatness of virtue cannot rise any further, as there is no increase to what is the greatest. Nowhere will you find something that is straighter than what is straight (rectius recto), just as nothing is truer than what is true or more balanced than what is balanced out. (epist. 66.7f.)

With statements such as these Seneca tries to explain the distinctive quality of a Stoic good as something that can neither be intensified nor diminished. Arguably, this quality can be called the theoretical core of Stoic ethics, and Seneca accords it due prominence by making it the subject of the first and, at the same time, most extensive technical discussion of a Stoic tenet in the Epistulae morales.19 2. Virtue, Happiness, and Nature Like other Stoics, Seneca insists that there are certain absolute goods with absolute value, belonging to a category of their own and completely different from what they call “preferred indifferents” (ἀδιάφορα προηγµένα), such as good health, a long life, intelligence, or safety and sustenance. These 19 With 13 Oxford pages, Epistle 66 on the equality of Stoic goods is longer than any letter before it. The debate continues, in particular in Epistles 67, 71, 74, and 76. Three of these (66, 71, and 76) are discussed in Inwood 2007a. On Epistle 66, see also Hachmann 2006.

306

jula wildberger

indifferents have some relative value, which can be intensified and of which there can be more or less, according to the circumstances in which they become available for choice (ἐκλέγειν) among various alternatives.20 But only virtue is to be sought (αἱρεῖσθαι) under all circumstances and without any reservation or qualification.21 The value of virtue is always the same and of such a different nature that Stoics contend that virtue is not only sufficient to achieve happiness but the only thing that makes a person happy. Even in the short definitions of the Stoic end, which we find in doxographic writings, the close connection between virtue and happiness, i.e., the good life, is evident. The following is the most extensive doxography we have of early Stoic definitions and accounts of the end:22 Therefore, Zeno in his book On the nature of man was the first to say that living in agreement with nature (ὁµολογουµένως τῇ φύσει ζῆν) is the end, which is living in accordance with virtue (κατ’ ἀρετήν). For nature leads us toward virtue. So, too, Cleanthes, in his book On pleasure, and Posidonius and Hecato, in their books On ends. Further, living in accordance with virtue is equivalent to living in accordance with experience of what happens to/by nature, as Chrysippus says in On Ends, Book 1. For our own natures are shares (µερή) of the nature of the whole. Therefore, living following the lead of nature (ἀκολούθως τῇ φύσει) comes to be the end, which is in accordance with the nature of oneself and that of the whole, engaging in no activity usually forbidden by the universal law, which is the right reason pervading everything and identical to Zeus, who himself is identical with the director of the administration of existing things. And the virtue of the happy man and his good flow of life are just this: always doing everything on the basis of a concordance with the divinity (δαίµων) that is with each man [i.e., his own divine mind] with23 the volition of the administrator of the whole.24

The text distinguishes clearly between a foundational definition introduced by Zeno and later developments and explications that seem to go back to Chrysippus. In Seneca we find reflections of both: “living in agreement with nature” is expressed with various composite forms of sentire, e.g., naturae assentiri. Chrysippus’s “living according to the lead of nature” is rendered most closely with naturam sequi.25 20

See, e.g., Diog. Laert. 7.104 f., Sen. epist. 31.6. On the reservation with which the sage selects indifferents, see Ar. Did. in Stob. 2, p. 115 Wachsmuth, Wildberger 2006: 273 with 871 n. 1324. 22 An important parallel account is given by Ar. Did. in Stob. 2, p. 75f. Wachsmuth. 23 The Greek preposition πρός with the accusative indicates that the divinity that is within each man is attuning itself to the volition of the administrator and not the other way around. 24 Diog. Laert. 7.87 f., trans. Long and Sedley 1987 (63C1–4), with alteration. 25 The reason for this change of terminology, from agreement to following a lead, is 21

ethics iv: wisdom and virtue

307

The Stoic happy life is a virtuous life. As we learn from the first part of the quoted passage, this is so because the happy life is a life in agreement with nature, while nature itself leads human beings toward virtue. According to Chrysippus’s influential discussion, which is summarized in the second part of the quoted passage, the word “nature” in the definition of the end refers to two things: to the nature of the whole, i.e., the principle God26 in a certain function, and to the individual nature of every single human being. When using the term “nature” to refer to the principle God, or shares of this principle that constitute individual bodies within the cosmos, Stoics emphasize that the principle God, or the share of it, is a life form, i.e., something that is moving itself out of itself in an ordered way and is able to give birth to things that are similar to itself. Accordingly, individual natures, which are such procreated things, behave in the same manner as the natures from which they were procreated (Diog. Laert. 7.148). A nature is a body that has an inherent aim and drives the living being to goal-directed movements. The Stoic concept of nature is what comes closest to our modern concept of life. We might also say that the aim inherent in a nature is a biological function. Each life form has a share of God, a nature of a particular kind, and with this nature it has its particular biological function. To perform this function is to live according to one’s own nature, and the life form that is performing it to perfection has attained the kind of virtue accessible to that class of life forms and thus its highest possible good.27 That this holds true of the human good as well is enhanced in the definitions given by Panaetius (frg. 96 van Straaten) and anonymous postPoseidonian Stoics: In addition to these men [scil. earlier Stoics] Panaetius declared the end to be “living in accordance with the tendencies bestowed on us by nature.” […] Some of the later Stoics represented it as follows: “The end is living in agreement with the constitution of man.”28

The various species and classes of living beings on earth—plants, beasts, humans—all have their particular form of perfection, their propria virtus, as Seneca calls it (epist. 41.7). Now, human nature is rational. “And since reason

explained in Wildberger 2006: 273 f. For Senecan translations of the different terms, see Wildberger 2006: 872–874 n. 1325. 26 In modern discussions, this principle is often called “Logos” by one of its other appellations; in a wider sense, the word “god” is also used to refer to the cosmos as a whole. 27 For further details and Seneca’s as well as Hierocles’s discussions of οἰκείωσις (“appropriation”), see Smith, infra, pp. 343–361, and Bees 2004. 28 Clem. Al. strom. 2.21.129.4 f., trans. Long and Sedley 1987 (63J1. 3).

308

jula wildberger

[…] has been bestowed on rational beings, to live correctly in accordance with reason comes to be natural for them. For reason supervenes as the craftsman of impulse”29—impulse being the highest faculty of the nature of beasts. Thus, like his fellow Stoics, Seneca conceives of human virtue as a state at the top of a hierarchy of natural functions: Praise in a man what is his very own, what can neither be taken away nor given to him, what is particular to man and his very own possession (proprium). You ask what it is? It’s the mind and perfected reason within this mind. Man is a rational animal. Therefore, that which is good for him reaches its highest degree of perfection once he has completed what he was born for.30

Their own natures lead human beings toward virtue (Diog. Laert. 7.88) by setting inherent goals for themselves. According to its nature, every living being has some things it can do, some biological faculties and functions, and it is part of that nature that the living being develop these faculties and actually perform the functions it is made for. In man, this constitutes the faculty “reason” and the functions that derive from this faculty. Reason is something that humans share with the gods and the principle God; it is the highest faculty any nature can have. Humans are created by God to behave in the same manner as the nature from which they were created, i.e., God.31 This is what the happy and virtuous man does when he always acts in such a way as to achieve concordance of his own divine mind “with the volition of the administrator of the whole” (Chrysippus in Diog. Laert. 7.88), and it is also what Seneca is talking about when he says: “What is reason, then?”—The imitation of Nature.—“And what is man’s highest good?”—To behave according to the volition of Nature. (epist. 66.39)

3. Virtue and Wisdom The biological functions—in Stoic terms, the proper functions (καθήκοντα)— of humans and beasts are different because their highest faculties are different. This is why human beings should not seek examples for their behavior in the animal world, “when you have at your disposal the divine cosmos, which man, alone of all animals, can understand in order alone to imitate it.”32 Passages such as this one show the close connection between 29 30 31 32

Diog. Laert. 7.86, trans. Long and Sedley 1987 (57A5). Sen. epist. 41.8; compare, e.g., epist. 76.10, 124.23. A more detailed account of this is given in Wildberger 2006: 225–231. Sen. dial. 4 (= de ira 2).16.2, trans. Cooper and Procopé 1995.

ethics iv: wisdom and virtue

309

virtue and knowledge both for Seneca and for the Stoics in general. If one is to follow “the volition of the administrator of the whole,” one must know what that administrator wants.33 And if one is to act rationally in accordance with one’s own nature, one must have a correct understanding of that nature, too. It is therefore not surprising that the four cardinal virtues (justice, wisdom, self-control, and courage) were defined as knowledge (ἐπιστήµη) of certain fields of human action.34 Seneca, too, defines one of them, courage, as knowledge (scientia) of dangers and of what is good and bad (benef. 2.34.3, epist. 85.28). However, in a striking passage in Epistle 90, he goes far beyond any extant statement by any other Stoic. Whereas the traditional account assumes that, in principle, every human being can develop into a sage, Seneca explicitly denies that early man could have been in possession of real virtues. They were innocent because of their ignorance of things. It makes a big difference whether someone does not want to do wrong or does not know how. They lacked justice, they lacked practical wisdom, they lacked selfcontrol and courage. Their primitive life had some aspects that were similar to these virtues. Virtue [itself] can only be achieved by a mind that has been instructed, thoroughly taught, and brought to the highest point [scil. of complete perfection] by means of continuous training. (epist. 90.46)

At that primitive time, when philosophy had still to be invented (compare 90.1), when there was no “systematic method by means of which knowledge about human and divine things [could] be acquired” (epist. 31.8), it was not yet possible to develop in oneself the intricate technical knowledge of which a virtue consists. This, at least, is Seneca’s view: the very same intellectual progress that teaches human beings vices like greed and injustice is also the necessary prerequisite for real goodness. Stoic knowledge (ἐπιστήµη) is an exactly defined epistemic state: it is a knowledge that can neither be lost nor changed.35 The same is true of the knowledge (scientia) that is virtue, according to Seneca. In spite of its length, the following passage is quoted in full because it also shows that the knowledge that constitutes virtue is not purely contemplative but has a practical impact: it is knowledge about values, a knowledge that transforms the mind and not only shapes but determines its volitions. Like Socrates 33

Chrysippus in Diog. Laert. 7.88. See also Menn 1995 and, more generally, Küppers 1996. See, e.g., Ar. Did. in Stob. 2, p. 59, lines 7–11 Wachsmuth. There, practical wisdom is defined as “knowledge of what to do and what not to do and what is neither [scil. something to do or not to do], or as knowledge of good things and bad things and of what is neither for an animal that is by its nature a political animal and rational.” The definition of courage is “knowledge of what is terrible and not terrible and what is neither.” 35 Ar. Did. in Stob. 2, p. 73, lines 19–23 Wachsmuth. 34

310

jula wildberger

and like his fellow Stoics, Seneca is clearly an intellectualist. For him this means that one must not only practice ethical behavior but seriously study philosophy in order to become a good person. Virtue is knowledge about other [things and people] and about oneself. One must learn something about virtue in order to learn virtue itself. An action will not be right (recta) unless volition has been set right. For action comes from volition (voluntas). Volition, in turn, will not be right unless the disposition of the mind (habitus animi) has been set right. For volition comes from that disposition. Further, the disposition of a mind will not be in the best possible state unless it has learned the laws of life as a whole and worked out in detail what is to be judged about each thing, unless it has verified [its understanding of] things. Peace of mind (tranquillitas) can only be achieved by those who have acquired immutable and certain judgment. (epist. 95.56)

The virtuous disposition from which virtuous action can arise is a mind that possesses knowledge. This knowledge is, by definition, immutable and true. Its content is variously indicated, either as knowledge of things (e.g., epist. 31.6) or of “things divine and human,” i.e., the subjects studied by philosophy (e.g., epist. 31.8, 74.29, 89.5), or as knowledge of what is good or bad: “By one thing only the mind reaches perfection: through unchangeable knowledge of good things and bad things” (epist. 88.28). 4. Forms of Agreement with Nature In descriptions of the good life Seneca sometimes presents wisdom as a necessary condition for happiness. At other places he directly identifies wisdom with a life in agreement with nature, e.g., in the following passage from De vita beata. With its reference to a universal law that is to be [scil. understood and] followed, it is reminiscent of Chrysippus’s description of the end (Diog. Laert. 7.88). In the meantime—this is a point on which all Stoics agree—I assent to the nature of all things; never to deviate from it and to shape oneself according to its law and example, this is wisdom. The happy life, then, is a life in accordance with one’s own nature […] (dial. 7 [= vit. beat.].3.3)

Like Chrysippus, Seneca presents the highest good in the quoted passage as a life in agreement with both one’s own nature and the nature of all things. It is striking how abruptly he moves from one to the other, as if the connection between the two were completely obvious.36 He evidently 36

This is true even if the particle ergo has only transitional force.

ethics iv: wisdom and virtue

311

expected his readers to supply an explanation such as the one given by Chrysippus, namely that our individual natures are parts of the nature of all things. In the following paragraphs (dial. 7 [= vit. beat.].3.3–15.4), in which Seneca proposes no fewer than eleven further definitions and descriptions of the happy life, agreement with the nature of all things is never mentioned again.37 To some extent, this can be explained by the fact that the passage is a sustained argument against the Epicurean position according to which virtue is only a means to achieve pleasure, pleasure itself being the real end. Seneca wishes to beat the Epicureans at their own game, arguing, as far as possible, on the basis of concepts that are acceptable to both Stoics and Epicureans. Living in agreement with one’s own nature could be understood by an Epicurean as a way of living a pleasurable life. But Epicurean theology does not admit the idea that there is a universal nature that one should agree with. However, at the end of the passage, universal agreement returns in a form that is frequent in Seneca but, as far as I know, not attested in quite this manner for any Stoic before him. The happy person must follow God as a soldier obeys his general, and in particular those commands that bid him suffer physical pain, loss, and death.38 He must “obey the orders of God (deo parere) and cheerfully take upon himself whatever happens to him, never complain about fate, always interpret his own misfortunes in a charitable way” (dial. 7 [= vit. beat.].15.4). Agreement with the nature of all things often has this subjective ring in Seneca. The relationship between God and man is personalized. “Living according to the lead of nature” becomes “following God” (deum sequi ; e.g., epist. 96.2) and “obeying God” (e.g., also epist. 71.16). The sage follows God like a person.39 God is a general who gives orders (epist. 107.9f.), a father, and a friend (dial. 1 [= prov.].1.5 f.). Human life is seen from within, from the perspective of the sufferer who is hurt or tempted and bravely bears his lot as something assigned to him. Never did that fully developed man who had achieved virtue say anything bad about Fortune, never did he react to an incident with despondency; believing that he was a citizen of the whole world, and its soldier, too, he submitted to all trials as if they were orders given to him. Whatever happened, he did not 37

See also de Pietro (forthcoming). Compare, e.g., epist. 71.16, 76.23; further references Wildberger 2006: 875 n. 1332, 839 n. 1228. Similar imagery is to be found in Epictetus (Long 2002: 168f.). 39 Sen. epist. 107.11, Seneca’s famous adaptation of Cleanthes’s prayer to Zeus and Fate (Epict. ench. 53.1 = SVF 1.527). Only Seneca calls God parens. See also, e.g., epist. 104.23, where the human mind appears like a child trying to keep pace with the big person it follows. 38

312

jula wildberger turn away from it as if it were something bad and brought upon him by chance but accepted it as a personal assignment. “Whatever this is like,” he said, “it’s mine: it’s rough, it’s hard, so let’s bend to the task!” (epist. 120.12)

More rarely, Seneca assumes a cosmic perspective, letting the virtuous mind soar high above human affairs in a vision of how insignificant all this is. The human mind thus elevated realizes that the realm of all nature is its true abode, that it came into being to acquire knowledge about God and, thus, its own greatness as well (Sen. nat. 1 pr. 6–17).40 It is interesting to compare such passages with the beginning of Lucretius’s De rerum natura, Book 2, where an Epicurean looks down on the struggles and unnatural desires of his contemporaries.41 The Epicurean feels pleasure because he has arrived at a true understanding of human affairs, or rather his own affairs, and knows that he need not suffer either pain or fear; he himself is in a godlike state (Lucr. 2.8), and knowledge about the natural world outside him is just a means to achieve this aim of perpetual bliss (Lucr. 2.61). In Seneca, too, the study of nature helps to develop the greatness of mind that is necessary to withstand the lures and assaults of Fortune.42 Nevertheless, this study is also an aim in itself, something to be sought for its own sake. Wisdom is more than understanding your own nature and how its needs can be satisfied. For Seneca, as well as for other Stoics, wisdom is knowledge of both divine and human affairs.43 5. The Integration of Epicurean Features All the same, Seneca does integrate Epicurean features into his portrait of the happy person. And this, too, may be one of his original contributions to the Stoic discourse. 44 However, Epicurean spoils are always either perfectly compatible with Stoicism or modified by Seneca in such a way as to suit the Stoic system. One of them is the idea of a limit, a finis: the highest good is conceived of as an end where all wishing comes to a halt because a person

See also, e.g., epist. 65.15–21, 102.21 f., Williams 2012 and Gauly, infra, pp. 363–378. Lucretian influences on Seneca are discussed, e.g., in Guillemin 1952, La Penna 1994, Ronnick 1995, and Althoff 2005. 42 E.g., Sen. epist. 117.19, 58.26–29, Hadot 1969: 251–254, Gigon 1991. 43 Sen. epist. 31.8 (quoted supra), epist. 89.5; compare, e.g., Aëtius 1 pr. 2. 44 Seneca himself highlights how unusual it is for a Stoic to turn to Epicureans for instruction: epist. 2.5, 12.11. On Epicurus in Seneca, see Hermes 1951, Setaioli 1988, and, e.g., Mutschmann 1915, Schottlaender 1955, Maso 1979–1980, Setaioli 1997b, Grilli 1998, Graver (forthcoming) and Wildberger (forthcoming). 41

ethics iv: wisdom and virtue

313

has achieved all there is to achieve (epist. 15.11): “Set yourself an end/limit (finem) that you cannot surpass, not even if you wanted to.” Yet, whereas the Epicurean limits his desires by reducing them to what is necessary and natural, Seneca limits human desires by redirecting them toward what is truly good and what can be achieved by the mind on its own. Like Epicurus, Seneca can praise the wealth that is gained by limiting oneself to the fulfillment of a few elementary natural desires (epist. 4.10f.) and contrast such natural desires with the insatiable desires arising from wrong opinions (epist. 16.9): “Natural desires are limited (finita). Those arising from a wrong opinion have nowhere to stop. Nothing marks the boundary of what is wrong.” Nevertheless, Seneca retains the crucial difference between Stoicism and Epicureanism: the Epicurean limit is reached quantitatively with a surplus of pleasure or the complete removal of pain; the Stoic limit can only be achieved by realizing that a human being needs a different quality of good. It is not a sufficient quantity of pleasure or other indifferents that makes a Stoic happy. Like an Epicurean, a Stoic may—as a first step— reduce his desires to a natural minimum that can easily be satisfied under most circumstances (e.g., Sen. epist. 15.11, 20.7), but this is not enough. The recommendation to “reduce yourself to small things from which you cannot fall” (Sen. epist. 20.7) is coupled with the urgent advice to give upall former wishes and hopes, even those for small things (epist. 45.10–12, 110.18–20), and strive for one thing only: “to be content with oneself and the goods that grow within oneself.”45 What the Stoic is striving for is to understand his own nature, i.e., the thing he is made to be: a man with perfect reason. He needs this self-knowledge in order to achieve the only thing that can make him happy: to be what he is by nature. This is why the Stoic sage, who lives in agreement with his nature, needs nothing apart from himself. He does not want anything other than to be what he is, and he is content with himself because he has realized that he himself, i.e., his perfect, virtuous mind, is the best thing he can ever have. This is the condition that Seneca wishes Lucilius to achieve. Undisturbed by erroneous desires and fears of the wrong things he shall enjoy what is really good for him: himself. I wish you that you might become available to yourself, that your mind, which is now agitated by wandering thoughts, might finally come to a halt and assume a stable position, that it begins to like itself and that, after understanding what 45 Sen. epist. 20.7; compare epist. 9 and Evenepoel 2007. Seneca’s awareness of the theoretical differences becomes evident also in Epistle 118, where he offers a discussion of quantitative change that becomes qualitative; see Schmidt 1960.

314

jula wildberger the true goods are (which are obtained as soon as they are understood), it may no longer need an addition to its lifetime. (epist. 32.5).

Another adaptation of Epicurean ideas is the emphasis on the serenity and tranquility of the perfect mind. This is reminiscent of the smooth, level, and pleasurable movements in the soul of the Epicurean sage who is in a state of inperturbedness (ἀταραξία).46 A related phenomenon is Seneca’s stressing the joy (gaudium) of the sage.47 Yet again, he takes care to point out that this joy is the good feeling (εὐπάθεια) that the Stoic sage has when he perceives the presence of a good and that is in stark contrast to the worthless elation of a fool (epist. 59.1 f.). Finally, as Seneca takes equal care to stress, even this Stoic joy is not the end humans should be aiming for; it is only something that follows as a welcome consequence once the end has been reached (epist. 59.16, dial. 7 [= vit. beat.].9.2). One other feature of the Senecan sage might owe something to Epicureanism: the Stoic sage is a social animal, and he would not be perfect without social virtues like justice and the knowledge that he is part of a wider community. In Seneca, however, this Stoic sage often shows features of a more private kindliness.48 The natural environment of the Senecan sage seems to be the world of personal friendship rather than the political state. Given the importance of friendship within the Epicurean community, it is significant that when Seneca first introduces his own Stoic sage in Epistle 9, it is to show that his sage is a much better friend than any Epicurean ever could be. 6. Indifferents and the End: According to Nature vs. in Agreement with Nature Most frequently, however, the sage is presented by Seneca as a complete world in and of himself, like an impervious sphere. All that he does not wish to admit bounces off him without lasting effect.49 This state of selfcontained success and invincibility is also described as liberation from the slavery suffered by those who are obsessed with their own body and the external things that Stoics regarded as indifferents with only relative value or

46 For example, in Epistle 95.46 (quoted supra); see also Epistle 92.3. 6 (on the related concept of ἀοχλησία = “undisturbedness”), Hadot 1969: 126–141, and on tranquillity in general Striker 1996: 183–195. 47 Sen. epist. 4.1 f., 23.2 f., 27.3, 59.1 f., 59.14–16. 48 E.g., dial. 7 (= vit. beat.).4.2, epist. 115.13f. See also Ganss 1952: 62–83 and Gallina 1997–1998. 49 Epist. 9.2, 53.12, dial. 2 (= const.).3.5, Wildberger 2006: 934f. n. 1533.

ethics iv: wisdom and virtue

315

non-value.50 What such a person has learned is the fundamental difference between “according to nature” (κατὰ φύσιν) and “in agreement with nature” (ὁµολογουµένως τῇ φύσει).51 Humans should select preferred indifferents not just because it is natural, i.e., according to nature, for humans to select them, but because selecting what is natural is in agreement with nature. The sage does not select preferred indifferents for their own sake, because they would be valuable for him by themselves. He selects them because a consistent selection of what is according to nature will allow him to lead a consistent life, in agreement with nature, and thus a good life (e.g., epist. 92.11–13). This point is highlighted in the definitions that some of Chrysippus’s successors in the 2nd and 1st centuries bc had proposed of the Stoic end. Diogenes [of Babylon]: “reasoning well in the selection (ἐκλογή) and disselection of things in accordance with nature;” Archedemos: “to live [perfectly] completing all proper functions (καθήκοντα);” Antipater: “to live continuously selecting things in accordance with nature and disselecting things contrary to nature.” He also frequently rendered it thus: “to do everything in one’s power continuously and undeviatingly with a view to obtaining the predominating things which accord with nature.”52

The things in accordance with nature in these definitions are indifferents, not goods. For the Stoics quoted here, the highest good, and thus virtue, consists in the perfect and incessant activity suitable for obtaining indifferents with some positive value and avoiding indifferents with some disvalue. These definitions were developed in a debate with Academic critics who pointed out a problem. Stoic virtue is in danger either of becoming an empty term if it disregards indifferents as things that do not contribute to the acquisition of the highest good or of ending up becoming a position similar to that of the Peripatetics, according to whom there is a hierarchy of goods, with virtue standing at the top, but all the goods contributing to happiness.53 Seneca is aware of these theoretical problems. With Aristo’s criticism of practical advice he also rejects Aristo’s unorthodox claims that happiness can be achieved through indifference toward indifferents alone, i.e., by 50

For example, epist. 51.8 f. There is some overlap between the terms; see Wyszomirski 1993: 48, Wildberger 2006: 873 n. 1326. 52 Ar. Did. in Stob. 2, p. 76, lines 9–15 Wachsmuth, trans. (partly) Long and Sedley 1987 (58K). 53 See, in particular, Striker 1991, Barney 2003, Bénatouïl 2006, and the sources collected in Long and Sedley 1987, section 64. 51

316

jula wildberger

simply understanding what is good, what is bad, and what is indifferent and consistently choosing the good things (epist. 94). As Seneca expresses it in a recurring phrase, indifferents are materia virtutis, both the stuff from which virtue is developed and thus made (see below) and the matter on which virtue expresses itself and performs its art of living.54 All the same, attributing relative value to indifferents and finding a content for virtue is not one of Seneca’s central philosphical concerns. Texts in which Seneca demonstrates that the Stoic sage can or should lead a normal material life are rare in his œuvre (epist. 5, De vita beata). More frequently, he advocates a moderately ascetic lifestyle and encourages his addressees and readers to disregard what is not a good. The Stoics quoted in the definitions above use the technical terms “selection” and “disselection” for the proper evaluation and choice of indifferents. The terms contain a reference to a selection process because, when dealing with indifferents, we usually have different options to choose from. In order to effect the selection, it is therefore necessary to assess the relative value of each available indifferent under the given circumstances. It is symptomatic, then, of Seneca’s assessment of the real problems his philosophical writings should deal with that the verb “to despise” (contemnere) acquires the almost technical meaning of “evaluating indifferents correctly as somthing with relative and thus irrelevant value (in comparison to a really good or bad thing).” 55 The crucial point for Seneca is that one must stop attributing too much importance, i.e., too much positive or negative value, to indifferents because that is what most people get wrong. 7. Virtue and Roman Manliness It has, of course, been noticed that this ascetic focus brings Seneca close to contemporary Cynicism, whose proponents, on their part, often seem to have shown considerable sympathy for Stoic ideas.56 All the same, Seneca’s ascetism has also a distinctively Roman quality. In a striking passage (epist. 66.49–53), Seneca first acknowledges that, with regard to what is really good,

54 Epist. 66.15, 71.21, 85.39, dial. 7 (= vit. beat.).21.4, 7.22.1, 12 (= cons. Helv.).6.2, benef. 4.19.4 (a different sense in 90.46). The expression goes back to Chrysippus in Plut. mor. (On Common Notions against the Stoics) 1069e = SVF 3.491: ὕλην τῆς ἀρετῆς. In Epistle 92.5, Seneca may be referring to Antipater’s position in the debate about the value of indifferents; he says that Antipater accords too much importance to them. 55 E.g., epist. 8.5, 23.7, 32.4, 39.4, 56.11, 58.28, 62.3, 65.22, 66.1, 71.19, 71.37, 73.14, 74.13, dial. 1 (prov.).6.1, dial. 7 (= vit. beat.).4.3, 7.20.3, 7.21.1 f. 56 On the Cynic Demetrius, in particular, see Billerbeck 1979.

ethics iv: wisdom and virtue

317

it makes no difference whether virtue expresses itself in a pleasurable or a difficult matter, but then states that he personally would tend to prefer “those [goods] that seem harsher to those that are soft and effeminate.”57 This is, first of all, an argumentum a maiore ad minus. If there were any difference at all, adverse goods would even be better. Therefore, they must at least be equal, as is required by the thesis that all goods are equal in whatever matter they are expressed. It is also an argument that makes hortatory use of values we particularly associate with such Roman conceptions of virtue as fortitude and military valor: Seneca quotes the most Roman forms of praise, the archaic formula “macte virtute esto” (epist. 66.40) and contrasts the heroism of a Mucius Scaevola (epist. 66.51–53) with the enjoyment of certainly delightful but not very manly physical pleasures. Seneca’s praise of the Stoic vir invictus, the invincible man who stands upright amid ruins, who equally despises pain and gold, draws on the Roman discourse of manliness. Like Cicero, Seneca uses the Roman term vir bonus to refer to the sage. Like Cicero, he represents and discusses virtuous behavior through traditional Roman exempla. Seneca even goes beyond what we find in his predecessor when, in Epistle 120, he develops a theory of how humans formed a concept of good by observing such exemplary figures and paints the picture of a sage that shows the very Roman traits of civic responsiblity, fortitude, readiness to suffer hardships, and discipline (120.10–13). This is particularly remarkable because the epistemological ideas developed in this letter are not attested in any other Stoic source and may very well have originated with Seneca himself.58 Other passages, too, bear witness to the same phenomenon that, in Seneca, Stoicism and typically Roman manliness blend harmoniously: among philosophers, only the Stoics offer a philosophy for real men (dial. 2 [= const.].1.1); the Roman philosopher Quintus Sextius, who did not want to adhere to any Greek school, is characterized as a Stoic in essence, if not by profession (epist. 64.3), and as someone who practices philosophy in Greek but with the character and mindset of a true Roman.59 Roman color can easily be applied by replacing, e.g. athletic imagery with imagery drawn from 57

Epist. 66.49, trans. Inwood 2007a. Seneca is drawing on the Stoic theory of concept formation (Inwood 2007a: 324, note on epist. 120.3, Inwood 2005b). But neither the application of that theory to the concept of good nor the concept formation by simultaneous use of combination and idealization are attested elsewhere. On the contrary, the parallel account we find in Cic. fin. 3.20–22 (see below) presupposes that the notion of “good” derives from observation and conceptualization of behavior as it actually occurs. 59 Epist. 59.7: virum acrem, Graecis verbis, Romanis moribus philosophantem. 58

318

jula wildberger

the military life or gladiatorial combat.60 At other points, Seneca expresses his perception of a difference by contrasting ineffective quibbling with the stern language of command characteristic of a society where authority and example promote uncontested values, where speech does not argue a point but helps to muster the strength to actually do what the audience already knows to be right and necessary (epist. 82.20–24). Even here, however, Seneca indicates that there is no fundamental clash but that the Greek philosophers should learn from the Romans. One of the adduced examples is the Spartan Leonidas with his men (21), and the whole passage is introduced as a suggestion of how Seneca, as a Stoic, would prefer to present Zeno’s syllogisms more effectively in order to create real, heartfelt belief in their truth (19f.). 8. Kalon vs. honestum One of the most fundamental changes Stoicism underwent when being translated into Latin was the Roman interpretation of the Greek Stoic term καλόν.61 As Seneca insists again and again, only what is καλόν is good: unum bonum quod honestum. In translating καλόν as honestum, Seneca follows the lead of Cicero and others, thus helping to cement a linguistic usage that is responsible for a frequent misunderstanding—a misunderstanding that surely was intended by some Roman interpreters of Stoicism who used it to support their own value systems.62 It arises from the fact that the Latin word “honestum” no longer retains any trace of the meaning “beautiful” that is central to the Greek word “καλόν,” in general, and to καλόν as a Stoic concept, in particular. The relation between good and καλόν is outlined in Diog. Laert. 7.101, where καλόν is defined as “virtue and what partakes in virtue.”63 From such definitions, together with the interpretation as honestum, i.e., honorable, the word is often understood to refer to morally praiseworthy, honorable acts, such as saving one’s fatherland, protecting the innocent, or pronouncing a just judgment. Often, these honorable acts are conceived in antithesis to

See, e.g., Lavery 1980, Sommer 2001, Kroppen 2007. See, e.g., Fischer 1914: 11, Eisenhut 1973: 66, Powell 1995: 299. 62 For Cicero, see, e.g., Dyck 1996: 31, Atkins 2000: 513. 63 Compare in Seneca, e.g., epist. 67.10, 71.5, 76.11. This completed reason is called “virtue” and is, at the same time, what is honorable (honestum); 76.19, 82.12: “[…] whatever of these things has been approached and handled by virtue, is turned into something honorable and glorious;” 82.14, 85.17. 61

ethics iv: wisdom and virtue

319

self-interested behavior or behavior that is in some way ignoble, e.g., because it is regarded as unlawful or as a base expression of animal instincts. Now, for a Stoic, all these acts by themselves are neither καλόν nor its opposite, αἰσχρόν, but only things whose selection or disselection may be realizing a proper function of a human being. Proper functions are a class of predicates, namely those which it is in accordance with human nature to effect. The sage will therefore perform such proper functions, but only under suitable circumstances. The objects of proper functions are only to be selected in preference to other indifferents. A real Stoic honestum or καλόν, however, is to be sought unconditionally, absolutely, under all circumstances. And the thing that is a real honestum or καλόν can only be something that involves real Stoic virtue: either a perfect mind itself or something that partakes in some way of such a perfect mind, e.g., by being one of its actions. One might, of course, wonder why something that ordinary people would regard as an honorable act is not a good thing for a Stoic. What is it that virtue adds to the act itself to turn it into something so beautiful and honorable (καλόν) that it becomes a good? A single act, may it be ever so praiseworthy, cannot be καλόν in itself. It stands on its own and is only a fragment. To be meaningful and the kind of good a human being can achieve, it must be part of a good life. It must be set in a context of a consistent harmonious whole. For a Stoic, this whole can only be virtue. Whatever we do without virtue lacks agreement and consistency because virtue itself is consistency (Diog. Laert. 7.89, Sen. epist. 74.30). This is why Seneca so often criticizes inconsistency and discontinuity, the lack of a coherent context for one’s actions, as the hallmark of the fool.64 The Stoics themselves inherited the term καλόν from archaic and classical Greek ethical discourse and seem to have invested it with various meanings. A series of explanations are given for why virtue or the good are said to be καλόν: because virtue naturally calls (καλεῖν) those who seek her (Ar. Did. in Stob. 2, p. 100, lines 21f. Wachsmuth), because it has all the numbers that nature is looking for, or because it is perfectly symmetrical (Diog. Laert. 7.100). All these explanations presuppose that, for the Stoics, the word still retains at least some components of the meaning “beautiful”: the καλόν is attractive; it is in some way completely worked out; it is symmetrical. In a similar manner, one of the three definitions of the word refers to καλόν as something that functions as an ornament,65 this last sense being the one that is applied to the sage. 64 65

For example, epist. 20.1–6, 23.8, 45.6, 52.1 f. Diog. Laert. 7.100: τὸ ἐπικοσµοῦν.

320

jula wildberger

All the same, these sources do not yet tell us all: we still do not know what it means to a Stoic that something is beautiful. Unfortunately, no extant doxographic source gives us the complete Stoic definition of beauty (κάλλος). All the same, it can be reconstructed from a passage in Plotinus’sOn Beauty that is directed against the Stoic concept as an influential position contrary to Plotinus’s own. Whereas for Plotinus beauty derives from the One, for the Stoics beauty presupposes a plurality of things. It is “symmetry of the parts with each other and with the whole.” 66 Plotinus attributes this definition to “almost everyone.” All the same, it can be shown both that this was at least accepted by the Stoics and that they used it in ethical discussions. Chrysippus, for example, is reported to have defined the beauty of the soul as a symmetry of its parts, a definition that recurs in Arius Didymus’s Handbook of Stoic Ethics.67 Accordingly, we can reconstruct68 that a life is beautiful for a Stoic when it is a coherent whole (ὅλον) with parts, e.g., its single acts, and when all these parts are symmetrical and in harmony both with the whole and with each other. Once this is the case, every proper function a person effects is turned into a “perfect achievement” (κατόρθωµα), while the mind of that agent is now a virtue. From this, one can conclude that, for a Stoic, something is καλόν in the ethical sense if it is a constituent of the beauty of the agent who has it. As such it must necessarily be a part of a whole, which explains why there cannot be isolated occurrences of it and why a single occurrence of καλόν is always part of a coherent whole, a complete good life. This specific nature of the καλόν and thus of what is good is also taken account of in two descriptions of the process by which a human being grasps the concept of the good. The one is to be found in the Stoic section in Cicero’s De finibus (3.20–22). It explains how a concept of what is good is developed from self-observation. The young man observes how he is consistently acting according to nature and thus begins to understand what is good for him. For a human being’s first affiliation (conciliatio)69 is toward those things that are in accordance with nature. But as soon as he has acquired an understanding, or rather the conception, which the Stoics call ennoia, and has seen the order 66

Plot. 1.6.1: συµµετρία τῶν µερῶν πρὸς ἄλληλα καὶ πρὸς τὸ ὅλον. Galen, De placitis Hippocratis et Platonis 5.2.47; compare also 31–33, Ar. Did. in Stob. 2, p. 63, lines 1–5 Wachsmuth; Cic. Tusc. 4.31. 68 It must be pointed out that what follows is my reconstruction, which I believe to be consistent with most current readings of Stoic ethics. There are, however, scholars who offer a different account, e.g., Brennan 2005. 69 This is Cicero’s term for οἰκείωσις. 67

ethics iv: wisdom and virtue

321

and, so to speak, harmony of his actions, he comes to value this far higher than all those objects of his initial affection; and he draws the rational conclusion that this constitutes the highest human good, which is worthy of praise and desirable for its own sake. Since that good is situated in what the Stoics call homologia (“agreement” will be our term for this […]) […] that good […] is the only thing desirable through its intrinsic nature and value, whereas none of the first objects of nature is desirable for its own sake. (Cic. fin. 3.21)

The other account is one we find in Seneca, in Epistle 120, which was already mentioned above. It complements the account we find in Cicero in that it explains how a concept of the good can be developed by observing not oneself but other human beings. Here, humans understand what the good is by contracting single observations to a vision of the exemplary sage. This vision comes into being precisely at that moment when single traits are combined in such a way that an image of coherent, invariable behavior is formed. Moreover, he was always the same and consistent with himself in every act; no longer good just by design, but so thoroughly habituated that he not only could act rightly but could not act other than rightly. We understood that in him virtue was complete. […] On the basis of what, then, did we come to understand virtue? It was shown to us by this man’s orderliness, fitting beauty (decor) and consistency, the mutual agreement of all his actions […].70

It is a characteristic of Seneca’s descriptions of what is good that he prefers wealth of detail to focus and clarity. It is also characteristic that the sage we envision at this moment is a Roman hero, someone who combines all the virtues represented by the great traditional exempla, who lives a Roman upper-class life and understands his role in the cosmos as a militia with God as his commander. The beauty and agreement of this Roman sage’s soul blend indistinguishably with the good old virtue of stubborn constantia, the steadfastness a vir Romanus shows when intent on his purpose. All the same, we can see that the foremost and basic, or rather essential, feature of the good is precisely that Stoic agreement (ὁµολόγια) which comes about when everything fits together, when the parts of a life, its single actions, form a coherent whole with which they harmonize as beautifully as they harmonize with each other. Although Seneca uses the traditional Roman term honestum, he makes sure that his reader does not forget how beautiful that honestum is—but also that he sees at the same time how manly Stoic beauty can be.

Sen. epist. 120.10 f., trans. Inwood 2007a, with alterations.

322

jula wildberger Nothing is better than virtue, nothing more beautiful. No, I cannot see anything more beautiful for Iuppiter to find on earth […] than to watch Cato […].71

In sum: Seneca’s writings demonstrate familiarity with and adherence to essential Stoic tenets concerning virtue, wisdom, and the good life. They also show signs of independent innovation, such as assigning a much more important role to intellectual progress and the theoretical knowledge that comes with it; personalizing the idea of “following nature”; the incorporation of Epicurean ideas about limits, tranquility, and joy; the hortatory use of a Cynic term “contempt” instead of the Stoic terms “selection” and “disselection”; and applying the Roman method of ethical argument by examples to design a new explanation of how a concept of good is formed. Most importantly, however, Seneca integrates all these elements into a model that builds on and incorporates an already existing tradition of Roman philosophy and is suitable for the reality in which he and his peers were living, a model that merits being called both “Roman” and “Stoic.”

71 Sen. epist. 67.16 and dial. 1 (= prov.).2.9. Compare, e.g., also dial. 3 (= de ira 1).12.5, 7 (= vit. beat.).15.2, epist. 66.6, 66.21.

ETHICS V: DEATH AND TIME

Catharine Edwards The fear of death, for Seneca, casts over human life a terrible shadow. To accept death is the hardest lesson but the most important one for those seeking the happiness of philosophical calm. This can only be done, however, if we learn to manage correctly our understanding and use of time. The relationship between attitudes to death and to time often comes to the fore in Seneca’s writing and is the prime concern of his treatise on the shortness of life, De brevitate vitae, composed probably 49–50ad.1 People complain that life is too short, observes Seneca, but any life is long enough if used properly. Much of the treatise is concerned with the carelessness with which people give away their time; people live their lives, he claims, as if they were never going to die: tamquam semper victuri vivitis (3.4), a reproach cast vividly in the second person. The treatise concludes with exhorting Seneca’s addressee Paulinus to abandon his public career at once and devote himself to philosophical leisure. Those who fill their days to an advanced age even with the law courts, the Forum, and the responsibilities of public office (to say nothing of the pursuit of pleasure) do not really experience life (20.5): No one keeps death in view, no one restrains his hopes. Some indeed make plans for those things that lie beyond life—great hulking tombs and dedications of public works and offerings for funeral pyres and ostentatious funerals. Yet, in truth, the funerals of such men should be carried out by the light of torches and candles, as though they had lived but the shortest time.2

The signs that mark the death of a publicly distinguished man at an advanced age are juxtaposed with those of the death of a little child; as Seneca has repeatedly asserted earlier in the treatise, even a very old man’s death, when he has not spent his life wisely, feels premature (cf. 3.3, 7.10, epist. 77.20).

1 On the background to this treatise, see most recently Williams 2003: 19f., as well as Traina 1984: xv. 2 Nemo in conspicuo mortem habet, nemo non procul spes intendit; quidam vero disponunt etiam illa, quae ultra vitam sunt, magnas moles sepulcrorum et operum publicorum dedicationes et ad rogum munera et ambitiosas exequias. at me hercules istorum funera, tamquam minimum vixerint, ad faces et cereos ducenda sunt.

324

catharine edwards

Only the wise man, one who is conscious that he has used his time well, can approach death with a steady step, certo gradu (dial. 10 [= brev.].11.2). The preoccupations of De brevitate vitae surface repeatedly in the Epistulae morales, the collection of letters written in the early 60s ad, which turned out to be Seneca’s final work, composed in the ominous shadow of Nero’s displeasure—written, it might seem, in borrowed time.3 Time, death, and the relationship between them are concerns with which Seneca opens the first of his Epistulae morales. He exhorts his addressee Lucilius: tempus […] collige et serva, “gather and save your time” (epist. 1.1).4 Epistle 12, which concludes the first book in the collection, considers at length how time should be conceptualized and presents the contemplation of death as playing a crucial role.5 Time, for Seneca, figures among the key concerns of philosophy (epist. 88.33). But it is a relatively abstract concept, which can only be fully grasped by those whose philosophical progress is quite advanced (epist. 90.29). Earlier Stoics seem to have been notably preoccupied with time in the context of physics.6 Chrysippus is said to have argued that no time is present as a whole or exactly.7 When he chooses, Seneca is quite capable of engaging with the philosophical technicalities of time. Epistle 49 opens with a poignant account of how a visit to familiar places in Campania has made Seneca feel much more acutely the absence of his friend Lucilius. This emotive opening is a prelude to a discussion of time that touches suggestively on the more technical aspects. Punctum est quod vivimus et adhuc puncto minus. sed et hoc minimum specie quadam longioris spatii natura derisit […], “The time we spend living is a moment, even less than a moment. But this briefest time nature has mocked by making it appear of greater duration” (49.3). Seneca goes on to argue, however, that it is precisely the brevity of life that makes it foolish to waste time on technicalities of dialectic. Mors me sequitur, fugit vita; adversus haec me doce aliquid, “Death is at my heels, life runs away; teach me something that will help me confront this” (49.9). The technical conceptualization of time is useful insofar as it underpins Seneca’s insistence on the urgency of his philosophical project.

3 On the chronology of Seneca’s works and the circumstances under which they were written, see Griffin 1992. On Seneca’s treatment of time in his works generally, see Grimal 1968, Armisen-Marchetti 1986, Gagliardi 1998. 4 On this letter see Gagliardi 1998: ch. 3, Richardson-Hay 2006 ad loc. 5 These concerns underlie all the letters but manifest themselves notably in Letters 4, 12, 23, 24, 26, 49, 61, 69, 70, 71, 77. Death has been seen as a particular theme of the third book of letters (22–29). 6 On the complexities of this see Goldschmidt 1979, Brunschwig 2003. 7 SVF II 509. Helpfully discussed by Schofield 1988.

ethics v: death and time

325

According to the traditional Stoic scheme, time is one of four incorporeal things (along with the sayable, void, and place).8 Seneca acknowledges time’s incorporeality (dial. 10 [= brev.].8.1; epist. 58.22) but does not insist on it. Armisen-Marchetti (1995b: 548) argues, in her perceptive discussion, that Seneca’s prime concern is with human time, lived time, rather than cosmic time.9 Spatial imagery has an important role to play in Seneca’s conceptualization of time. Linear images tend to feature in his discussions of the individual human life, often conceived as a cursus with a fixed end-point, while cosmic time is usually conceived of in terms of cycles, on the model of the cyclical motion of the planets (e.g., dial. 10 [= brev.].8.5; epist. 107.8f.).10 Lived time is sometimes measured in terms of change and decay. Epistle 12, for instance, dwells vividly on a house falling into disrepair, some overgrown trees and finally the human body, whose perceptible signs of aging constitute insistent reminders of time’s irrevocable passage.11 But an important part of Seneca’s approach to the correct conceptualization of time is to encourage a shift in how human time is to be understood from the linear to the circular, from the existential to something approaching the cosmic. A cosmic model of time is brought into play in De brevitate vitae, where Seneca declares that the passage of present time can no more suffer delay “than the universe or the stars, whose perpetual unceasing motion never lets them rest in the same position” (10.6).12 The relationship between human and cosmic time lies at the heart of Epistle 12, where Seneca observes: Tota aetas partibus constat et orbes habet circumductos maiores minoribus, “Our space of life is divided into parts; it consists of large circles enclosing smaller” (epist. 12.6). He repeats Heraclitus’s opaque observation:parem esse unum diem omnibus similitudine; nihil enim habet longissime temporis spatium, quod non et in uno die invenias, “One day is equal to all days through resemblance, because the very longest space of time possesses no element that cannot be found in a single day” (12.7).13 Two possible interpretations of this are offered. First, each day is the same length, and made up of the same

Cf. SVF II 331, 521, 1142. Cf. Traina 1984: x–xi. Seneca like Marcus Aurelius later is, in Rist’s (1972: 287) terms, less interested in time “viewed primarily as a problem in physics” but rather concerned with time as “a moral problem”. 10 Armisen-Marchetti 1995b: 550–552. Unusually in epist. 36.10 life itself is seen as cyclical— the time will return when we shall be restored to the light of day. 11 Edwards 2005a. On Letter 12 see also Henderson 2004: 19–27. 12 Nec magis moram patitur quam mundus aut sidera, quorum inrequieta semper agitatio numquam in eodem vestigio manet. 13 On the circles, see Habinek 1982. This issue is also explored in Ker 2009a. 9

326

catharine edwards

divisions of time as any other. Second, each day has the same shape as any other, as light succeeds darkness, to be succeeded again by darkness. It is on the latter basis that one might understand a kind of equivalence between a day and a lifetime, as Habinek suggests. Angustissimum habet dies gyrum, sed et hic ab initio ad exitum venit, ab ortu ad occasum, “The day is the smallest circle, but this too has its beginning and its end, its sunrise and sunset” (12.6). Another implication of this resemblance seems to be that because each day of one’s life is like the last day, it should not be too onerous to treat it as if it were the last day: sic ordinandus est dies omnis, tamquam cogat agmen et consummet atque expleat vitam, “every day should be regulated as if it concluded the series, as if it consummated and filled out our life” (12.8). An individual life seen as a circle may be experienced as complete, perfect, whenever it comes to an end. Earlier Stoics, as we have seen, had debated at length how to define the present. Seneca seems inclined to treat the individual day as the most productive way of conceptualizing present time: singuli tantum dies […] praesentes sunt, “Only one day at a time can be experienced in the present” (dial. 10 [= brev.].10.4). We might see this, he asserts, as a philosophical variant on the poet’s motif of carpe diem; we should not focus on preparing for the future but live today rightly (dial. 10 [= brev.].9.3).14 The individual day is the focus of scrutiny according to the technique of self-examination Seneca repeatedly advocates, attributing it in De ira to the philosopher Sextius (dial. 5.36.1–3): Sextius used to do this, and when the day was over and he had retired to bed he would put these questions to his soul: “What faults of yours have you cured today? What vice have you resisted? In what way are you improved?” […] I have adopted this strategy and every day I plead my cause before myself as judge.15

For Seneca, the single day is the unit of time best adapted to a philosophical approach to life.16 While this emphasis is also to be found in other Stoic writers, for instance, Epictetus (diatr. 3.10.2),17 it is developed furthest in 14 On the contrast between Seneca and Horace’s treatments of carpe diem, see Williams 2003: 22. 15 Faciebat hoc Sextius, ut consummato die, cum se ad nocturnam quietem recepisset, interrogaret animum suum: “quod hodie malum tuum sanasti? Cui vitio obstitisti? Qua parte melior est?” […] utor hac potestate et cotidie apud me causam dico. 16 Cf. epist. 4.5, 16.1. This passage plays a key role in Foucault’s The care of the self (1986: 46, 61 f.). On the practice of daily self-scrutiny, see Hadot 1995, Edwards 1997 and (offering an illuminating account of the De ira passage) Ker 2009b. For Ker, the strategies of time-control advocated by Seneca are deeply implicated in the set of techniques by which the Roman aristocracy maintained its social power. 17 Epictetus refers to Pythagorean practice in this context and Pythagorean writings may also have influenced Seneca. Cf. Ker 2009b.

ethics v: death and time

327

Seneca. Indeed, as Foucault and others have observed, the practice might seem to underlie Seneca’s treatment of his daily experiences in the Epistulae morales (perhaps most explicitly in epist. 83). Ker (2009b: 185) suggests that one might detect in the De ira passage “a fusion of day and self as the object of scrutiny.” There is a kind of equivalence between control of time and control of the self set out even in the opening passage of the first letter in the collection, where Seneca urges Lucilius: vindica te tibi, et tempus […] collige et serva, “Lay claim to yourself and gather and save your time.”18 Seneca returns again and again to the excoriation of those who fail to value time correctly, who waste their own time. The denunciation of their failings is one of the principal themes of De brevitate vitae. They spend little of their lives in actually living (2.2). The letters, too, return repeatedly to the concern with time wasted. In Epistle 122, Seneca compares to the dead those who fritter away their time in the self-indulgent pursuit of pleasure (2 f.): Though they pass the night-time hours with wine and perfume, though they spend every minute of their unnatural waking hours in eating dinners—and those, too, cooked separately to make up many courses—they are not really banqueting, they are conducting their own funerals.19

The luxurious anticipate their own deaths, not only in the sense that they may be shortening their lives but also in their preoccupation with the meaningless experiences of the body rather than with what is truly good.20 The repetitive and unsatisfactory pleasures of the mortal flesh should be a matter of indifference to one who is properly focused on life’s only true goal, the pursuit of virtue. As Seneca asserts in Epistle 12, one who wastes his time is not truly alive; immo mortuus est, “indeed he is dead” (12.9). Past, Present, Future For Seneca, time is a supremely valuable possession, re omnium pretiosissima (dial. 10 [= brev.].8.1). Indeed, he sometimes characterizes it as the only thing that belongs to us: omnia, Lucili, aliena sunt, tempus tantum nostrum est, “No other things, Lucilius, belong to us; time alone is ours” (epist. 1.3). One must properly take account of one’s time: rationem facere (dial. 10 [= brev.].17.5; cf. 18

Cf. Grimal 1968. Licet in vino unguentoque tenebras suas exigent, licet epulis et quidem in multa fericula discoctis totum perversae vigiliae tempus educant, non convivantur, sed iusta sibi faciunt. 20 Cf. epist. 60.3f., 65.16. One might trace here the influence of Plato’s Phaedo (esp. 65 f.), where Socrates is made to argue that the body is a tomb and the philosopher only truly lives insofar as he frees himself from the body’s needs. See further Edwards 2007: 172–176. 19

328

catharine edwards

dial. 9 [= tranq.].3.8, epist. 1.4). A multitude of images drawn from commerce are used to emphasize time’s value. As Armisen-Marchetti stresses, these also serve to undermine the traditional Stoic characterization of time as incorporeal. The idea of time as a commodity makes it seem fixed and static— and is in considerable tension with Seneca’s stress on the fleeting nature of time (1995: 552 f.). Seneca offers a variety of techniques to enable the would-be philosopher to take possession of time (1.2, 101.8). The very process of writing letters, in itself (at least as practiced by Seneca) a form of self-scrutiny, could be seen as a means to this end.21 The focus here is primarily on present time. De brevitate vitae, in particular, develops at length important distinctions between past, present, and future time: in tria tempora vita dividitur: quod fuit, quod est, quod futurum est. It is present time, often, as we have seen, conceptualized in terms of the individual day, which we must value and exploit to the full.22 In contrast to the fleeting nature of the present, past time is certum, sure (dial. 10 [= brev.].10.2). It is an everlasting and untroubled possession (10.4). Fortune, which forever threatens the present and the future, has no dominion over the past. The past, therefore, has the capacity to be a source of certain happiness—at least for the would-be philosopher. Again, we must take possession of it. But the manner in which we effect this in relation to the past is different. We must allow ourselves (as those who are too busy, whether with work or pleasure, fail to do) the leisure to enjoy its recollection (dial. 10 [= brev.].10.4f., cf. epist. 83.2). Memory plays a key role here.23 But only those who have lived all their lives well are in a position to take pleasure from looking back (10.3f.). And as he comments in the Epistles, it is only contemplation of the past that enables us to formulate a productive plan for the future (epist. 83.2). Elsewhere, however, Seneca sometimes chooses rather to stress that time that is past no longer exists: usque ad hesternum, quicquid transit tempus, perit, “Even including yesterday, whatever time is passed is lost” (epist. 24.20). A later letter describes both past and future times as aliena (74.73).24 In

Sangalli 1988: 55. On the broader implications of his use of epistolary form, see Wilson

2001. 22 Marcus Aurelius lays a similar emphasis on the need to focus on the present. Indeed his To himself offers a similarly episodic model of self-scrutiny. This is suggestively discussed by Hadot 1998: 131–137. 23 The emphasis Seneca places on memory here is developed further in his discussions of the role of memory in overcoming the pain of bereavement in cons. Helv. and cons. Marc. (cf. Armisen-Marchetti 1995b: 554). 24 Sangalli 1988: 59 sees Seneca as influenced by Epicureanism here. Cf. Grimal 1968.

ethics v: death and time

329

reminding his reader how much time has already been wasted, Seneca seeks to underline the urgency of making good use of whatever time remains. It is the imminence of death that renders time so valuable, so precarious. It is by reminding ourselves of death’s imminence that we may be galvanized to make the best use of our time.25 The thought of death must be our constant companion. Time must be valued, but we can never depend on it—future time cannot be counted on. Hope poses a significant threat to the mental tranquility that should be the philosopher’s goal (dial. 9 [= tranq.].2.7–9). How long we live is not in our power, Seneca insistently reminds his reader.26 “The man who is spurred ahead by hope of anything […] is troubled and unsure of himself” (epist. 23.2). Hope is always accompanied by fear. Anxiety for the future creates intense wretchedness (epist. 98.6). And concern with the future serves as a dangerous distraction from the present, another cause of wasted time. One who thinks too much of the future spends his life getting ready to live rather than living (epist. 45.12f.). If we are to derive full value from the present, we must free ourselves from anxieties about the future. Above all, many people’s lives are blighted by the fear of death.27 This must be overcome if we are to enjoy life. A key strategy here is the Praemeditatio futurorum malorum (cf. Armisen-Marchetti 1986). Arguing that unexpected misfortunes are felt as more grievous blows than those for which one is prepared, Seneca advises his readers to make mental preparation for the possibility of poverty, of losing one’s loved ones, one’s home, but above all for death (epist. 30.18, 70.17 f.). One should make a habit of rehearsing these events in one’s imagination, so that one is never taken by surprise. The imagination of one’s own end, filled out in gruesome detail, is to be dwelt on and embraced.28 The most appalling of future events transposed by imagination into the present can thus be robbed of their power. Another way to conceptualize the experience of death is to think of it as a very gradual process, a process in which we are already far advanced. We die a little every day, Seneca advises his correspondent, in the first of the Epistulae morales: “What man can you show me who values his time, who takes account of the worth of each day, who understands that every day he is dying?” (1.2).29 Seneca at once reminds his readers that past time is lost time. 25

Marcus Aurelius offers similar comments, if not so insistently as Seneca (cf. M. Aur. 2.5.2). E.g., epist. 92.25, 93.4–7. 27 Here too Seneca has much in common with Lucretius’ version of Epicureanism, cf. Edwards 2007: ch. 3. 28 See Edwards 2007: 107. 29 Cf. epist. 24.20 f., 58.24. 26

330

catharine edwards

It is already in the possession of death. Yet by this means he also presents a picture of death itself as an already familiar part of our lives rather than the great unknown. Here, too, Seneca focuses on rethinking our disposition toward death by transposing it from the future to the present. Seneca insists repeatedly that the length of one’s life is not significant (epist. 77.20). Death, he claims, should not be seen as an intrinsically bad thing. Is there no case to be made, we might wonder, for regret at, for example, good deeds unfinished? A crucial consideration here is that for Stoics virtue does not need the dimension of time to be complete (epist. 78.27, 93.4). Behavior is judged on the basis of intention rather than result: consilium rerum omnium sapiens, non exitum spectat (epist. 14.16). The wise man lives fully in the present moment (cf. Armisen-Marchetti 1995b: 565). Accepting Death The wise man never does anything unwillingly; dying well is dying willingly, Seneca observes (epist. 61.2, 82.17f.). The philosopher, then, accepts death. His disposition toward death colors the whole of his existence. But it is most evident at the moment when he meets his own end. The question of how one should die has a particular prominence in the Epistles. It is here that we find articulated most explicitly a view (which can also be found in the writings of other authors of the Principate) that the moment of death, above all, expresses an individual’s true value. Mors de te pronuntiatura est, “death will pronounce judgment on you” (epist. 26.6).30 It is because dying is such a significant experience that one must prepare oneself with particular care to face death: egregia res est mortem condiscere “It is a great thing to learn thoroughly how to die” (26.6).31 This is what philosophy primarily offers (cf. dial. 10 [= brev.].15.1, epist. 4.6). Seneca’s use of this claim as a means of countering the fear of death might seem paradoxical. But his argument is that only one who has learned to overcome the fear of death can die well. Examples of courageous ends have a key role to play here. Seneca explores in detail instances of individuals who encounter death from disease with great bravery. His friend Bassus, for example, overwhelmed by the infirmities of old age, is praised at length for seeing death coming and welcoming it (epist. 30.9). The death of Socrates, condemned to drink hemlock in 30 31

On this as a general cultural preoccupation see Edwards 2007. There is perhaps an echo of Plat. Phaid. 64a.

ethics v: death and time

331

an Athenian prison recurs several times, as does that of Regulus.32 Most striking, perhaps, is Seneca’s repeated celebration of the suicide of Cato the Younger, notably in Epistle 24.33 By rehearsing in our minds the deeds of such individuals we can perhaps spur ourselves to equal their bravery when the time comes. A Time to Die Accepting death may sometimes, as in Cato’s case, entail choosing death. One might say that suicide can offer the most graphic evidence that one has overcome the fear of death. 34 Seneca’s frequent references to and examples of suicide are an aspect of his writing that has disturbed (and fascinated) many readers. They need to be seen as a key part of his project to overcome the fear of death (Griffin 1992: 384). The implication of numerous passages in the Epistles is that to take one’s own life at the moment one chooses may sometimes be a good death. Seneca concludes Epistle 69 with further observations on death: hoc meditare et exerce, ut mortem et excipias et, si ita res suadebit, accersas, “consider and practice this—how you may welcome death and, if circumstances recommend, invite it” (epist. 69.6). The following letter, Epistle 70, offers a lengthy and sustained exploration of the right time to die. Seneca, in Epistle 69, invokes Epicurus’s advice: meditare mortem, “think on death.” Yet the Epicureans apparently condemned suicide under almost all circumstances—despite their doctrine that “death is nothing to us” (Kyria doxa 2, cf. Warren 2001: 92). When he killed himself the philosopher Diodorus was criticized, according to Seneca, for not following the teachings of Epicurus (dial. 7 [= vit. beat.].19.1).35 By contrast, Stoicism in imperial Rome, at least in Seneca’s rendering of it, seems to endorse, even encourage, suicide under certain circumstances. Arthur Darby Nock famously referred to “the Stoic cult of suicide” (1933: 197). Seneca’s views on the appropriateness of suicide are to some extent shared by other Stoics (even if his concern

Socrates: epist. 13.14, 67.7; Regulus: epist. 67.7, 12. See too epist. 13.14, 98.12, dial. 1 (= prov.).2.12, dial. 9 (= tranq.).16. Edwards 2007: 87–90. 34 For some—in other cases, paradoxically, suicide can actually be motivated by the fear of death, epist. 4.4, 24.23. 35 Though nota bene epist.12.10 f. On Epicurean attitudes to suicide, see also Hill 2004: ch. 3 who stresses that some texts offer a rather different picture, most notably Cic. fin.1.49 where the Epicurean Torquatus asserts that the individual may leave life whenever he or she chooses, as though leaving the theatre. 33

332

catharine edwards

with suicide is particularly intense).36 Epictetus acknowledges Stoic teaching that suicide could be justified under intolerable circumstances, although he seems to insist on a theological endorsement.37 Closer still to Seneca is the view Cicero puts in Cato’s mouth in De finibus (3.60–62) and that outlined by Diogenes Laertius.38 According to Diogenes in his account of Zeno and later Stoics (7.130), they considered self-killing to be an appropriate action, if it would save a friend’s life, if it would benefit one’s country, or if it would allow one to escape from painful or incurable disease. Seneca invokes this as Stoic tradition. In Epistle 104 he comments that Socrates can teach us how to die when it is necessary, Zeno before it is necessary (104.21).39 According to Stoic theory, as set out by Diogenes Laertius (7.130), one might simply calculate whether the natural advantages of living are outweighed by the corresponding disadvantages.40 Seneca presents himself as readily resorting to such a process of calculation, in considering whether life continues to be worth living in the face of the physical and mental afflictions of old age (epist. 58.34f.). The term ratio, in the sense of calculation, recurs frequently in Seneca’s discussions of when is the right time to die (accounting imagery that also figures significantly in Seneca’s thinking about time, as we have seen).41 Should one anticipate the executioner or not? Sometimes this may be the appropriate course. But on other occasions to wait is better. An important example here is that of Socrates (70.9):

36 According to Griffin 1992: ch. 11. However for Rist, Seneca’s interest in suicide far exceeds that of other stoics. “Seneca’s wise man is in love with death”, comments Rist 1969: 249. For a comprehensive account of Seneca’s comments on suicide see Tadic-Gilloteaux 1963. Hill also discusses these texts in detail, arguing that Seneca “produces very little that is philosophically innovative” with regard to suicide (2004: 147). 37 Cf. 3.24.101f. Long 2002: 203f. comments: “Epictetus shows none of Seneca’s fascination with suicide, nor does he treat it, like Seneca, as the supreme test of Stoic freedom.” Cf. Droge and Tabor 1992: 34–37. 38 Though Rist 1969: 239–241 argues that according to the position set out by Cicero’s Cato, only the sapiens is ever in a position to know when it is right to kill himself. On the vagueness of this Ciceronian account, see Hill 2004: 36–41. 39 As Griffin 1992: 373 suggests, it makes most sense to interpret Seneca’s Zeno not as making an arbitrary decision but as perceiving the increasing weakness of his body (cf.epist. 58.34). 40 Cf. Cic. fin. 3.60f. On the notion of the balance sheet, see van Hooff 1990: 122, Griffin 1986: 200. 41 E.g. epist. 14.2, 24.24, 98.16. Griffin 1992: 376–380 discusses some specific examples of such calculations in the letters. On the discourse of rationes in relation to planning one’s death see also Plin. epist. 1.12.3–5 on the death of Corellius Rufus.

ethics v: death and time

333

Socrates could have brought his life to an end by abstaining from food rather than dying of the poison. Yet he passed thirty days in prison with death in prospect and not with the thought that anything could happen, that such an extended period brought many hopes but in order that he might show himself obedient to the laws and let his friends benefit from the last days of Socrates.42

Interestingly, Seneca chooses not to engage with the argument Socrates is made to advance in the Phaedo against suicide, that it is only permissible when one has received a divine sign.43 Rather he stresses Socrates’s wish to demonstrate his own respect for the laws of Athens. At the same time, the desire to benefit others, even though one might experience greater pain oneself, is also shown as a laudable motive for letting the law take its course rather than rushing to embrace death.44 The example of Drusus Libo that follows is altogether more ambiguous. Seneca seems at first to be reproaching him for not following his aunt’s advice and awaiting execution rather than taking his own life, after his conspiracy against the emperor was discovered. But Seneca then shifts tack: manus sibi attulit, non sine causa, “He laid violent hands on himself—and not without reason” (70.10). What point is there in living for another few days at one’s enemy’s pleasure? Significantly, this line of debate leads Seneca to the claim (epist. 70.11): And so you cannot make a general pronouncement on the matter of whether, when an external force decrees death, you should anticipate it or wait for it. For there are many considerations which may incline a person in one direction or the other.45

There is no general answer.46 Thus, careful consideration is always needed. Moreover, the process of reasoning is itself particularly valuable. This is a key aspect of the contemplation of suicide, which could be seen as, for Seneca, the most important philosophical exercise the would-be philosopher undertakes.47 42 Socrates potuit abstinentia finire vitam et inedia potius quam veneno mori. Triginta tamen dies in carcere et in expectatione mortis exegit, non hoc animo tamquam omnia fieri possent, tamquam multas spes tam longum tempus reciperet, sed ut praeberet se legibus, ut fruendum amicis extremum Socraten daret. 43 On the Phaedo’s discussion of suicide, see Warren 2001. On Seneca’s engagement with this see further Edwards 2007: 105. 44 Compare the example Seneca offers in Letter 98 of an elderly friend who, despite suffering pain, continues to live while he may be of service to his companions (98.15–18). 45 Non possis itaque de re in universum pronuntiare, cum mortem vis externa denuntiat, occupanda sit an expectanda. Multa enim sunt quae in utramque partem trahere possunt. 46 See Inwood 2005a: 106, 113 on the discussion of situational factors in epist. 71. 47 See Hill 2004: 151–157.

334

catharine edwards Death and Freedom

Death is to be accepted. Sometimes it is to be chosen. For Seneca death has a positive value for the opportunity it can offer to exercise virtue. The thought of death can also, under some circumstances, serve as an important source of hope—perhaps the only hope the philosopher may legitimately entertain. For death can offer a very particular kind of freedom, libertas. In Epistle 24, Seneca makes Cato, on the point of taking his own life, exclaim (epist. 24.7): “O fortune,” he said, “you have achieved nothing by impeding all my enterprises. Until this time, I fought not for my own liberty but for that of my fatherland, nor did I act with such persistence so that I might be free but so that I might live among the free. Now that our state has no future, let Cato be led to safety!”48

Similarly, Seneca has Jupiter in De providentia declare that Cato’s sword can give him libertatem, quam patriae non potuit, “the freedom it could not give his fatherland” (2.10).49 Seneca’s marked emphasis on the freedom suicide can offer could be read as a counter to the concerns of some Stoics who concluded that “if we are supposed to live according to nature, we should wait for nature to release us from life.”50 Cato’s death seems to have prompted an intense debate about the acceptability of suicide (cf. Plut. Brut. 40.4). The freedom death can offer is repeatedly stressed in the letters more generally. Death offers libertas recedendi, “the freedom to withdraw” (22.5f.). Thus death is something to be valued rather than feared: Mihi crede, Lucili, adeo mors timenda non est, ut beneficio eius nihil timendum sit, “Believe me, Lucilius, so little is death to be feared that, thanks to death, nothing is to be feared” (24.11). Epistle 26 develops this idea at some length: “meditare mortem”; qui hoc dicit, meditari libertatem iubet, “ ‘Think on death’: one who says this instructs us to think on freedom” (26.10).51 And Seneca criticizes those philosophers who exclude the possibility of committing suicide: hoc qui dicit, non videt se libertatis viam cludere, “One who says this does not see

48 “Nihil”, inquit, “egisti, fortuna, omnibus conatibus meis obstando. Non pro mea adhuc sed pro patriae libertate pugnavi, nec agebam tanta pertinacia, ut liber, sed ut inter liberos viverem. Nunc quoniam deploratae sunt res generis humani, Cato deducatur in tutum.” 49 Cf. too epist. 95.72. In epist. 14.12 f., however, Seneca sets out the view that libertas was already lost when Caesar and Pompey were in conflict and that it was not appropriate for the philosopher to take part in the struggle for power between them. 50 As Griffin 1992: 375 suggests. 51 Seneca here claims to be quoting Epicurus. Further examples in the Letters include: 66.13, 16; 70.14, 24f. De providentia also returns to this theme (dial. 1.6.7): adtendite modo et videbitis quam brevis ad libertatem et quam expedita ducat via, “only observe and you will see what a short and easy path leads to liberty”. See too dial. 6 (= cons. Marc.).20.2f.

ethics v: death and time

335

that he is shutting the gate to freedom” (70.14). The slightest of weapons will achieve this end: scalpello aperitur ad illam magnam libertatem via et puncto securitas constat, “A small blade opens the way to great liberty and peace of mind can come through a pin prick” (70.16). In Stoic philosophy, freedom (eleutheria in Greek, libertas in Latin) had come to have the sense of “total independence of the person from all passions and from all wrong desires.”52 Such an understanding of freedom could reinforce the appeal of death as a means of escape from any situation, no matter how oppressive. A key issue here must be agency.53 The option of death guarantees that action is always possible, however constrained one’s circumstances may be. As Seneca comments (epist. 26.10): One who has learned to die has unlearned slavery. He is superior to all powers, and certainly beyond their reach. What to him are prison, guards and fetters? He has an open door.54

Here, then, is at least part of the value in thinking on death, in calculating and recalculating whether and for how long one’s life may be worth living. Such exercises serve to keep the possibility of freedom forever before one’s eyes. At the same time there seems to be an ambivalence here, highlighted by Seneca’s pervasive use of military imagery. At one point, Seneca comments with regard to the freedom offered by the possibility of suicide: si pugnare non vis, licet fugere, “if you do not want to fight, you can run away” (dial. 1 [= prov.].6.7). This surely reveals a tension in Seneca’s thinking on suicide.55 For the Socrates of Plato’s Phaedo, suicide was no more to be contemplated than deserting one’s guard post (62b).56 Yet that seems to be just what Seneca is advocating in this passage from De providentia. 52 Bobzien 1998a: 339. As Inwood comments, this constitutes an “internalisation of social and political reality”. On this issue, see particularly Inwood 2005a: ch. 11, ‘Seneca on freedom and autonomy’. 53 Inwood 2005a: 306. Contrast the view of Hill 2004: 11, who argues that in Roman discussions of suicide the central issue is not agency but rather honour. His approach rightly emphasises the Roman tendency to categorise together voluntary and enforced suicides. Yet even in the case of the latter there might be considered some scope for agency which though limited is nevertheless highly valued. See further Edwards 2007: ch. 4. 54 Qui mori didicit, servire dedidicit; supra omnem potentiam est, certe extra omnem. Quid ad illum carcer, et custodia, et claustra? Liberum ostium habet. 55 Lavery 1980: 150 comments: “the suicide would appear to be a deserter in battle and a soldier who surrenders to fortune”. Another aspect of this problem is discussed by Griffin 1992: 380 f.: “If the virtue of the wise man’s actions lies in its intentions, not its result, what danger of disgraceful action can he be said to avoid through suicide?” 56 The term phrouria can also have the sense of “prison” as well as “guard-post”.

336

catharine edwards

The most extreme—and notorious—formulation of Seneca’s celebration of suicide comes in his treatise on anger, De ira. Seneca has been describing situations in which anger will inevitably arise and what the consequences might be of concealing or revealing it. Praexaspes has been punished for advising king Cambyses that he should moderate his drinking; the king demonstrates his steadiness of hand by shooting an arrow—through the heart of Praexaspes’s son. Praexaspes praises the king’s aim—he thus demonstrates that anger can be restrained under the most extreme provocation. Harpagus, the object of another king’s cruelty, finds at the king’s banquet that he has been served and has eaten the bodies of his own children. He, too, moderates his anger, flattering the monarch (dial. 5 [= de ira 3].14f.). While these stories purport to show that anger can always be concealed—ostensibly a good thing—they also reveal some profound difficulties for Seneca’s position.57 Ultimately, he cannot bring himself to endorse the restraint of either Praexaspes or Harpagus. Praexaspes is a slave in mind animo […] mancipium (de ira 3.14.3). The gods should curse him. In relation to Harpagus, Seneca comments that he should try quaerere dignam tam truci portento poenam, “to find a punishment worthy of such monstrous ferocity” (3.15.2).58 For these men, urges Seneca, suicide by any means would surely be the best option. It is to them he offers this chilling advice (3.15.4): Wherever you turn your gaze, there is an end to your troubles. Do you see that cliff? From there you can drop to freedom. Do you see that sea, that river, that well? Freedom lies in its depths. Do you see that stunted, twisted, barren tree? Freedom hangs from it. Do you see your throat, your gullet, your heart? They are the means to escape slavery. Are the ways out I’m showing you too troublesome? Do they require too much bravery, too much strength? Do you ask what may be the way to freedom? Any vein in your body!59

Detachment, Seneca understands, is and should be impossible. He cannot quite bring himself to advocate any act of resistance to tyranny other than suicide; the individual cheats the tyrant of the pleasure of his murder—the most 57 As Nussbaum 1994: 437 (cf. 435) emphasises: “The twistings and turnings of the text contain a far more complex message.” 58 Nussbaum 1994: 434 stresses the vehemence of Seneca’s language here and comments: “Seneca never seriously doubts that a parent will feel anger inside himself at these incidents, nor does he even try to suggest that it would be a good thing if he didn’t.” 59 Quocumque respexeris, ibi malorum finis est. vides illum praecipitem locum? illac ad libertatem descenditur. vides illud mare, illud flumen, illum puteum? libertas illic in imo sedet. vides illam arborem brevem, retorridam, infelicem? pendet inde libertas. vides iugulum tuum, guttur tuum, cor tuum? effugia servitutis sunt. nimis tibi operosos exitus monstro et multum animi ac roboris exigentes? quaeris quod sit ad libertatem iter? quaelibet in corpore tuo vena!

ethics v: death and time

337

effective punishment he can devise.60 Nevertheless, the decision to withdraw from the world by deciding on suicide, motivated as it is by anger, constitutes the Stoic as one deeply implicated in the world and what happens in it. The act of choosing death could convey a specifically political message.61 To celebrate death as a means of escape is to undermine the power of a regime that seeks to control its subjects through the threat of lethal punishment. This political dimension is explicit in the Stoic Epictetus’s discussion of suicide, where keen students want to demonstrate by their own deaths that tyrants have power over no one (1.9.15). Seneca alludes to the general moral weakness that afflicts his contemporaries. Yet even now some show enough spirit to seek security in death (epist. 24.11): Think about our own times, whose inertia and fastidiousness we complain about. They will include persons of every rank, of every degree of fortune, of every age who have cut short their own trouble with death.62

It is interesting that Seneca does not, in the Epistles, refer explicitly to specific exempla of self-killings from times closer to his own.63 But this more general claim certainly adduces self-inflicted death as a means of displaying qualities opposed to the moral weakness exemplified by languor and delicia. The political overtones of libertas (with which Seneca so closely associates death) are never wholly absent. In political terms, this is a kind of resistance but one that in some respects carries a heavy price.64 In Seneca’s writing we see what appears to be an increasingly extreme form of the Stoic depreciation of life. At 71.12, for instance, political change is, on one level, to be equated with the change of the seasons, something over which one has no control whatever, something that must simply be accepted. We may well feel uneasy at the implications of a philosophy that effectively discourages its adherents from taking any initiative to change a social order they find repugnant. And yet, once no choice was left, Stoicism, especially as developed in Seneca’s writing, could offer a means to make sense of a horrible death, to appropriate it as part of a virtuous life. And even before death was imminent, to think over in advance 60

See Nussbaum 1994: 436 f. The limitation of suicide, however, is that it can never make the same kind of statement on behalf of social justice that could be conveyed by a more active kind of resistance, such as an attack on the king. See Nussbaum 1994: 436 and Barton 1994: 59. 62 Respice ad haec nostra tempora, de quorum languore ac deliciis querimur; omnis ordinis homines suggerent, omnis fortunae, omnis aetatis, qui mala sua morte praeciderint. 63 Though the death of Cremutius Cordus, Marcia’s father, is discussed briefly in cons. Marc. 1.2, while that of Julius Canus receives extended treatment in tranq. 14.4–10. 64 See Nussbaum 1994: 468. 61

338

catharine edwards

how one might die was to prepare oneself against the worst, to assume an armor that might prove invincible. Seneca is by no means an enthusiastic advocate of suicide under all circumstances. In Epistle 24, having first referred to Epicurus’s criticism of those qui mortem concupiscent, “who desire death,” Seneca himself explicitly criticizes those who are obsessed with death. The brave and wise man should avoid that libido moriendi, “longing for death,” which has afflicted so many (24.25).65 “The idle and abject,” ignavos iacentesque, finding life tedious, often fall victim to a desire to die. The diurnal pleasures of the flesh slip readily into torments (24.16).66 At the same time, in Epistle 24, he also concedes that it is sometimes the noblest individuals, generosos atque acerrimae indolis viros, who are overtaken by the desire for death. While apparently condemning those who are simply tired of life, he expresses sympathy with those who despise it.67 In Epistle 30, which, beginning with the particular case of Bassus, discusses death in old age, Seneca praises the inspiration offered both by those who call for death—qui deposcunt mortem—and those who meet it in a state of calm and good cheer—qui hilares eam quietique opperiuntur (30.12). He goes on to qualify his praise for the former: illud ex rabie interdum ac repentina indignatione fit, “this first attitude is sometimes derived from frenzy and sudden anger.” Yet this is not invariably the case, as interdum makes clear. Such statements seem to betray a profound ambivalence on Seneca’s part. There is perhaps an acknowledgment that the wise man might legitimately want death. An endless life, after all, would be a life without meaning.68 Certain people say to themselves, claims Seneca (epist. 24.26): How long will these things go on? Shall I keep on waking up and going to sleep, being hungry and being full, getting cold, getting hot? There is no end to anything but all goes round in circles, one thing connected to another, each succeeding the one before? Night comes on the heels of day, day on the heels of night. Summer lapses into autumn, winter follows autumn, spring puts an end to winter. Everything passes away so that it returns again […].69 65

For Stoic criticism of the desire for death, see also Epictetus 1.9.12 and 2.15.4–12. Hill 2004: 175–178 offers a suggestive discussion of Seneca’s fastidiosi, stressing the influence of as well as the contrast with Lucretius. 67 Disapproval of those who kill themselves for frivolous reasons, out of boredom or under the influence of extreme emotion: cons. Helv. 10.9 f., tranq. 2.14f., de ira 2.36.5f., epist. 4.4. 68 Discussing Letter 12, Habinek 1982: 68 helpfully cites Bernard Williams’ argument about the meaninglessness of endless life, set out in his 1973 essay ‘The Makropoulos case: reflections on the tediousness of immortality’. 69 “Quousque eadem? Nempe expergiscar dormiam, esuriam fastidiam, algebo aestuabo. Nullius rei finis est, sed in orbem nexa sunt omnia, fugiunt ac secuntur. Diem nox premit, dies 66

ethics v: death and time

339

While earlier in the letter Seneca explicitly criticized those on whom excessive familiarity with the routines of life weighed heavily, this passage could be taken to express greater sympathy with this perspective.70 It is tempting to see an echo here of Seneca’s discussion of the structures of time in Epistle 12, where time is understood in terms oforbes, “circles” (12.6). The insistent, repetitive demands of the flesh intensify the philosopher’s disdain for the body. As Plato’s Socrates advised in the Phaedo (63a–64b), embodied life has little to offer the philosopher, who should always be preparing for death. Writing in Time Seneca’s mode of philosophy is largely paraenetic. His work offers an approach to wisdom, which is to be achieved by slow maturation, the outcome of lengthy spiritual exercises. This is a process that operates in and through time (Armisen-Marchetti 1995b: 545). As Grimal (1968: 109) suggests, it is in part Seneca’s preoccupation with the experience of everyday life that informs his particular concern with temporality. The very concept of a series of Epistles itself implies composition over time. Seneca’s letters describe incidents that appear to hook them into their author’s quotidian experience. Epistle 64, for instance, begins “Yesterday you were with us,” and describes a convivial evening of fireside talk with a group of friends. The letters of Cicero earlier and (later) Pliny the Younger, though quite possibly edited after their original time of composition, present themselves as compositions firmly situated in a particular time. Although Seneca’s letters, by contrast, do not contain the kind of references to specific events that would allow their precise dating (much to the frustration of modern scholars), nevertheless they appear as a sequence composed in order over an extended period, most notably by evoking the gradual philosophical development of Lucilius.71 Waiting for Nero’s centurion, Seneca will have been especially alert to the possibility that each letter he added might prove to be the last in the collection (as it is, it seems the final letters he wrote have not survived).72 At whatever point the series is interrupted it will be complete, he asserts—like noctem, aestas in autumnum desinit, autumno hiemps instat, quae vere conpescitur; omnia sic transeunt et revertantur. nihil novi facio, nihil novi video; fit aliquando et huius rei nausia.” 70 We might compare an observation offered as consolation for the inevitability of death in Letter 77.16: “Your pleasures have been exhausted; none of them is a novelty.” 71 On this contrast see further Edwards 2005b. 72 Aulus Gellius (12.2.3) refers to a now lost twenty-second book of letters.

340

catharine edwards

the life of the wise man. In Epistle 12, Seneca advised: sic ordinandus est dies omnis, tamquam cogat agmen et consummat atque expleat vitam, “every day should be regulated as if it concluded the series, as if it consummated and filled out our life” (12.8). It is death’s imminence that makes urgent the need to balance life’s account at the end of every day. Each day should be treated as if it were our last. This thought recurs later in the letters (epist. 61.1 f.): I am behaving as if each day were a complete life. Of course, I’m not seizing it as my last but I look upon it as if it could be my last. This is the frame of mind in which I am writing to you now, as if death might call me away, just as I am writing.73

The claim is made still more insistently in Epistle 101: qui cotidie vitae suae summam manum inposuit, non indiget tempore, “One who puts the finishing touch to his life every day is never in need of time” (101.8). But can the life of the proficiens, one who is merely on the road to philosophical understanding, be understood as complete? Is there not a profound tension between the exhortation to see life as whole, whenever it terminates, and the sense of a philosophical journey toward sapientia, a journey that death might well interrupt before the goal is attained? Yet there are perhaps other senses in which Seneca’s writings offer a more powerful challenge to the limitations of mortality. The opening of Epistle 64 moves from the recollection: Fuisti here nobiscum, “Yesterday you were with us,” to a different kind of engagement with temporality: mecum […] semper es, “you are always with me.” There is an important sense in which letters have the power to transcend constraints of both space and time. The act of writing can serve as a strategy to fix time, and thus to transcend death. The writers, too, can hope to overcome mortality. In Epistle 21, Seneca evokes the analogy of Cicero’s Epistles to Atticus, promising Lucilius renown similar to that of Cicero’s friend among future generations: “Time’s deep flood will roll over us; a few great men will put their heads above it and, though bound in the end to depart into that silence, will resist oblivion and for a long while maintain possession of themselves”74 (epist. 21.5). Achievement through writing will enable some talented individuals to maintain a presence far into the future. In his prediction se vindicabunt, “they will maintain possession of themselves,” Seneca uses a term that appeared in his exhortation to Lucilius 73 Id ago, ut mihi instar totius vitae dies sit. Nec mehercules tamquam ultimum rapio, sed sic illum aspicio, tamquam esse vel ultimus possit. Hoc animo tibi hanc epistulam scribo, tamquam me cum maxime scribentem mors evocatura sit. 74 Profunda super nos altitudo temporis veniet, pauca ingenia caput exerent et in idem quandoque silentium abitura oblivioni resistant ac se diu vindicabunt.

ethics v: death and time

341

in the opening sentence of the first letter in the collection: vindica te tibi (epist. 1.1). The longevity of great writing offers another kind of mastery of time, allowing the philosophical self, the author (and perhaps his correspondent too) the means to continue his existence far beyond the limit of human mortality. The relationship between time and philosophical writing is also a key concern in De brevitate vitae. Reading the philosophers, Seneca stresses here (echoing earlier poetic texts as well as Aristotle), can enable the reader to escape time: hi tibi dabunt ad aeternitatem iter […] haec una ratio est extendendae mortalitatis, immo in immortalitatem vertendae, “They will offer you the road to immortality […] This is the only means to prolong mortality, indeed to transform it into immortality” (15.4). The philosopher is freed from the limits that constrain others (the most significant of these being death) (15.4). The philosopher alone has the capacity to collapse distinctions between past, present, and future, to combine all times into one: longam illi vitam facit omnium temporum in unum conlatio, “Combining all times into one makes life long for him” (15.5). Philosophers teach us how to die (dial. 10 [= brev.].15.1) but at the same time communing with philosophers allows one to transcend time (15.4).75 In his Aporias: Dying—Awaiting (one another at) the “limits of truth” (1993), Derrida draws on the De brevitate vitae. Seneca’s intense engagement with death, his sense of the imminence of death, Derrida finds particularly good to model, as one contemplates “the rear-view mirror of a waiting-for-death at every moment” (1993: 55). Death limits time, death gives time its value, makes us value time. The vividness with which Seneca conveys this has appealed to many readers. But more than this, it is precisely in accepting the time-bound nature of human life, the inevitability of death that, for Seneca, we can come closest to the transcendence of both death and time. In one of the last of his letters to have survived, Seneca comments paradoxically that the human heart numquam magis divinum est, quam ubi mortalitatem suam cogitat, “is never more divine than when it reflects on its mortality” (120.14).

On this passage see Dionigi 1995a and Williams 2003 ad loc.

PHYSICS I: BODY AND SOUL

R. Scott Smith As a point of departure for our discussion of body and soul in Seneca we would do well to recall Quintilian’s assessment of Seneca’s capacity as a philosopher (inst. 10.1.129): in philosophia parum diligens, egregius tamen vitiorum insectator fuit, “in philosophy he was not meticulous enough, yet he was outstanding in his persecution of vices.” If Seneca was not concerned with providing a thorough account of the physical nature of the body and soul, we need not look far for a reason. His preoccupation with ethics left little time for the knotty problems of logic and physics; those problems, after all, simply obscure the truly important questions of moral improvement and the pursuit of the good life. So it is hardly a surprise that Seneca did not devote an entire treatise to the soul, as Aristotle and his Stoic predecessor Chrysippus had done, or investigate human physiology and psychology in a systematic way, despite having ample opportunity to do so in, say, the Naturales quaestiones. When he does mention the nature of the soul or body, it is usually in the context of other (usually ethical) discussions; the state of the evidence, therefore, makes it difficult to reconstruct his views on body and soul—if indeed he ever had a thoroughly thought out position on the matter. On analysis, it appears that he is largely an orthodox Stoic who tends to refrain from specifics, but one who can and sometimes does apply independent and critical judgment to certain problems. Thus, it seems methodologically best to proceed by judging Seneca’s position against that of his Stoic predecessors—keeping in mind that we are often not as well informed about the earlier Stoics’ conception of body and soul as we might like. Any analysis of body and soul in Seneca must begin with a review of the basic Stoic physical and metaphysical principles that will be discussed in more detail in the chapter “Physics II: Cosmology and Natural Philosophy” by Bardo Maria Gauly, infra, pp. 363–378. For just as the cosmos itself was held to be a compound of an active, intelligent, self-moving entity, and a passive, insensate, inert one, so too were human beings thought to be amalgams of active (soul) and passive (body) elements. According to the Stoics, all things within the cosmos operated on the same principles; human beings were no exception.

344

r. scott smith

Like the Epicureans but unlike Plato, the Stoics were committed materialists. What this means, in essence, is that all things that exist are corporeal, and that all events are a result of cause and effect based on corporeal bodies interacting.1 From a metaphysical standpoint there exists a single underlying substance that operates differently on two principles (archai), one active (to poioun or theos/logos) and one passive (to paschon or hyle).2 The active principle, identified by the Stoics under various names (god, nature, reason, intellect, mind, breath, creative fire, etc.), unifies and gives form to the unqualified passive matter. Both of these principles operate as, or one might say through, corporeal entities, that is, the elements (active = fire, air; passive = earth, water). According to the most developed theory, that of Chrysippus,3 the cosmos was composed of a compound of inert matter and active pneuma, a sort of warm vapor that unifies and gives form to unqualified matter by means of physical tension (tonos or tonike kinesis).4 The Stoics explained the cohesive workings of pneuma as “tensional movement” or breath “turning back on itself,” consisting of a simultaneous outward and inward flow that creates the tension required for unity.5 It is important to note that the active and passive principles are nowhere found in isolation. Both the active principle and inert matter, though discussed as if separate entities, are in fact inseparable in Stoic cosmology. Briefly, “acting upon” and “being acted upon” are two perspectives of the same process, one in which both principles participate.6 The distinction made between active and passive principles is merely conceptual. This pneumatic tension works both on the cosmic scale, holding together the whole cosmos, as well as within individual objects. But not all things are created equal, and pneuma works differently in different bodies. There developed in Stoicism, derived from Aristotle’s formulation (De anima 2.2f.), a hierarchy of corporeal objects in which beings were categorized according to more or less sophisticated forms of pneuma, the so-called scala naturae.7 At the most basic level there are inanimate objects that

For Seneca, see epist. 106.8, 117.7. Diog. Laert. 7.134, 139 (SVF 2.299f.), 7.150 (SVF 2.316). See Sambursky 1959, Lapidge 1973, Hahm 1977; for Seneca, see Wildberger 2006. 3 Hahm 1977: 163–169. 4 SVF 2.441, 444, 448, 451. 5 SVF 2.442, 451, 452. Specifically, the outward flow contributes form and qualities, the inward flow unity. 6 Scarpat 1965: 126–155, Lapidge 1973. 7 The locus classicus is Origen, De principiis 3.1.2 f. (= SVF 2.988; cf. 2.989). See Hahm 1977: 164f., Long 1982: 37–39, Inwood 1985: 18–27, Annas 1992: 51–54, Wildberger 2006, 1: 205–243, Graver 2007: 19–21. 2

physics i: body and soul

345

merely have coherence, said to be held together by pneuma in the form of hexis (“coherence”), for instance a stone or a lump of iron. A more complex form of pneuma called physis (“growth” or “nature”) embodies vegetation, which in addition to coherence has metabolism and may change through growth. Even more sophisticated are animals embodied by soul-pneuma (psyche), which bestows, in addition to coherence and metabolism, sentience and auto-locomotion. Human beings, as a special class of animals, are endowed with the same powers as mute animals, but their psyche operates rationally. Thus, when we say that according to the Stoics a human being is composed of body and soul, this is just another manifestation of their physical conception of how the world works. The human soul is the active principle, a rarefied and sophisticated kind of pneuma, one that unifies, animates, and qualifies the passive body in a specific manner, which endows us with, in addition to the functions of lower forms of being, reason. The soul comes into existence when the animal organism is born8 and occupies the same physical space as the body until their separation at death. Galen attributes to Chrysippus a specific definition of soul: “the pneuma innate in us, continuous, running through the whole body, so long as the breath of life remains in the body” (SVF 2.885; cf. Diog. Laert. 7.156 = SVF 2.774, SVF 2.778). To a modern reader the notion that one corporeal form can pass through another, without recourse to void, runs counter to modern scientific axioms. The Stoics explained this process as an example of “complete mixing,” whereby two bodies “extend through one another wholly, but in a way that each retains its own substance and qualities,” just as fire might pass through iron yet each retain its own essence (SVF 2.473).9 The human soul, a superfine corporeal body (soma leptomeres, SVF 2.780) made of warm air, completely mixes with and gives unity to the heavier corporeal passive elements. In other words, there is no portion of body that is not at one and the same time occupied by soul—just as a stone is fully pervaded by hexis. It is worth reiterating that the body is not a vessel or container for the soul (as the Epicureans held), but rather the soul is the substance that holds the 8 Two sources, probably derived ultimately from Chrysippus, inform us that the embryo prior to birth is equivalent to vegetation guided by physis (Hierocles col. 1.15–28 BastianiniLong, Plut. mor. [= De Stoic. repug.] 1052F–1053C). As the fetus approaches birth, the pneuma becomes more and more refined, and at birth this refined pneuma is either “hardened” (Hierocles) or “cooled” (Chrysippus), resulting in the immediate change from physis to psyche. If I read epist. 124 correctly, Seneca hints at the non-animal nature of the fetus: see onus (epist. 124.8) and his comparison of plant life (ch. 8, 11); cf. epist. 121.17. 9 See Annas 1992: 47–50, Long 1982: 38 f., Long 1999: 561 f.

346

r. scott smith

compound together, as Posidonius notes in his criticism of the Epicureans: “it is not bodies which hold souls together, but souls bodies, just as glue controls both itself and what is outside it.”10 Seneca on Body and Soul Seneca, who is an orthodox Stoic in his metaphysical concept of the world (epist. 65.2), likewise adheres to the Stoic concept of the scala naturae outlined above, as demonstrated in an illustrative passage (epist. 58.14): How do I divide body? Thus: either animate or inanimate. Now, how do I divide animate beings? Thus: certain animate beings have animus, others only anima; or another way: certain ones have impetus, move, and pass through space; others, planted in the soil, are nourished by roots and grow.11

To these categories we may add Seneca’s earlier definition of inanimate objects such as stones as “lacking anima” (58.10; cf. dial. 4 [= de ira 2].26.4). These classifications correspond, with some variation of terminology, to those found in the Stoic sources outlined above. Those inanimate objects the Stoics said were held together by hexis (“coherence”) Seneca calls “lacking anima,” the principle of life.12 Those beings that are guided by physis (“growth”) Seneca simply describes as “having anima.”13 Those more sophisticated living beings (animals) embodied by soul-pneuma (psyche) are said to have animus, the ruling part of the soul, which endows the animal with sentience and autolocomotion. Although Seneca does not include humans in this passage, he often comments on the unique position that they hold at the top of the scala naturae because of their rational natures. Because their psyche operates rationally, they have a share in the divine (epist. 65.24): “the position that god holds in this cosmos is the position the soul (animus) has in a human being. What is materia there, is in us body (corpus).” Seneca maintained, just as

Posid. frg. 149 Kidd-Edelstein (trans. Kidd). For Seneca, see nat. 2.6.6. Corpus quomodo divido? ut dicam: aut animantia sunt aut inanima. rursus animantia quemadmodum divido? ut dicam: quaedam animum habent, quaedam tantum animam, aut sic: quaedam impetum habent, incedunt, transeunt, quaedam solo adfixa radicibus aluntur, crescunt. Cf. epist. 76.8–11. 12 He is aware of the concept of hexis: see nat. 6.16.1; cf. 2.6.2, 2.6.6. 13 Wildberger 2006, 1: 210 suggests that Seneca’s categories here are contaminated with Peripatetic elements (Aristotle thought plant life was empsychos “ensouled,” hence anima), but Seneca is merely applying the Latin lexicon to Stoic concepts. See further idem, 2: 759–760 nn. 1006–1008. 11

physics i: body and soul

347

other Stoics did, 14 that the human soul is but a fragment of the cosmic soul. He states that (epist. 66.12) “reason is nothing other than a part of the divine spiritus sunk into the human body,” and on numerous occasions he notes the soul’s divine origins or divine nature.15 It is necessary to point out that all objects are, according to strict Stoic cosmology, part of and pervaded by cosmic pneuma in some form, but there is something intrinsically different about humans, the capacity to reason, that connects them more closely to the completely perfected, fully rational god of the cosmos. Before pressing on with a closer examination of the human body and soul in Seneca, we should take a moment to reflect on the philosophical lexicon he employs when discussing such matters. Although the physical conception of the Stoic psyche outlined supra seems clear enough, we shall presently discover that the word “soul” is ambiguous, vague, and often unhelpful. Difficulty arises partly because there is an inherent ambiguity in the Greek word psyche generally and in its Stoic use specifically, but also because Latin had at its disposal numerous pre-existing terms for what confers life and mental function on a human being (e.g., animus, anima, spiritus, mens). As we have just seen, Seneca shuns verba e verbis translations of hexis and physis, choosing to use the common Latin term anima (“breath” as “principle of life”) as the feature differentiating animate from inanimate objects. Anima and animus, to be sure, had a long life in Latin prior to Seneca16 and were readily adopted by Latin translators of Greek philosophy. The Epicurean Lucretius, for instance, uses the term animus for the controlling center of the soul (located in the heart) but reserves anima for the non-sentient soul-material that pervades the rest of the body.17 Seneca similarly employs the term animus for the Stoic concept of the mind as a center of psychic activity (so also mens);18 in some passages he clearly equates it with the Stoic hegemonikon, the Stoic directive faculty of the soul located in or around the heart, which

14 Diog. Laert. 7.143 (= SVF 2.633), “from the fact that our soul is a fragment of it [scil. the cosmic soul]”; cf. SVF 2.634. 15 Dial. 6 (= cons. Marc.).24.5; 11 (= cons. Pol.).9.8; 12 (= cons. Helv.).6.7, 11.6; epist. 31.11 (“what else would you call the animus but god living in a human body?”), 41.2, 41.5, 79.12, 92.34, 120.14; cf. nat. 1. pr. 11. Seneca, interestingly, also applies the soul/body dichotomy to the political structure of emperor/subjects at the beginning of clem. 1 (suggesting the divine nature of the emperor’s position?). 16 See, especially, Reis 1962. 17 Lucr. 3.94ff. The available Latin lexicon, with its implicit dualism in the terms animus and anima, easily fit into the Epicurean scheme of the logikon and alogon parts of the soul. See Lathière 1972, Bailey 1947 ad loc. 18 See Grimal 1992b.

348

r. scott smith

he occasionally translates with a technical term (principale19). Anima, on the other hand, never stands for the ruling part of the soul. It denotes specifically the substance that provides life.20 The essential difference may gleaned from Seneca’s own sententia (epist. 4.4): difficile est … animum perducere ad contemptionem animae, “it is difficult to persuade the mind (animus) to regard life (anima) lightly” (cf. nat. 6.32.4). The ambiguity in the Stoic use of psyche also presents a difficulty. It is one thing to say that according to Stoic physical principles a human is composed of body and soul, quite another to identify exactly how these general physical principles play out in all the complexities of a human organism. Is the psyche to be viewed as a singular substance that subsumes hexis and/or physis in addition to providing sentience, locomotion, and (in humans) rational thought? Or does pneuma operate on three different levels independently, with the lower levels of hexis and physis cooperating somehow with a higher form of soul located in the ruling part? If there was a singular, orthodox position on this question, we cannot recover it. The evidence is ambiguous at best, and it appears that Seneca’s Stoic predecessors also wrestled with this problem. Sextus Empiricus (adv. math. 7.234, not included in SVF) reports that the Stoics used the term psyche in two distinct senses, to denote 1) the substance that holds together the whole compound, and 2) specifically the ruling center, i.e., the hegemonikon (for convenience, termed soul1 and soul2 below). Sextus goes on to explain that the Stoics, when speaking of a human being as a composite of soul and body or describing death as the separation of soul from body, were specifically referring to the soul in the second sense. There is not enough space here to review fully the debate concerning the functions of the psyche in early Stoic thought,21 but it is sufficiently clear that Seneca regarded soul2 (the leading part) as distinct from soul1 (the pneuma that held the body together). In his extensive treatment of the various workings of spiritus (pneuma) in nat. 2 he emphasizes that the coherence of human bodies is due to spiritus (corpora nostra inter se cohaerent. quid enim est aliud quod teneat illa quam spiritus? “our bodies 19

Dial. 3 (= de ira 1).3.7, epist. 92.1, 113.23, 121.10 and 13. It is anima that is regularly said to leave the body at death (Wildberger 2006, 2: 759 n. 1007); in the few cases where animus is said to leave, it is always with the implication of sensation of the event (e.g., dial. 9 [= tranq.].14.9). Seneca occasionally conflates anima and animus as the entity that leaves the body at death (see, e.g., epist. 57.7–9 and Berno 2006a ad loc.). See also the discussion of Sextus Empiricus below. 21 See Bonhöffer 1890: 69, 106f., Voelke 1965, Long 1982, Inwood 1985: 25f., Graver 2007: 19–21. 20

physics i: body and soul

349

are unified; what else would it be that holds them together other than spiritus?”: 2.6.6), i.e., pneuma as hexis. One might argue that this spiritus is the same as that which performs the higher psychic functions, but this cannot be right, for later Seneca refers specifically to the spiritus “by which the earth is held together and unites its parts and which is also present in stones and dead bodies” (qui inest etiam saxis mortuisque corporibus: nat. 6.16.1). What separates dead Caesar from live Caesar is the absence of at least a secondary (anima) and perhaps a tertiary (animus) level of soul-pneuma that either dissipates or leaves the body intact after death. For Seneca, then, the body is not merely unqualified passive matter held together by spiritus, but that matter plus the spiritus that gives coherence and form to that matter. Anima further animates and provides the warmth characteristic of living things;22 soul2 (animus) endows the living thing with perception and the impulse to move. Seneca’s view on the matter—as far as the evidence suggests—is similar to Long’s interpretation of the earlier Stoics’ position: the body is “earth and water informed by cohesive and vegetative (soul) pneuma, but not specific soul pneuma.”23 Since Seneca privileges the ruling part of the soul as the carrier of our identity and the locus of moral responsibility, a clear dualism arises not between soul1 + 2 and body (inert elements), but between soul2 and the amalgamation of body and soul1. Since spiritus and anima have nothing to do with decision making, they are hardly significant in ethical matters. For all its importance as a life-giving force and its participation in perception and movement, soul1 might as well be, and was considered by Seneca, just part of the body. The Relationship of the Body and the Soul Despite the layered complexity of the Stoic view of the soul, for sentience and locomotion to occur there has to be some cooperation between the ruling part of the soul (hegemonikon/animus) and the rest of the body (inert elements + spiritus/anima). There was considerable debate among the Stoics themselves concerning the psyche (wholly conceived) and its parts and

Nat. 2.10.3; see Hahm 1977: 70. Long 1982: 40, Annas 1992: 55f.; for a contrary view see Bonhöffer 1890: 69f., 106, Inwood 1985: 25f. It would be misleading not to point out here that there may have been (and probably were) competing (or eccentric) views of the soul among Seneca’s Stoic predecessors and that Sextus may have been privileging one particular position over another. 23

350

r. scott smith

faculties, but Zeno’s division of it into eight parts became orthodox:24 the ruling part (hegemonikon), the parts involving the five senses, and those two parts having to do with voice and reproduction. Of these, the ruling part was said to be located in or around the heart, the other seven being pneumatic outgrowths of and extending out from the ruling part into the body, which the Stoics likened to the tentacles of an octopus.25 Another source calls these branches “perceptive exhalations.”26 These subordinate parts are both extensions of soul1 and markers of their functionality vis-à-vis the ruling part. For example, the part responsible for the sense of sight connects the eyes to the ruling center, that responsible for hearing leads from the ears, and so on. The Stoic conception of pneumatic branches stretching out from a command center has often been likened to the modern conception of the central nervous system.27 Seneca, however, nowhere engages in an examination of the parts of the soul (he comes closest at epist. 90.29), concentrating primarily on the hegemonikon/animus as the locus of decision making. Whether he was in fundamental agreement with the (more or less orthodox) view outlined supra is impossible to tell, although it stands to reason that this would have been part of his Stoic training. But Seneca’s silence on the matter is especially noticeable, and it highlights his preoccupation with the ruling part (animus). This is, no doubt, rooted in his privileging of ethical issues. The ruling part, after all, is what ultimately matters when assigning moral or ethical value to actions, since it and it alone is responsible for all decision making. Seneca occasionally provides a glimpse into the physical makeup of the animus. It is (epist. 50.6) “spiritus in a particular state” (Greek πνεῦµά πως ἔχον). Elsewhere (epist. 57.8) it is noted to be a superfine substance (ex tenuissimo constat), more rarefied than fire (tenuior est igne), and capable of passing through any corporeal substance (cf. dial. 12 [= cons. Helv.].11.6f.). Yet, when all is said and done, not even the stuff nor the whereabouts of the ruling part is truly significant—all that matters is what the animus does. Consider epist. 121.12, one of the many places where Seneca reveals his skepticism: “We know that we have an animus; what animus is, where it is located, what its nature is, or how it comes to be, we do not know.” Again at nat. 7.25.2: “that we have an animus, which drives us and

24 25 26 27

SVF 1.143; 2.827 f., 830–833, 836, 874; cf. Sen. epist. 92.1. SVF 2.836. SVF 1.141. E.g., Annas 1992: 61 f., Long 1999: 562–572.

physics i: body and soul

351

holds us back, everyone will admit. What this director and master of us is, however, no one will inform you any more than where it is” (cf. epist. 88.34; 90.29).28 Despite Seneca’s reluctance to assign a location to the animus, the role it plays in his psychology is clear enough. At its most basic level it is the nexus of the physical web of pneuma stretching out into the body that endows the organism with perception and locomotion. It is also the physical locus of the constellation of powers that lead to action: perception/presentation, assent, and impulse. Since the physics of action and emotions are discussed by Margaret Graver (supra, pp. 257–275), I intend here to focus only on the relationship between the ruling part of the soul and the rest of the organism, beginning with self-perception and the relationship of the hegemonic animus and the body of the organism. Epist. 121 is the most illuminating source for Seneca’s view on the relationship between the ruling part of the soul and the rest of the organism. In this letter Seneca, drawing from the theories of the Stoics Archedemus and Posidonius (121.1), argues that all animals, including humans, are aware of their own constitution at birth (constitutionis suae sensus: 121.5), a function of the specific physical connection between the ruling part of the soul (soul2) and the rest of the body. Seneca then provides what is apparently the Stoic definition of constitutio (121.10): “the ruling center of the soul in a specific relation to the body.”29 This intimate connection explains, for instance, how a bird intuitively knows it has a wing and not an arm, or how animals are able to move all their limbs expeditiously from the moment of birth. It also explains why a toddler will try to stand despite the pain of falling down or why a tortoise that has been flipped over will try to right itself in order to return to its natural disposition (121.8). The constitution of an animal—that is, what the organism fundamentally is at any one time—is intuitively and immediately understood by the ruling part. The ruling part, moreover, is aware of itself (121.12) as the locus of that awareness.

28 Related to Seneca’s agnosticism on this point is that he does not venture into the complex physiological relationship among the heart, blood, pneuma, hegemonikon, and the nerves, nor does he explain how respiration and circulation contribute or fail to contribute to the sustenance of the soul-pneuma (although see the suggestive comments at nat. 3.15.1f. and 6.14.1). One might compare the agnostic position found at Cic. ac. 2.124: sed redeo ad animum et corpus, satisne tandem ea nota sunt nobis, quae nervorum natura sit, quae venarum? tenemusne quid sit animus, ubi sit? Cf. Cic. Tusc. 1.24. 29 Constitutio […] principale animi quodam modo se habens erga corpus; see Inwood 1985: 313 n. 40.

352

r. scott smith

Seneca unfortunately does not analyze this process in detail, but the underlying physical principles can be reconstructed from an invaluable second-century ad papyrus containing Hierocles’s account of animalian self-perception,30 an account that has important points of contact with the Senecan letter. Hierocles’s Elements of Ethics begins with the foundational argument that animals have self-perception at birth. His exposition is thankfully systematic and clear: 1) animals are composites of body and soul, both corporeal and able to be impacted; 2) body and soul are completely mixed, resulting in a mutual “grasp”; 3) the soul stretches outward from and back to the ruling center, striking all the parts of the body, which reciprocates the impact because both entities are corporeal and offer resistance; 4) the resulting experience (pathos) from this contact is carried back from the outer parts to the ruling part (hegemonia) located in the chest, resulting in continuous perception of all parts of both body and soul. This, Hierocles says, is the equivalent of the animal’s perceiving itself.31 In other words, the animal is aware of itself because of the physical push and pull caused by the physical contact between the corporeal soul and corporeal body constantly registered, even if unconsciously, in the directive faculty of the soul. Although Seneca nowhere articulates this particular physical construction of the soul, he is clearly committed to a model like that of Hierocles since it forms the underpinning of the concept of oikeiosis (see discussion below) he is concerned with in epist. 121.32 If, after all, such an awareness of one’s constitution—one that is, moreover, constantly changing—is immediately attendant at birth (121.6) and lasts until death (121.15), it follows that there must exist a process whereby the ruling part of the soul is intimately aware of the coextension and specific nature of the body/soul compound to which it belongs.33 Seneca’s position is based on such a physical arrangement even if he does not go into detail here.

Bastianini and Long 1992. See Inwood 1984, Long 1993, Graver 2007: 23f. Hierocles col. 4.39–53 B.-L. The term borrowed from modern neurology for such inwardlooking self-awareness is “proprioception.” See Long 1993: 96–101 with earlier bibliography. 32 For recent commentary on this letter, see Inwood 2007a. 33 Seneca and Hierocles (col. 5.52–6.24 B.-L.) also agree that self-perception is a prerequisite for the perception of externals (aisthesis/sensus). Seneca notes that an animal must “perceive that thing through which they perceive other things as well” (121.12) and that there has to be a point of reference—the self—to which other things are referred (121.17). If, for example, I reach out my hand and touch a book, there must be an awareness of the hand that is doing the touching, and I must be aware that that hand belongs to me. Perception only becomes meaningful when it is self-referential. See Annas 1992: 71f. 31

physics i: body and soul

353

But the animal not only perceives that which belongs to itself, it also naturally feels an attachment to, or one might say ownership of, the body/soul compound, a sense of individuality that separates the organism from everything else around it. This brings us to the important Stoic concept ofoikeiosis,34 a notoriously slippery term that is variously translated as “appropriation,” “orientation,” or “ownership,” although no single English word adequately expresses the full meaning of the word. In essence, oikeiosis is “an animal’s innate disposition to be concerned about and motivated by what belongs to itself.”35 This includes, at the most instinctual level, self-preservation (cf. Cic. fin. 3.16ff.). The Stoics claimed that (Diog. Laert. 7.85 = SVF 3.178) “the first impulse of an animal is toward self-preservation, since Nature from the beginning endears the animal to itself.” The same passage of Diogenes quotes Chrysippus verbatim, “the first thing every animal is concerned with [proton oikeion] is its own constitution and its awareness of it.” Seneca follows suit (epist. 121.21), “each animal cares for its own safety, and it seeks what will benefit it and shrinks from what will harm it” (cf. Epikt. ench. 31). For humans, before the onset of rationality at around age fourteen,36 our primary motivation is rather like that of mute animals: we are driven by the natural impulse for survival and thus are concerned primarily with our bodily existence. Once, however, we have become fully mature adults, our constitutions have fundamentally changed, and our primary drive is no longer rooted in biological survival but in the health and preservation of the rational animus. In other words, the preservation of our rational selves takes precedence over our bodily selves. This is an important step, and it has great ethical implications. But despite this new set of ethical priorities centered on the rational animus, there remains an innate concern for the body, the residue, perhaps, of that initial impulse toward self-preservation. Seneca opens epist. 14 by asserting that there is a natural concern for the body:

34 See discussion of this topic in Pembroke 1971, Inwood 1984: 184–201, Engberg-Pedersen 1990, Long 1993. 35 Long 1993: 97. 36 At birth the human psyche is virtually equivalent to that of a mute animal, whose ruling centers are underdeveloped and non-rational, with one important exception: an infant has the potential to become a rational creature. See Seneca epist. 124.9, “there is a non-rational creature [mute animal]; another that is not yet rational [infant]; yet another that is rational but not yet consummate” (cf. epist. 49.11). Around fourteen, the soul finally becomes mature when “rationality supervenes on the soul as the craftsman of impulse” (Diog. Laert. 7.86), that is, when a human is considered fully rational and so could reasonably be held responsible for her or his actions. Seneca at epist. 118.14 calls the infant inrationalis but the pubes rationalis.

354

r. scott smith Yes, there is in us an innate affection (caritas) for our bodies, and yes, we are their guardians. I am not saying that we should not indulge them, only that we should not be slaves to them […] (2) we ought to behave not as though we must live because of our bodies, but as though we cannot live without them.

This sentiment that there is a natural, instinctual concern for self-preservation is frequently echoed in Seneca’s letters (e.g., 36.8, 82.15, 121.24). But as rational animals we are expected to act accordingly, and other factors take precedence over survival. Although our body might urge us to drink or eat, as a rational agent we may reason that this may not be the appropriate course of action, and we may decide not to take food or drink. Or, although we may wish to preserve our biological lives, it may be more appropriate to preserve our independence and so we will choose to forfeit our life to achieve that end. As Seneca sums up in the letter (epist. 14.2), the care of the body should be such that “if reason, dignity, or honor demands it, we are to throw it into flames.” As our physical makeup changes and we become fully rational adults, our responsibility shifts from the corpus to the animus, and one of philosophy’s main goals is to turn our attention specifically in that direction. The Relative Value of the Body and the Soul The early Stoics, it seems, did not dismiss the body as an unimportant appendage (good health, for instance, was still advantageous, even if secondary to virtue), although they clearly considered it to be inferior to the soul.37 But there is very little to suggest that they denigrated the body38 like later Stoics such as Seneca and Epictetus, who were more willing to emphasize the sort of body/soul dualism advocated by Platonism.39 In epist. 65 Seneca emphasizes the supreme importance he places on the soul to the disadvantage of the body.40 In opposition to the soul, which he calls “the better” or “best” part of a human being (65.18; cf. dial. 4 [= de ira 2].14.2, epist. 74.16, 76.9, 78.10; nat. 1. pr. 14), more powerful and precious (65.23), and naturally superior (65.24),41 he often degrades the body with great vigor. He defends his contemplation of the natural world by arguing that it “lifts up

SVF 3.136. The closest we come is at SVF 1.529, where the body acts as a “bitter tyrant” demanding daily tribute and threatening disease and death if its needs are not met. 39 See Dobbin 1998: 70 f., on Epict. 1.1.7; cf. Husner 1924: 25–27, Long 2002: 157f. 40 Scarpat 1965: 239–258; most recently Inwood 2007a. 41 At epist. 92.33 the animus is regarded as a procurator of the “necessary burden” that is the body. 38

physics i: body and soul

355

and relieves the soul (animus), which, oppressed by its heavy load, desires to free itself […]. For this body is the soul’s (animus) burden and punishment; the soul is weighed down by the body’s weight and imprisoned” (65.16).42 Later in the letter (65.17–24) he calls the body “a grim and murky domicile” that confines and enslaves the soul, and, like Epictetus, refers to his body with the diminutive corpusculum (“little body”), a pointed word that will recur twice in the next letter.43 The body can blunt the powers of the mind (dial. 4 [= de ira 2].10.1, 20.2; epist. 15.2, 58.33, 88.19), but if properly trained the animus can remain agile despite the loss of bodily vigor (epist. 30.13, 66 passim). Seneca’s low estimation of the body’s value can be seen when he expatiates on the free will of a slave despite his legal status (benef. 3.20.1): Whoever thinks that slavery penetrates the whole man is mistaken. The better part of him is exempt. Our bodies are subject to and awarded to masters, but the mind (mens) is wholly independent: so free and unshackled is it that it cannot be confined by this prison in which it is enclosed such that it is prevented from using its own impulse. […]44

The body—i.e., inert elements + soul1—has no bearing on the freedom or quality of the person, here identified entirely with the ruling part of the soul. A person’s worth is equivalent to the quality of the animus (epist. 76.32): “if you want to determine the true worth of a person and get to know what sort of man he is, inspect him naked: forget his estate, forget his honors and all of Fortune’s deceptive gifts, let him even shed his very body—inspect his soul (animus), its quality, its capacity, whether it is great on its own or because of some external factor.” Departing from traditional Roman values, Seneca argues that the same animus can be present in an eques, a libertinus, or a servus (epist. 31.11). The body is nothing more than clothing; even the most beautiful garment cannot cover up an inferior body (epist. 92.13). Seneca’s position is comparable to what we find later in Epictetus, who, as Long in his

42 Body as prison: dial. 6 (= cons. Marc.).20.2, 11 (= cons. Pol.).9.3, 12 (= cons. Helv.).11.7, epist. 102.30; cf. Plat. Phaid. 62B. 43 Corpusculum as a derogatory word: dial. 12 (= cons. Helv.).11.7, benef. 4.13.1, epist. 23.6, 24.16–18, 41.4, 58.29. I regard the cluster of letters 65–66 (as well as 23–24) as bound tightly to each other around this theme; Claranus in epist. 66 offers a positive exemplum of the theme “the independence of the mind from the body” treated in the previous letter. See Maurach 1970: 137 f.; Scarpat 1965: 72. For an extensive treatment of Seneca’s metaphorical descriptions of the body (slave, domicile, cave, vessel, clothing, etc.) see Husner 1924. 44 Errat, si quis existimat servitutem in totum hominem descendere. pars melior eius excepta est. corpora obnoxia sunt et adscripta dominis; mens quidem sui iuris, quae adeo libera et vaga est, ut ne ab hoc quidem carcere, cui inclusa est, teneri queat, quominus impetu suo utatur.

356

r. scott smith

recent treatment notes (2002: 157), “frequently denigrates our ‘bodily’ aspect, often calling the body a corpse or mud or earth, and using diminutives ‘little flesh’ or ‘little body’.” This condemnation of the body is, of course, part and parcel of elevating the soul as the true self that has autonomy and is free from all externals. In his programmatic first Discourse, Epictetus outlines the differences between “what is up to us” (ta eph’ hemin) and what lies outside our power. External forces we cannot control; what we are in control of is our mind. So strong is the identification of the animus as the true “self” that Seneca can write (dial. 10 [= brev.].19.1), “[do we not want to learn] where nature will settle us [nos] when we have been sent from our bodies?” This mind/body dualism found in Seneca should not be confused with psychological dualism, wherein there are separate, competing powers or faculties in the soul itself (such as those articulated by Plato or Aristotle), which could produce conflicting impulses to action. In his clearest exposition of the question,45 Seneca fits squarely within the orthodox Stoic view that the animus (hegemonikon) is a solitary faculty and is alone responsible for human action. Whether someone is guided by right reason or passion is not a matter of one impulse overpowering another, or of accepting one impulse over another, but of the current state of the rational soul. Yet it is easy to conflate the mind/body dualism with psychological dualism in Seneca’s case, for example at epist. 71.27:46 I do not remove him [illum, scil. the sapiens] from the category of humanity, nor do I say that he is impervious to pain like some cliff that does not admit any feeling. I remember that he is composed of two parts. One is non-rational; this is bitten, burns, and feels pain. The other is rational; this has unshakeable opinions, is fearless and unconquerable.47

Yet, despite lexical appearances (pars rationalis/inrationalis), this has nothing to do with an irrational part of the soul (animus). The pars inrationalis refers, as Seneca himself later notes (71.29: note the repetition of the verb dolere), to the body and not a separate faculty of the animus: et tremet sapiens et dolebit et expallescet; hi enim omnes corporis sensus sunt (“the sapiens will tremble and pain and grow pale, for all of these are sensations of the body”). Once

Dial. 3 (= de ira 1).8.1f., discussed in full by Graver (supra, pp. 268f.); see also Inwood

1993. 46

Zöller 2003: 134; so, apparently, Pohlenz 1948–1949: 308, Ganss 1952: 32. The key sentence: Memini ex duabus illum partibus esse compositum: altera est inrationalis, haec mordetur, uritur, dolet; altera rationalis, haec inconcussas opiniones habet, intrepida est et indomita. It is important to note in this discussion that illum refers specifically to the sapiens, not his animus as Zöller (supra n. 46) takes it. 47

physics i: body and soul

357

one recognizes the important distinction between animus (soul2) and body (inert elements + spiritus and anima), all apparent difficulty disappears.48 The animus is the power that controls what we could call voluntary action.49 The body may be responsible for feeling pain (registered, of course, in the animus), but the sapiens will not be influenced by that pain, and will instead act in accordance with his inconcussae opiniones despite it. Death and the Survival of the Soul Death occurs at the moment when the soul and body separate. Chrysippus, like Plato, speaks simply of the “separation of the soul from the body” (SVF 2.604, 790, 815; cf. 1.137, 146, Plat. Phaid. 67D), and at the time of the soul’s departure the animal is said to cease living and become merely a corpse. Seneca defines death specifically as “the moment when the anima departs from the body” (quo [tempore] anima discedit a corpore: dial. 1 [= prov.].6.9),50 but he also vividly portrays the soul “bursting forth” and escaping through any earthly substance (erumpit; […] animo […] per omne corpus fuga est: epist. 57.8; cf. dial. 6 [= cons. Marc.].23.2).51 The soul is occasionally said to escape through the mouth or a wound, but the latter instance of this may be no more than a bon mot.52 This notion of the soul actively “leaving” the body may be a reflection of Seneca’s metaphysical conception of the soul as a portion of the divine: it is an active, vigorous entity always in motion, striving to return to the place from which it descended.

See further Hadot 1969: 91 f., Kidd 1988: 666 ad frg. 184, Inwood 2007a. Seneca provides an extensive list of the kinds of responses that are involuntary, many of which are purely bodily (see dial. 4 [= de ira 2].2.1–5, 3.2, 4.2; epist. 57.4), e.g., shivers caused by freezing water or revulsion at the touch of certain objects. That thesapiens will feel pain is a natural and physical response to, say, being cut by a sharp knife, just as the animus cannot completely block out a loud, crashing noise—both the cut and the noise register as physical alterations of soul2. What is “up to the sapiens,” however, is the response to the cut or noise. More difficult to explain are the involuntary affectations that we would describe as mental in nature; see Graver’s discussion (supra, pp. 257–275). Because of space constraints I have here omitted discussion of the “will” in Seneca, whether it was a separate faculty of the soul (termed “traditional will” by Inwood 2000) or simply an index for multiple functions of the mind (“summary will”): I refer the reader to Pohlenz 1941: 112–118, Gilbert 1963, Rist 1969: 219–232, Voelke 1973: 161–190, Dihle 1982: esp. 134 f., Kahn 1988, Inwood 2000, Zöller 2003. 50 Cf. nat. 2.59.3. See also n. 20 supra. 51 Seneca perhaps plays with this idea when describing Claudius’ death at apoc. 4.2–3; see Weinreich 1923: 53–55. 52 Dial. 5 (= de ira 3).19.4, epist. 76.33, 95.72; cf. Ps.-Quint. decl. 2.18.20. 49

358

r. scott smith

The separation of the soul from the body is instantaneous (momentum at epist. 24.9; cf. 77.13), so much so that it is imperceptible (brevius est quam ut sentiri tanta velocitas possit: dial. 1 [= prov.].6.9). Although Seneca vehemently denies that anyone could know the nature of death (epist. 91.21), he was fascinated with the idea. With great admiration he records Julius Canus’s desire at the moment of death “to observe, at that swiftest of moments, whether the animus senses that it is departing” (dial. 9 [= tranq.].14.9). At epist. 30 he likewise reports eagerly listening to the Epicurean Bassus, in the throes of old age and with an infirm body, dissertating on death, speculating that it would come without sensation, for it was implausible that death, which led to non-sensation, would itself be sensible (30.5 f., 9, 13 f.). A reconstruction of Seneca’s view concerning the fate of the soul after its separation from the body is elusive, for Seneca nowhere provides us with a systematic account and often presents conflicting claims about the survival of the soul. René Hoven has categorized four distinct strains of thought concerning the fate of the soul after death in Seneca’s writings: 53 1) the soul survives for a limited time (orthodox Stoicism?); 2) death is either an end or a transition (the Socratic alternative); 3) death is non-existence (Epicurean); and 4) after death the soul strives to return to its divine origin and/or returns in new bodies (a sort of Pythagorean/Platonic mysticism). These differing claims are doubtless a function of Seneca’s reluctance to commit to a position on an unknowable subject that has little relevance (in his mind, at least) to ethical questions—compare his reluctance to commit to a locus for the animus noted above. It should be pointed out, if Seneca himself did not present a singular, confident view of the afterlife, neither did his Stoic predecessors, who were not in agreement concerning the soul’s survival after death. Diogenes Laertius (7.157 = SVF 2.811) records the differing opinions of Cleanthes, who argued that all souls survived until the ekpyrosis (the great conflagration), and of Chrysippus, who limited survival to the souls of the wise, presumably because of their stronger pneumatic tension. Another text (SVF 2.815) provides the detail that Chrysippus believed that after death the soul becomes spherical. The school’s founder, Zeno, also held that souls survived after separation from the body, calling the soul (SVF 1.146) “long-lived pneuma […] but not wholly immortal; for the soul is worn away into oblivion by the long span of time.” Unfortunately, a lacuna in the text makes it impossible to determine exactly what Zeno thought happens to the soul after the separation. The positions

Hoven 1971: 109–126. See also Motto 1955a, Setaioli 1997a.

physics i: body and soul

359

of later Stoics on the matter are even harder to determine. Panaetius seems to have argued for the mortality of the soul (frg. 83 van Straaten = Test. 120 Alesse; cf. Diog. Laert. 7.156 [= SVF 2.774], where individual souls are said to be “subject to destruction”), but whether this denies a limited survival of souls after the separation of body and soul is controversial.54 The position of Posidonius is even less clear, resting on one much disputed passage, Cicero, div. 1.64 (= frg. 108 Edelstein-Kidd). It was also once fashionable to claim that Posidonius advanced a highly mystical eschatology, which lies behind Cic. Tusc. 1, the Somnium Scipionis, Verg. Aen. 6, as well as Sen. cons. Marc. (dial. 6) and epist. 102.55 But in the absence of definitive evidence for Posidonius’s view it is more profitable to treat each of these texts on its own merits. In Seneca’s case, rhetorical aims and literary ambition may, and often do, take precedence over maintaining internally consistent views. Such is the case of cons. Marc.56 Just chapters after Seneca firmly claims that death is equivalent to non-existence (nihil est et omnia in nihilum redigit: 19.5; cf. epist. 54.4), he embarks on an eloquent and elaborate mystical depiction of a young son’s soul rising to meet his ancestors. This passage has a clear rhetorical objective, aimed at consoling Marcia at the passing of her young son, Metilius. It recalls the Somnium Scipionis and the Platonic concepts therein, but with some considerable divergences which render these conventional sentiments compatible with Stoicism. In the consolation Seneca asserts that Metilius’s early death is actually a boon since his short time on earth means that his soul will be less contaminated by contact with the body: “[those who die early] have only carried a little filth, a little burden with them” (23.1). Seneca later resumes this line of thought: Metilius is in a better station, everlasting, freed from his terrestrial bonds and the bodily stain that had accrued while he was alive (24.5–25.2);57 for a short time he will remain above us while he is cleansed of his mortal filth, after which he will be carried aloft to “run with the blessed 54 The vast majority of scholars incline to the view that Panaetius’s position was that the soul perished at death, but recently the case has been reopened: see Alesse 1997: 255f. for bibliography for and against. 55 For a thorough review of the scholarship until 1971, see Hoven 1971: 95–102; cf. Reinhardt 1921: 471–474. For a concise argument against Posidonian influence for the cons. Marc. see Manning 1981: 133–135; cf. Reinhardt 1921: 471–474, Favez 1928: xxxv–xlviii. See also Bocciolini Palagi 1979 on epist. 102. 56 One also sees the same conflict within the first two choral odes in Troades; the nihilism of the second ode, which espouses the Epicurean notion of non-existence, cannot be said to represent Seneca’s own thoughts on the matter. 57 Plat. Phaid. 66B–C, 80E–81C, 82C, 83D. At nat. 6.32.6 Seneca reports that upon death “a better and safer place awaits” the dead.

360

r. scott smith

souls (animas)” and be met by the sacred assembly composed of Scipios and Catos. The Platonic influences, especially from the Phaedo, are clear,58 but Seneca’s purgation is preparation for an extended afterlife and not for rebirth.59 Seneca has also molded his account to include the ekpyrosis, the great conflagration during which all elements are reduced to pure fire.60 Does this passage commit Seneca to anything more than literary ambition? Hardly. Nor can it be said that it represents his view of the afterlife. Seneca remains firmly non-committal to any particular view—betraying that uncertainty so famously articulated in the Socratic alternative found in the Apology (40C) and which is tersely summarized by Seneca himself: “What is death? Either an end or a transition” (mors quid est? aut finis aut transitus: epist. 65.24; cf. 93.10, dial. 1 [= prov.].6.6). Elsewhere he elaborates (epist. 24.18): “death either consumes us or frees us; if we are freed, the better part remains with its burden taken away; if consumed, nothing remains, and our benefits as well as our troubles have been removed.” It is Seneca’s wont to discuss the soul’s survival in purely hypothetical terms (e.g., either X or Y; if the soul survives, then Z).61 His main point concerning death is simply this: whether the soul survives or not, we are freed from the suffering that is attendant on our bodily existence.62 If it is productive to reduce Seneca’s belief in the afterlife, it may well be summed up as “a mixture of doubt and desire to believe.”63 But no more than that. Conclusion The evidence suggests the following conclusions: 1) Seneca does not systematically investigate the physiology of the human body or analyze the physical nature of the soul, refusing even to commit to a location for the animus. 2) He was nevertheless familiar with and followed the basic Stoic conception of the physical structure of body and soul (i.e., passive/active elements). 3) 58

Plat. Phaid. 113E–114A, Gorg. 524E ff., rep. 10.614A ff.; cf. Cic. rep. 6.29, Verg. Aen. 6.738–745. Occasionally Seneca toys with the idea of metempsychosis (epist. 102.24, 108.19ff.). 60 Some have ascribed such an eschatological view to Posidonius (e.g., Abel 1964), but this seems unnecessary given the context. It seems to me especially difficult to accept Posidonian influence here for two reasons: 1) at 23.2 Seneca cites Plato by name and clearly alludes to Phaid. 64A and 67D; 2) the reference to the Scipios (along with other echoes) clearly points to the strong literary influence the Somnium Scipionis had on Seneca’s consolation. If Posidonius is at work here (which is not at all clear), it is through the intermediary of the Ciceronian text. 61 Cf. epist. 57.7–9 (57.7 = SVF 2.820) and Berno 2006a ad loc. 62 Dial. 6 (= cons. Marc.).19.4f., benef. 7.1.7, dial. 11 (= cons. Pol.).9.2, epist. 99.29f., nat. 6.32.12. 63 Leeman 1951: 177. Cf. I. Hadot 1969: 91 with n. 74. 59

physics i: body and soul

361

Seneca accepts that there are different levels of spiritus/pneuma at work in the human organism (without exploring the physics of the relationship). 4) The “body” was considered to be the inert elements + spiritus (that held together the organism) and anima (that which provides life). 5) This creates a physical dualism between body (in the above sense) and animus, reflecting Seneca’s ethical interest in distinguishing purely natural/involuntary action from voluntary action, that is, “what is up to us.” 6) Platonic influences can be detected in Seneca’s degradation of the body and the elevation of the animus; whether this was part of the early Stoics’ view, was owed to an intermediary like Panaetius or Posidonius, or was Seneca’s own contribution cannot be ascertained. Finally, 7) appropriate actions are predicated on the physical constitution of a human organism; the fully rational adult is to be concerned with his rational, not biological life.64

64 I should like to thank Professors Stephen Brunet and Margaret Graver for many helpful suggestions and criticisms. Naturally, all remaining infelicities are mine and mine alone.

PHYSICS II: COSMOLOGY AND NATURAL PHILOSOPHY

Bardo Maria Gauly

Preliminary Remarks The ancient Stoics distinguished between three areas of philosophy: ethics, physics, and logic; the Greek terms τὸ ἠθικόν, τὸ φυσικόν, τὸ λογικόν (Diog. Laert. 7.39 = LS 26B) are rendered into Latin by Seneca as philosophia moralis, naturalis, and rationalis (epist. 89.9): Prima componit animum; secunda rerum naturam scrutatur; tertia proprietates verborum exigit et structuram et argumentationes, ne pro vero falsa subrepant. (“The first gets the soul into shape, the second inquires into the nature of the world, and the third clarifies the specific meaning of words and their linking into a line of argument; this will prevent false assumptions from insinuating themselves as true ones.”) Two alternative divisions quoted by Diogenes Laertius show that the Stoics subsumed under the term physics not only topics classed as natural philosophy today, but also questions we would assign to metaphysics. The first division, which Diogenes calls “the specific one” (εἰδικῶς), lists teachings about bodies, principles, elements, gods, boundaries, space, and the void; the second division, the generic one (γενικῶς), comprises doctrines of the world, the elements and the causes.1 In this article, I will first discuss natural philosophy in general, which deals with everything that exists or subsists (metaphysics), and then cosmology, which inquires into the existing world order. According to the Old Stoics, all three parts of philosophy should be considered an organic unity. The whole of philosophy is sometimes compared to a living creature, logic corresponding to the bones, ethics to the flesh, and physics to the soul. In another simile, philosophy is viewed as a fertile field, logic as surrounding fences, ethics as fruit, and physics as soil or

1 Diog. Laert. 7.132 = LS 43B. The only classification of physics in Seneca resembles the specific one of Diogenes (epist. 89.16): The first division is made between non-bodies and bodies; bodies are, in turn, divided into the creating force and the resulting elements. According to certain philosophers, as Seneca points out, the topic of the elements is divided into matter, the cause of all things, and the elements themselves.

364

bardo maria gauly

trees. These comparisons do not clarify which topic should be studied first or which is the most important, and Diogenes Laertius, who hands them down to us, informs us that there were different opinions within the school.2 According to Plutarch, Chrysippus not only claimed that the course of studies should begin with logic, continue with ethics, and end with physics, but also declared theology to be the last and holiest topic of physics and philosophy as a whole by calling it τελεταί (“initiatory mysteries”).3 As for the later Stoics Panaetius and Posidonius, physics came first within the curriculum (Diog. Laert. 7.41 = LS 26B), but only for Panaetius does this seem to imply that it was considered less important.4 When we come to the Romans, we are confronted with the almost unanimously held conviction that Roman philosophers in general and Seneca in particular confine themselves to teaching moral philosophy for practical purposes and that they neglect physics.5 But even if there is some truth in this, there is something to be said about the way in which Seneca expounds the worth of physics. For a Stoic, physics is closely connected to ethics because human action aims at living in accordance with nature. The meaning of this maxim is not self-evident, since it could relate either to the nature of the world or to man’s own nature. The Stoics, however, seeing man’s rational soul as part of or a reflection of the order of the cosmos did not seem to worry about this possible ambiguity; nor did Seneca.6 Despite this strong link between human action and nature, the formula of living in accordance with nature, which is often referred to in the Dialogi or the Epistulae morales,7 is never

2 Diog. Laert. 7.39–41 = LS 26B. A third comparison (ibid.) makes philosophy an egg, whose shell is logic, whose white is ethics, and whose yolk is physics. 3 Plut. mor. 1035A–B = SVF 2.42 = LS 26C. But Diogenes Laertius makes him put logic first, physics second, and ethics third (7.40 = LS 26B). 4 Rist 1969: 174f. (about Panaetius); Posidonius’s wide learning in the study of nature is at variance with the assumption that he could have had a low opinion of physics. 5 Donini calls it “one of the most successful fables convenues in Classical studies” (1988: 26). For example, one could cite Lapidge 1978: 184 f. 6 Grimal 1978c: 252f.; cf. Gill 2006: 150f. Both interpretations are to be found in his writings; cf. epist. 104.23: Gloriosum et excelsum spiritum [scil. natura dedit] quaerentem ubi honestissime, non ubi tutissime vivat, simillimum mundo, quem quantum mortalium passibus licet sequitur aemulaturque. (“Nature gave us a glorious and high spirit, which seeks to live as honestly, not as safely as possible, closely resembling the world, which it follows and emulates as far as its mortal steps allow.”) Dial. 7 (= vit. beat.).3.3: Beata est ergo vita conveniens naturae suae. (“Happy is, therefore, life according to its own nature.”) For the Old Stoics, see Diog. Laert. 7.87–89 = LS 63C. 7 E.g., Sen. dial. 7 (= vit. beat.).8.1 f., epist. 5.4, 41.8, 45.9.

physics ii: cosmology and natural philosophy

365

mentioned in his only work on natural philosophy, the Naturales quaestiones.8 But even if the nature of their relationship is never discussed at length, both physics and ethics are highly esteemed by Seneca, in particular when compared to logic, which is discarded as a waste of time (e.g., epist. 49.5). Their relative values are not easy to determine, since there are different estimations within Seneca’s writings. Even in the works on moral philosophy there are passages that show great respect for cosmological speculation (epist. 95.10): Philosophia autem et contemplativa est et activa: spectat simul agitque. Erras enim si tibi illam putas tantum terrestres operas promittere: altius spirat. ‘Totum’ inquit ‘mundum scrutor nec me intra contubernium mortale contineo, suadere vobis aut dissuadere contenta: magna me vocant supraque vos posita.’ (“Philosophy is theoretical and practical; at the same time it is watching and taking action. For you are wrong if you think that it promises only earthly work; its spirit aims higher. ‘I am inquiring’, it says, ‘into the whole world and I am not confining myself to living with man, content to give you advice or warning: Great things soaring above you are inviting me.’”)9 This high esteem is regularly connected with cosmology (not physics in general), and, more specifically, with contemplation of the heavens and the stars, with reflection on the divine realms and their secrets.10 By contrast, when discussing the Stoic doctrine of the corporality of the virtues, Seneca ends by disavowing such musing: As with the study of logic, the preoccupation with the niceties of the physical principles is said to be worthless inasmuch as it does not help the man who strives for virtue.11

8 There are, to be sure, epilogues or digressions giving moral advice, but these are loosely attached to the treatises on natural phenomena; for this problem, see Gauly 2004: 87–134. 9 Cf. Althoff 2005: 18: “Der ethische Nutzen der Philosophie erscheint hier also erstaunlicherweise abgewertet gegenüber der reinen physikalischen Forschung.” 10 Cf. Sen. nat. 1 pr. 1–4, where only two areas of philosophy are contrasted with each other, moral philosophy and theology, which is then conceived as cosmology. A higher value is put on theology (§1): Altior est haec et animosior (“This is nobler and bolder.”) The questions that theology seeks to answer are these (§3): quae universi materia sit, quis auctor aut custos, quid sit deus, totus in se tendat an et ad nos aliquando respiciat, faciat cotidie aliquid an semel fecerit, pars mundi sit an mundus […]. (“What is the matter of the universe, who is its creator or guard, what is god, is he confining his attention to himself or does he sometimes take care of us, is he doing something daily or has he completed his work once and for all, is he part of the world or the world itself?”) It is reflection on these problems that makes life worth living (§ 4):Nisi ad haec admitterer, non fuerat operae pretium nasci. The identification of cosmology and theology recalls Chrysippus’s words on theology being the culmination of physics (Plut.mor. 1035A–B = SVF 2.42 = LS 26C). 11 It is compared to playing a game (epist. 106.11): Latrunculis ludimus. In supervacuis subtilitas teritur: non faciunt bonos ista sed doctos. (“We are playing board games. Subtlety is wasted on unnecessary things that do not render you good, but learned.”)

366

bardo maria gauly

Technical discussions about details of terminology or categorization are disapproved of, contemplation of the marvels and order of the world are viewed as edifying. One last preliminary remark: Seneca has not written a systematic discourse on natural philosophy as a whole. The Naturales quaestiones, the only work that deals with topics from physics, discusses only some portions of cosmology. The Stoic doctrines of the principles of physics are never presented in an orderly way. This does not mean that Seneca did not know them, but that he was more concerned with understanding the world as it exists and with guiding man to virtue. Therefore, if we want to reconstruct the way Seneca thinks of physics and cosmology, we often have to examine passages dispersed throughout his works. Even so, it is difficult to get a clear picture of his ideas. Physics This is true in particular with regard to the basic doctrines of physics; there is next to nothing in Seneca’s writing about crucial points of Stoic metaphysics as the theory of principles. Only when discussing Platonic and Peripatetic tenets of causality does Seneca comment on cause as a Stoic principle (epist. 65.2 = LS 55E = SVF 2.303): Dicunt, ut scis, Stoici nostri duo esse in rerum natura ex quibus omnia fiant, causam et materiam. Materia iacet iners, res ad omnia parata, cessatura si nemo moveat; causa autem, id est ratio, materiam format et quocumque vult versat, ex illa varia opera producit. Esse ergo debet unde fiat aliquid, deinde a quo fiat: hoc causa est, illud materia. (“Our Stoics say, as you know, that there are two things in the world through which everything comes into existence, cause, and matter. Matter lies inert, ready for everything, but resting as long as nobody puts it in motion; cause, however, i.e., reason, forms matter and turns it into whatever it likes, thereby producing a variety of creations. These are thus inevitably generated from something and by something; the latter is cause, the former matter.”) The relation of matter to cause is then explained by a comparison; since human art imitates nature, the bronze that is formed into a statue corresponds to matter, the artist who shapes it corresponds to cause (epist. 65.3). This is a correct, if sketchy account of the Stoic principles. As for matter, the passive principle, Seneca’s translation of the Greek adjective ἄποιος (e.g., Diog. Laert. 7.134 = LS 44B), which has often been understood to signify “unqualified” (e.g., by LS), suggests a different meaning, viz. “inert,” “not

physics ii: cosmology and natural philosophy

367

active.”12 Neither here nor elsewhere does Seneca seem to be interested in elucidating the notion of matter more thoroughly.13 He has far more to say about the active principle, which he calls “cause” or “reason” in epist. 65.2. The body of the letter argues first against Aristotle’s four types of causes, then against Plato, who is said to “add” a fifth type, the idea.14 Since matter is simple, Seneca continues, cause must be simple, too; thus he reduces the Academic and Peripatetic variety of causes to one cause only, the efficient cause, which he defines as “reason” or “God.”15 The identification of the active principle with God is repeated near the end of the letter in order to teach a moral lesson (epist. 65.23f.), the subordination of the human body to the soul (§23): Potentius autem est ac pretiosius quod facit, quod est deus, quam materia patiens dei. (“But more powerful and valuable is that which acts, viz. God, than matter, which is acted upon by God.”) One passage of the Naturales quaestiones goes further, giving a long list of terms that are supposed to be synonymous: Iuppiter is called not only rector custosque universi, but also causa causarum, which makes him identical with the active principle as defined in epist. 65.16 The Stoic doctrine of the corporality of the principles provides the premise of such an identification. Both reason and matter are bodies, because only bodies are able to act or to be acted upon.17 But since both principles are theoretically inseparable it seems more accurate to speak of two aspects of one substance (Lapidge 1978: 163 f.). Seneca never mentions the tenet of the corporality of the principles, but when he comes to discuss the question as to 12 Todd 1978: 140f. For a discussion of the Aristotelian origin of the Stoic doctrine of principles, see Scarpat’s commentary on Seneca’s letter (1965: 93–95). 13 As Chalcidius does in his version of Zeno’s doctrine (comm. in Plat. Tim. 292 = LS 44D): Matter is the substratum of everything existing; it is finite and subject to change, and it is neither generated nor bound to perish; cf. Diog. Laert. 7.134 = LS 44B. For another concept of matter in Seneca, see nat. 2.3f. (matter opposed to “parts” of nature) and Hine’s commentary: “perhaps […] a debased descendant of the Aristotelian and Stoic doctrine” (1981: 164). 14 Epist. 65.4–14. The surprising identification of Aristotle’s second cause with the opifex (artist), i.e., the δηµιουργός, seems to be due to the following comparison with an artist (Dörrie and Baltes 1996: 415 f.). Epist. 65 has, together with epist. 58, aroused a lively debate about its sources and its Platonic affiliations; see Bickel 1960, Dillon 1977: 135–139, Donini 1979: 151–208, Setaioli 1988: 126–140, Chaumartin 1993b, Isnardi Parente 1995, Küppers 1996, and Gauly 2004: 164–170. 15 Epist. 65.12: Quaerimus quid sit causa? Ratio scilicet faciens, id est deus. (“If we want to know what cause is, the answer is: reason, viz. God.”) 16 Nat. 2.45; arguing against the false assumption of the punishing god, Seneca creates the idea of Iuppiter as ruler of the universe, spirit of the world, fate, providence, first cause, and nature itself. 17 Diog. Laert. 7.134 = LS 44B (si vera lectio), Cic. ac. 1.39 = LS 45A, Aristocl. ap. Eus. Pr. Ev. 15.14.1 = LS 45G.

368

bardo maria gauly

whether good is corporeal, he quotes the teachings of anonymous Stoics that everything active is a body, a reasoning that he dismisses as useless (epist. 106.3–12). Seneca nowhere defines what a body is, but he shares the Stoic criterion for recognizing a body, viz. the ability to produce or feel an effect (epist. 106.3 and 106.8).18 Since everything existing is a body (including air), the movement of bodies requires an explanation; while the Epicureans taught the movement of atoms in the void, the Stoics developed the doctrine of ἀντιπερίστασις, which resembles the modern theory of displacement. Arguing for the unity of the air, Seneca explains this (nat. 2.7.2) by referring to fluids quae sic corpora accipiunt ut semper in contrarium acceptis refluant (“that take up bodies and replace them by constantly flowing back in the opposite direction”). Accordingly, there is no void within the existing world, but there is or may be one outside of the world.19 The void, which is beyond the cosmos, forms one of four ontological classes of the incorporeal. The others are the λεκτόν (“the sayable”), space, and time (LS 27D). These are to be considered as subsistent, not existent.20 In epist. 58.8–14, however, Seneca gives an account of ontology in which he contrasts bodies and the incorporeal before he subdivides the bodies into living and non-living beings; the living ones are in turn divided into animals and plants, and so on. By subsuming all these (including the incorporeal) under the highest class, quod est (Seneca’s translation of the Greek τὸ ὄν), Seneca diverges from Stoic orthodoxy, which would contrast the incorporeal and the existing, classifying both under the heading of τί (quid, “something”).21 We are therefore bound to distinguish two Stoic concepts of the world, first, the existing cosmos, and, second, the universe, which comprises the cosmos and the void around it.22 The cosmos is a rationally structured unity. This unity is assured by a sort of warming and invigorating power, which Zeno and Cleanthes identified as fire, more precisely as πῦρ τεχνικόν 18 In epist. 102.6 several classes of bodies are distinguished: continua corpora (like man), composita (like a ship), and ex distantibus (like an army); for these types and for different connections between bodies (as in nat. 2.2.1–4), see Wildberger 2006: 7–11. 19 Dial. 8 (= de otio).5.6: ‘Illud’ inquit ‘scrutor quod ultra mundum iacet, utrumne profunda vastitas sit an et hoc ipsum terminis suis cludatur.’ (“‘I explore,’ says he [man following his rational nature], ‘what lies beyond the world, asking if it is deep void or if even this space is confined by its own limits.’ ”) Cf. Lapidge 1978: 177, Wildberger 2006: 100–102. 20 For this distinction, see LS 1.163 f. 21 In epist. 58.15 Seneca speaks of “certain Stoics,” who give another definition of quid that comprises existing and imaginative entities. Why Seneca has conceived a different ontological scheme, is open to discussion; see Wildberger 2006: 94–99. 22 Wildberger 2006: 3; cf. Furley 1999: 412.

physics ii: cosmology and natural philosophy

369

(LS 47C = Cic. nat. 2.23–30).23 Chrysippus modified the theory by establishing πνεῦµα, consisting of fire and air, as the animating principle. This spirit permeates all bodies and gives them their inner tension and unity.24 It is the active principle, God, which—or who—operates in and by means of the πνεῦµα. Seneca follows Chryrippus, rendering πνεῦµα as spiritus (“spirit”). When he is pondering the question to whom man owes the beauties of nature (dial. 12 [= cons. Helv.].8.3), he mentions, in addition to God, reason or fate the divinus spiritus per omnia maxima ac minima aequali intentione diffusus (“the divine spirit that permeates everything great and small with even tension”). He is more explicit in the Naturales quaestiones, where he is pleading the unity of the air against the atomists. Having cited several phenomena (like the production of sounds by the tongue), which can only be explained by the theory of tension (2.6.3–5), he arrives at the effects that the tension of air has within individual bodies (2.6.6): Esse autem unitatem in aëre vel ex hoc intellegi potest, quod corpora nostra inter se cohaerent. quid enim esset aliud quod teneret illa quam spiritus? quid est aliud quo animus noster agitetur? quis esset illi motus nisi intentio? quae intentio nisi ex unitate? quae unitas nisi haec esset in aëre? quid autem aliud producit fruges et segetem inbecillam ac virentem erigit, arbores aut distendit in ramos aut in altum exigit, quam spiritus intentio et unitas? But that there is unity in the air may be understood even by the inner coherence of our bodies. For what else could it be that keeps them together if not the spirit? What else is it that makes our soul move? How could it move if not by tension? What tension could be there if not by unity? What unity could be there if not in the air? And what else makes fruit grow; what raises weak and green seed; what drives trees to spread their branches and to rise high, if not tension and unity of the spirit?25

The πνεῦµα, however, acts not only in every single body of the world, but also gives unity to the cosmos altogether.

23 For the relationship between the “creative fire” and the element see Lapidge 1978: 167, Furley 1999: 440 f., Wildberger 2006: 75–78. 24 The quality that every individual body receives from the spirit is called ἕξις (“tenor”); beyond this, plants are said to have φύσις, animals ψυχή (LS 47P). The same distinction is found in Sen. nat. 6.16.1. 25 Unitas seems to be equivalent to ἕξις (Wildberger 2006: 210 f.). Inquiring into Seneca’s terminology with reference to air, Bravo Díaz (1991) has noticed that there is ambivalence in his use of terms like aer and spiritus insofar as this usage is, despite his knowledge of the concepts of Stoic philosophy, not always technically strict.

370

bardo maria gauly

The whole world thus forms a single body, so that the individual bodies within form parts of a larger unit, a tenet that is the foundation of the Stoic doctrine of συµπάθεια. According to this, every change, every movement within the cosmos has effects on the rest of it (Wildberger 2006: 16–20). The world is a large organism, and for this cosmic body God, as the active and reasonable principle, performs a function analogous to the one the soul fulfils for man. The human soul in turn is considered to be part or offspring of divine reason, and Seneca attaches great importance to this kinship between man and God, as in epist. 92.30: Quid est autem cur non existimes in eo divini aliquid existere qui dei pars est? Totum hoc quo continemur et unum est et deus; et socii sumus eius et membra. (“Why would you not think that there is something divine in us who are part of God? This whole world around us is one and is God; we are both his companions and his parts.”)26 The question as to how man’s position within the world is to be established is also vital for Seneca’s cosmological reasoning. Cosmology Whereas the world on the whole, according to the Stoics, will last forever, the existing world order, which is sometimes called διακόσµησις, to distinguish it from the eternal κόσµος, is bound to dissolve into pure fire.27 This ἐκπύρωσις (conflagration), which is repeated at certain intervals, is not conceived as the destruction of the world, but as a reconstitution of the best possible state of the world, since all individual bodies are thereby transformed into divine fire. When the ἐκπύρωσις comes to an end, air is condensed to moisture bearing the seed from which will spring the new world.28 What Seneca has to say about cosmogony, about the world’s dissolution in fire, and about the cosmic cycle of conflagration and renewal shows that he is well aware of the Stoic doctrines. Nevertheless, the various statements referring to these issues, which are spread throughout his works, do not give a coherent picture of his ideas.

26 Cf. dial. 1 (= prov.).1.5, 8 (= de otio).5.5, 12 (= cons. Helv.).6.6–8, 12.11.6 f., epist. 41.1 f., 66.12, nat. 1 pr. 14. 27 For the term διακόσµησις, see SVF 1.98 (= LS 46G) and 1.107. Seneca uses mundus both for κόσµος and for διακόσµησις (Wildberger 2006: 478, n. 93). 28 Evidence for the Stoic tenets concerning conflagration and cosmic cycles can be found in LS 46 and 52.

physics ii: cosmology and natural philosophy

371

To start with, cosmogony is mentioned as a subject of man’s scientific curiosity in De otio (dial. 8.5.5), where there are allusions to the undifferentiated status of the universe after conflagration, to divine reason acting upon matter to create and order separate entities, and finally to the divine origin of man’s soul. The moment when the old order has dissolved and the new one has not yet begun to develop is captured in a beautiful picture in epist. 9: Speaking of the pure reason of the sapiens, which does not depend upon external circumstances, Seneca imagines (epist. 9.16) god (Iuppiter) cum resoluto mundo et dis in unum confusis paulisper cessante natura adquiescit sibi cogitationibus suis traditus (“reposing after the world has dissolved and the gods have merged and enjoying himself, engrossed in his thoughts, as nature is at rest for a while”). A more technical account is given in the Naturales quaestiones: Fire transforms the world “in itself”; after everything is consumed, fire gives way to moisture, yielding hope for a new cosmos.29 At the end of the Consolatio ad Marciam, the deceased Cremutius Cordus, speaking in a προσωποποιία, uses the idea of everything mortal being bound to perish and of the world’s periodic dissolution in fire leading to regeneration to give comfort to his daughter (dial. 6.26.6f.). After reminding her of natural disasters like earthquakes, floods, and conflagrations, which befall various parts of the world, thus indicating universal transience, he tries to console her by pointing out that everything that exists, including the souls of the deceased, will at some point dissolve in fire; comfort can be found in the thought that this conflagration does not end everything, but results in transformation and the beginning of a new world.30 There are other passages that do not seem to be reconciled so easily with the Stoic tenet that the ἐκπύρωσις entails renewal and regeneration. In the Consolatio ad Polybium, it is viewed as destruction and the return of chaos.31 A more elaborate picture of devastation is presented in the treatise on water 29 Nat. 3.13.1: Dicimus enim ignem esse qui occupet mundum et in se cuncta convertat: hunc evanidum languentemque considere et nihil relinqui aliud in rerum natura igne restincto quam umorem; in hoc futuri mundi spem latere. 30 The phrase in antiqua elementa vertemur (dial. 6 [= cons. Marc.].26.7): “we will be transformed into the old elements”) is, strictly speaking, not a correct statement of Stoic doctrine, but the general sense cannot be misunderstood. The notion of everlasting recurrence is alluded to when Cremutius talks of God deciding iterum ista moliri (ibid.: “to set in motion this process anew”). 31 Dial. 11 (= cons. Pol.).1.2: Mundo quidam minantur interitum et hoc universum quod omnia divina humanaque complectitur, si fas putas credere, dies aliquis dissipabit et in confusionem veterem tenebrasque demerget. (“Some threaten the world with doom, and if you are willing to believe, some day will destroy this universe that comprises everything divine and human and will make it plunge into the old chaos and darkness.”)

372

bardo maria gauly

in the Naturales quaestiones (3.27–30). Here, the disaster that will strike the world at a time determined by fate is a flood that will annihilate a mankind spoiled by vice. It will not lead to complete dissolution of the cosmos because the water, having destroyed mankind and earth, will ebb and make way for a new generation—that is again doomed to corruption. How this scenario is to be reconciled with Stoic ἐκπύρωσις and thus with Stoic orthodoxy is subject to debate.32 Suffice it to say that there are analogies (the periodic recurrence, the imposition by fate)33 as well as fundamental discrepancies (the catastrophe as punishment for wrongdoing).34 Wherever the existing world order is under consideration, it appears as a well-structured whole, rational and beneficial to man’s best interests. Anthropocentrism is the key feature of this conception; the world is visualized as a city to live in, as the common home of God and man (dial. 6 [= cons. Marc.].18.2),35 or as a body, an organic unity, ruled by divine reason, of which man forms part.36 Seneca speaks of the cosmos in its entirety in the context of carefully conceived images that depict man’s position within the beautiful order of the world so as to give encouragement or comfort. In the Consolatio ad Marciam Nature herself is said to inform the mourning mother about the condicio humana. A fictitious speech, delivered by Nature at the moment of man’s birth (dial. 6 [= cons. Marc.].17.6–18.8), gives an overall picture of the world, only to leave it to man to decide whether he wants to enter life (dial. 6.18.2): Intraturus es urbem dis hominibus communem, omnia complexam, certis legibus aeternisque devinctam, indefatigata caelestium officia volventem. Videbis illic innumerabiles stellas micare […]. (“You are about to enter the common city of gods and men, which comprises everything; it is subject to certain eternal laws and untiringly does its duties to the heavens. You will see there countless stars twinkling […].”) Having laid out the celestial phenomena, sun, moon, and the five planets, Nature proceeds to point out the miracles of the atmosphere and the earth. The variety and beauty of terrestrial marvels are contrasted with the dire consequences man must face

Cf. Gauly 2004: 235–266, Kullmann 2005: 142 f., Wildberger 2006: 56–58. In nat. 3.29.2 there is an explicit comparison between flood and conflagratio. 34 Despite the fact that there are some late testimonies that call the conflagration κάθαρσις (SVF 2.598), the Stoic ἐκπύρωσις does not seem to have performed a moral function (Gauly 2004: 247–253). 35 For Stoic cosmopolitism, see dial. 8 (= de otio).4 f., 9 (= tranq.).4.4, epist. 28.4, 102.21. 36 Epist. 95.52: Omne hoc quod vides, quo divina atque humana conclusa sunt, unum est; membra sumus corporis magni (“Everything you see, which comprises the divine and the human sphere, is one; we are part of a large body.”). See 92.30. For the unity of the cosmos and divine reason as principal “prior commitment” of Stoic physics, see White 2003: 127f. 33

physics ii: cosmology and natural philosophy

373

if he uses his abilities badly: diseases, dangers, and death then await him. But even so, Nature says in conclusion, man will opt for life if he is confronted with choice. By doing so, he will rationally accept natural law.37 The law of nature is here seen first and foremost as the law of mortality;38 man is destined to live not only on earth but also within the cosmos as a whole; all three spheres of the world are subjects of his observation and contemplation, his amazement and awe. In the Consolatio ad Helviam, the motif of contemplatio caeli, which makes life worth living, is more fully presented (dial. 12 [= cons. Helv.].8.2–6).39 Wherever he lives, man is able to direct his gaze to the heavens and to reflect upon his soul’s kinship to the divine sphere. The existing world is seen as the best possible world, the human soul as its most magnificent part inasmuch as it is contemplator admiratorque mundi (“the viewer and admirer of the world”).40 Man’s position within the cosmos is the prerequisite of this contemplation. Earth rests unmoving in the middle of the world (epist. 93.9) and it is from this center that man is able to watch the celestial sphere and to comprehend its order. The triple classification of the cosmos into a celestial, a meteorological, and a terrestrial sphere, which is hinted at in the passage cited from Ad Marciam (dial. 6.17 f.), is developed at the very beginning of the second book of the Naturales quaestiones, where cosmology is divided into astronomy,

37 Dial. 6.18.8: Respondebis velle te vivere. Quidni? immo, puto, ad id non accedes ex quo tibi aliquid decuti doles! Vive ergo ut convenit. (“You will answer that you wish to live. Of course, you will. But I am certain you will not enter a life that entails loss and dolor. So live under the agreement.”) Inwood has pointed out that the reasoning in this passage is strongly influenced by the Socratic example in Plato’s Crito (2005: 240–248). 38 Inwood 2005a: 240–248; for the notion of natural law in the Naturales quaestiones, see Kullmann 1995: 72–76 and Kullmann 2005. 39 For this motif see Küppers 1996 and Pfeiffer 2001. The idea that insight into the divine nature is the final aim of human existence is expressed in the praefatio to the first book of the Naturales quaestiones (nat. 1 pr. 4): Nisi ad haec admitterer, non fuerat operae pretium nasci. (“If I were not to gain access to this, it would not have been worthwhile being born.”) For this passage see Gauly 2004: 165 n. 126. 40 Dial. 12.8.4; cf. 12.8.6: Proinde, dum oculi mei ab illo spectaculo cuius insatiabiles sunt non abducantur, dum mihi solem lunamque intueri liceat, dum ceteris inhaerere sideribus, dum ortus eorum occasusque et intervalla et causas investigare vel ocius meandi vel tardius […], dum cum his sim et caelestibus, qua homini fas est, inmiscear, dum animum ad cognatarum rerum conspectum tendentem in sublimi semper habeam, quantum refert mea quid calcem? (“Therefore, as long as my eyes can view that spectacle for which they are insatiable, as long as I may look at the sun and the moon, behold the other stars, inquire into their rise, their setting and their distances, and find out why they move faster or more slowly […], as long as I am engaged with these and mingle as far as divine law allows with the celestials, as long as I direct my soul to the aspect of kindred beings and to heavens above, what does it bother me where I am walking?”)

374

bardo maria gauly

meteorology, and geography.41 The first of these deals with the form and matter of the heavens and the movement of the stars (nat. 2.1.1.), the second enquires into whatever lies between the heavens and earth, including not only meteorological phenomena like wind and rain but also earthquakes. Finally, water, land, and flora are defined as issues of geography (nat. 2.1.2). Seneca makes two specifications in his classification of cosmology; first, the earth is a subject not only of geography but also of astronomy as far as its position within the cosmos is of interest (nat. 2.1.4f.). Second, the topic of earthquakes is assigned to meteorology, since their cause, air, belongs to the atmosphere (nat. 2.1.3). Combining different spheres of the cosmos with different explanatory principles, Seneca’s structuring resembles Aristotle’s method at the beginning of the Meteorologika, in which he blurs the boundary between the meteorological and the terrestrial sphere.42 The earth rests unmoving in the center of the cosmos (dial. 1 [= prov.].1.2; epist. 93.9); it is spherical in shape (nat. 4b.11.2–4), and is a huge mass (dial. 1.1.2), but is of modest proportions if compared to the dimensions of the world (dial. 6 [= cons. Marc.].21.2).43 By assigning earth a central position, Seneca follows an old Stoic tradition (e.g., Cleanthes, SVF 1.537.7f. = LS 54I), but with one exception. In his treatise on comets, the idea of geocentrism seems to be challenged. Even if the phrasing does not exclude all possibility of doubt, the text suggests that the theory of geocentrism is a subject for discussion.44 More often than not, however, Seneca adheres to the traditional view of his

41 Nat. 2.1.1: Omnis de universo quaestio in caelestia sublimia terrena dividitur. (“The entire study of the universe is divided into research about celestial, meteorological, and terrestrial phenomena.”) 42 Aristot. meteor. 1.1; cf. Hine 1981: 124–127. 43 Cf. nat. 1 pr. 8–11, 4b.11.4; in all three instances it is called “a point” (punctum) within the universe; for this metaphor, see Gauly 2004: 181–186. 44 Nat. 7.2.3: Illo quoque pertinebit haec excussisse, ut sciamus utrum mundus terra stante circumeat an mundo stante terra vertatur. fuerunt enim qui dicerent nos esse quos rerum natura nescientes ferat, nec caeli motu fieri ortus et occasus, nos ipsos oriri et occidere. digna res contemplatione, ut sciamus in quo rerum statu simus, pigerrimam sortiti an velocissimam sedem, circa nos deus omnia an nos agat. (“It is also of importance to settle this question if we want to know whether the world is rotating while the earth is standing still or the earth is spinning while the world is standing. For there were those who said that we are moved by nature without noting it, and that it is not the movement of the sky that makes the sun rise and set, but that we ourselves rise and set. This question merits consideration so that we know what is our position within the world, whether we have been assigned a settled or a fast-moving dwelling, whether God drives everything around us or drives us.”) Despite the objections raised by Wildberger (2006: 490f., n. 127), the text seems to speak not about the earth rotating on its own axis; neither the wording oriri et occidere nor the last sentence seems to accord with such an understanding. See Gauly 2004: 188f.

physics ii: cosmology and natural philosophy

375

school, which has been rightly termed a Stoic “article de foi” (Aujac 1989: 1437). This view is connected with a belief in the anthropocentrism of the cosmos. Man is at the center of the world, bound to look up to heaven, where his soul comes from,45 while providence supplies him with everything he needs.46 Earth nourishes and animates the plants that grow on it and it is able to do so because it is itself animate (nat. 6.16.1): Non esse terram sine spiritu palam est: non tantum illo dico quo se tenet ac partes sui iungit, qui inest etiam saxis mortuisque corporibus, sed illo dico vitali et vegeto et alente omnia. (“It is obvious that the earth is not without air, and thereby I do not mean only the air that ensures coherence and interconnects its parts, which even rocks and dead bodies have, but that animating, lively and nourishing air.”) 47 The analogy between earth and a living organism is crucial for the explanation of various natural phenomena in the Naturales quaestiones, but Seneca seems well aware of the limits of analogy and of the problems that can result from metaphoric language.48 Since Aristotle, the second sphere of the cosmos, which comprises what lies between earth and moon, had been called τὰ µετέωρα (the meteorological region), but there had been no Latin equivalent. Referring to the atmosphere, Seneca sometimes uses aer, which in its proper sense signifies the element air, but he applies sublimia as well, a term he has coined to match the Greek τὰ µετέωρα.49 The air within the meteorological region is influenced by the respective adjacent spheres; it is dry, hot, and thin where it comes close to the celestial region; it is dense and misty where it receives the evaporations from the earth (nat. 2.10.2, cf. nat. 4b.10). The individual phenomena that are assigned to the meteorological region in the Naturales quaestiones (clouds and rain, wind, earthquakes, halos and similar phenomena, thunderstorms) seem to be random, although they are subject to reason and natural law (dial. 1 [= prov.].1.3): Ne illa quidem quae videntur confusa et incerta, pluvias dico nubesque et elisorum fulminum iactus et incendia ruptis montium verticibus 45 In De otio, man’s position at the center of the world is interpreted as proof of his destination for contemplatio rerum (dial. 8.5). For the divine origin of his soul, see ibid. 8.5.5. 46 The final chapter of the treatise on winds (nat. 5) presents the winds as providentiae opera (5.18.1: “works of providence”), as a benefit man receives from God, since it is due to the winds that the climate is temperate and fruits can ripen (5.18.13). 47 For the terms Seneca uses for life, see Wildberger 2006: 211. 48 Althoff 1997, who among others cites nat. 5.4.2 as an example (p. 104): Speaking about the origin of winds Seneca ridicules the analogy between evaporations from the earth and flatulence. 49 For Aristotle’s definition of meteorology, see meteor. 1.1; for the Senecan terms, see Hine 1981: 123–127, Bravo Díaz 1995.

376

bardo maria gauly

effusa, tremores labantis soli aliaque quae tumultuosa pars rerum circa terras movet, sine ratione, quamvis subita sint, accidunt, sed suas et illa causas habent […]. (“Not even those phenomena that seem to be random and accidental, viz. rain, clouds, the flashes and strokes of lightning, fires sent by the eruption of volcanic peaks, the shakings of an earthquake and other perturbations resulting from the turbulent region around the earth, occur without reason despite their suddenness, but have their own causes […].”)50 Thus, these phenomena command man’s admiration no less than the miracles of the celestial region and the marvels of the earth (dial. 6 [= cons. Marc.].18.3f.). The phenomena of the atmosphere are only once described as fortuitous. This is when, in the relevant treatise of the Naturales quaestiones, the comets are considered to be perennial celestial bodies and not the transient fiery phenomena of established tradition. They are seen in contrast to the random phenomena of the meteorological region, the argument in favor of their celestial status being their beauty.51 The atmosphere is seen as less fair, less pure, less bright than the heavens; but it connects the earth to the heavens (epist. 102.21): Aer humanis divina secernens etiam coniungit. (“The atmosphere not only separates man’s dwellings from the divine realms, but joins them.”)52 Not infrequently, it seems, Seneca views natural phenomena in a metaphorical way. This is particularly true for the highest sphere of the cosmos, the heavens, which are presented as realms of God, of pure reason, and of beautiful order (nat. 2.13.3 f.): Nihil enim illic [scil. in aethere] iniuria cogitur, nihil rumpitur, nihil praeter solitum evenit; ordo rerum est, et expurgatus ignis in custodia mundi summa sortitus oras operis pulcherrimi circumit. (“For nothing is there [viz. in the aether] compressed by force, nothing breaks, nothing uncommon happens. Everything is in good order there, and the purified fire, which has been allotted custody of the world’s highest region, runs around the edges of the finest creation.”) The passage quoted from the treatise on thunderstorms argues against the assumption that fire from the aether could descend into clouds and cause lightning; the celestial sphere itself is dealt with neither in the Naturales

50 Cf. nat. 1 pr. 14; for Seneca’s view about natural law, see Kullmann 1995: 72–76 and Kullmann 2005. 51 Nat. 7.27.6: Quorum formosior facies est quam ut fortuitam putes. (“Their appearance is too beautiful to be considered fortuitous.”) Cf. Gauly 2004: 157f. 52 Cf. nat. 4b.10: Editior aer, quo longius a terrarum colluvie recessit, hoc sincerior puriorque est. (“The greater the distance between the air in the height and the mud of the earth, the clearer and purer it is.”)

physics ii: cosmology and natural philosophy

377

quaestiones nor elsewhere in a coherent manner.53 Of the three topics of cosmology referred to in nat. 2.1.1., only the caelestia (“astronomy”) are not—with the exception of the comets—a subject of the work. As with his treatment of atmosphere, Seneca’s terminology is far from precise. Besides the term caelum (“sky, heavens”), Seneca uses aether as well as ignis (“fire”) to signify the highest sphere of the world. 54 There are, then, only incidental remarks concerning astronomy; when speaking of the rainbow, the sun is adduced as evidence for the shortcomings of man’s sensory perception, since it is far bigger than it appears to be and it moves far more quickly than it seems to (nat. 1.3.10). The comets, subject of the seventh book of the Naturales quaestiones, are the sole exception. While Anaxagoras and Democritus declared them to be an optical illusion and Aristotle regarded them as fiery phenomena of the atmosphere, Seneca favors the theory that they are perennial celestial bodies moving in regular, if unknown, orbits. I believe the reason why Seneca has discarded the established ancient ideas on comets and argues for a theory that comes close to the modern one is to be found in a metaphorical view of the phenomena. The comets are considered to be signs of the beautiful order of the heavens.55 Where the celestial region is spoken of, it is presented as a divine realm, as a sphere to which man’s sensory perception has limited access, as the origin of the human soul, where it longs to return to (epist. 102.23f.). As long as man is tied to his earthly existence, he is bound to reach for the heavens by means of his intelligent soul and to admire and worship the

53 Gross (1989: 318–320) has suggested that two books on the phenomena of the heavens have been lost; his hypothesis rests on Hine’s finding that in the Archetypus book IVb was numbered as the third book (1981: 4–6) and on a notice by Cassiodorus (inst. 2.6.4, frg. 13 Haase; cf. Lausberg 1989: 1928 f.) mentioning a lost work of Seneca’s entitled De forma mundi (“On the form of the world”). For different conclusions on book numbering and book order in the Naturales quaestiones, see Gauly 2004: 53–67. 54 Hine 1981: 123, Bravo Díaz 1995: 10–25. 55 See in particular nat. 7.27.6: Cometas non frequenter ostendit [scil. natura], attribuit illis alium locum, alia tempora, dissimiles ceteris motus: voluit et his magnitudinem operis sui colere. quorum formosior facies est quam ut fortuitam putes, sive amplitudinem eorum consideres sive fulgorem, qui maior est ardentiorque quam ceteris. facies vero habet insigne quiddam et singulare, non in angustum coniecta et artata, sed dimissa liberius et multarum stellarum amplexa regionem. (“Nature does not often show us comets; she has assigned them a different space, different times, dissimilar movements; she wanted to celebrate her great work also in the comets. Their appearance is too beautiful to be considered fortuitous, if you look at their dimensions or their splendour, which is more sublime and brighter than any other. Not least their form has something special and unique, since it is not narrowly limited and confined, but spreads out widely and comprises the region of many stars.”) For Seneca’s theory of comets, see Gauly 2004: 143–164.

378

bardo maria gauly

divine majesty (dial. 8 [= de otio].5.5–8; nat. 1 pr. 7–13). Having been relieved of its body and its impairments, the human soul will rise and look down upon the limitations of life on earth (dial. 6 [= cons. Marc.].24f.). Seneca’s cosmology first and foremost deals with man, his origin, and his end. Complete insight into the order of the cosmos is his aim—an aim not yet reached (nat. 7.25.4): Veniet tempus quo ista quae nunc latent in lucem dies extrahat et longioris aevi diligentia. ad inquisitionem tantorum aetas una non sufficit, ut tota caelo vacet. (“There will come an age when time and long-standing and thorough research reveal whatever is now concealed. The span of one human life is not enough for the investigation of such a demanding subject, even if it has been completely devoted to the study of the heavens.”)

PHYSICS III: THEOLOGY*

Aldo Setaioli

1. The first book of the Natural Questions opens with an enthusiastic praise of theology, conceived of as a personal search after the nature and ontological essence of godhead, beyond all representations transmitted by cultural and literary tradition or fostered by institutional and political conventions. In Varronian terms, one might say that here Seneca favors the theologia naturalis over both the theologia fabulosa and the theologia civilis;1 or, according to the terminology used by Seneca himself in his lost De superstitione, in this page he is concerned with the res, the “thing itself,” rather than the mos, the convention or custom of the generally accepted religious practice.2 Here theology is, to all effects, identified with physics:3 the cosmos is conceived of as God’s visible manifestation, and, as we shall see, knowledge of God is required in order to honor him correctly. For this reason, Seneca grants physics (i.e., theology) primacy over the two other branches of philosophy4—not merely over logic, but over ethics as well: obviously, a correct moral behavior and progress toward virtue (the main object of Seneca’s philosophical writing) cannot be achieved without gaining a correct knowledge of God and our relationship to him.5

Submitted for publication in 2007. For Seneca’s basic acceptance of this distinction, cf. Vottero 1998: 53f. 2 Sen. frg. 39 Haase = F 72 Vottero: omnem istam […] deorum turbam […] sic […] adorabimus ut meminerimus cultum eius magis ad morem quam ad rem pertinere. Cf. Mazzoli 1984: 986. 3 Cf. SVF II 42, where theology is presented as the crowning of physics. 4 Sen. nat. 1 pr. 2. Although Seneca refers only to physics and ethics, errores nostros discutit and ambigua vitae (cf. epist. 90.29) suggest that he has logic too in mind, though it is subordinate to ethics. 5 The Stoics, in fact, defined philosophy as ἐπιστήµην θείων τε καὶ ἀνθρωπίνων πραγµάτων (SVF II 36; cf. 35, 1017), a definition accepted by Seneca (divinorum et humanorum scientiam: epist. 89.5; cf. 90.3). These two components can hardly be separated. Contemplation, that is meditation directed at higher realities, cannot be separated from action (dial. 8 [= de otio].5.8). Only he who knows God can properly honor him (epist. 95.47), but we can do so only by 1

380

aldo setaioli

Seneca’s texts on God and our relationship to him are numerous and permit us to sketch a clear picture of his standpoint.6 To begin with, there are passages that testify to a strong religious sensitivity. The beauty of nature, in particular, gives him an indefinite religious shiver—quaedam religionis suspicio.7 The Greek Stoics, too, drew the idea of God from the beauty of the cosmos,8 but most of the times they favored rational inference: the perfection of the created universe testifies to the wisdom of a divine craftsman9—a sort of cosmological proof of God’s existence, which, of course, repeatedly appears in Seneca, too.10 The latter, however, aims to communicate the inner experience of the divine.11 Significantly, he experiences the same religious shiver both when he faces unspoiled and uncontaminated nature and when he pictures the ethical perfection of the soul of the good and wise man.12 Aesthetic and religious experiences are inextricably intertwined, as are sensible and moral beauty, inasmuch as they are complementary aspects of the divine.13 Seneca, however, is fully aware of the fact that this instinctual stage is indeed very far from the knowledge of God’s innermost essence, which, in his own words, “escapes the eyes and must be viewed by thought.”14 On a less emotionally subjective level Seneca resorts to one of his school’s standard arguments in order to prove the existence of God: the consensus of all nations in this belief. 15 He obviously refers to the Stoic doctrine of the

imitating him (epist. 95.50, cf. 90.34); see below. Clearly, Seneca’s interest for practical ethics cannot be totally suppressed in favor of a merely theoretical contemplation of God. 6 An essential collection of Seneca’s pronouncements on the subject is provided by Motto 1955b. Cf. also Motto 1970, s.v. “God,” “piety,” “prayer,” “superstition,” and “worship.” 7 Sen. epist. 41.3. 8 SVF II 1009. 9 Cf., e.g., SVF II 1010, 1106 f. 10 E.g., Sen. dial. 1 (= prov.).1.2–4, benef. 7.31.4, nat. 1 pr. 14f. 11 Cf. Bellincioni 1986: 33 (the essay first appeared as Bellincioni 1980), Armisen-Marchetti 1990a: 89. This work contains an important treatment of epist. 41.2–5 (for which see also Grammatico 1987: 144f.) and 115.3–6. In Seneca’s attitude Armisen-Marchetti sees a heritage of Roman native religious sensitivity (cf. Andreoni Fontecedro 2000: 189) as well as the echo of contemporary taste for the “numinous” in nature; Seneca, however, is closer to Cic. nat. 2.98 than to first-century ad descriptions. 12 Sen. epist. 41.2–5, 115.3–6. For the nod to Plat. Phaedr. 250d (cf. Cic. off. 1.15, fin. 2.52) cf. Armisen-Marchetti 1990a: 94, Setaioli 2008. 13 Cf. Setaioli 2007b: 52–57 for the texts quoted in the previous note and dial. 6 (= cons. Marc.).18, benef. 4.5, dial. 12 (= cons. Helv.).8.2–6. 14 Sen. nat. 7.30.3: effugit oculos; cogitatione visendus est (cf. below, n. 114). 15 Sen. epist. 117.6: multum dare solemus praesumptioni omnium hominum et apud nos veritatis argumentum est aliquid omnibus videri; tamquam deos esse inter alia hoc colligimus,

physics iii: theology

381

κοιναὶ ἔννοιαι—conceptions considered to be common to all mankind and an infallible criterion of truth16—as made clear by the term praesumptio, coined by Seneca as an exact calque of πρόληψις, used by the Greek Stoics to express this idea.17 2. As far as the ontological essence of God is concerned, Seneca accepts the Stoic concept of πνεῦµα (spiritus), the fiery breath pervading and animating every part of the universe.18 This breath is completely immanent in the material cosmos19 and although, as we shall presently see, it is conceptually distinct, as an active (τὸ ποιοῦν) rational power (λόγος, ratio), from matter (οὐσία) and its sheer passivity (τὸ πάσχον), it is itself material, to the point that, broadly speaking, it can be identified not merely with nature,20 but with the cosmos itself.21 Clearly, Seneca accepts, in principle, the monistic conception of Stoicism. His universe is unitary22 and his spiritus is not

quod omnibus insita de dis opinio est nec ulla gens usquam est adeo extra leges moresque proiecta ut non aliquos deos credat. Cf. Wildberger 2006: I, 27f. This was, of course, a traditional argument, found, for instance, in Cicero, nat. 2.5, 2.15, and in S. Emp. adv. math. 9.61– 71. 16 Cf. SVF II 473 (p. 154,29f.). For the application of this doctrine to the existence of God, cf., e.g., SVF II 337. See Verbecke 1993. 17 Unfortunately, the dictionary of Pittet 1937 did not proceed beyond the letter c; that of Borgo 1998 is limited to ethical vocabulary. Neither deals with the term praesumptio. Cicero, though repeatedly referring to the same doctrine, is content with vague and inaccurate circumlocutions. At nat. 1.43f. (anticipatio, praenotio) he refers to Epicurean, not Stoic πρόληψις. For Cicero’s renderings, cf. Li¸scu 1930: 115–128, Hartung 1970: 78–101, Lévy 1992a: 302–304. For Seneca’s contribution to philosophic terminology, see, e.g., Setaioli 2000: 97–109, Setaioli 2006–2007: 336. 18 Suffice it to refer to the lively exposition found at Sen. nat. 6.16.1. 19 At Sen. dial. 12 (= cons. Helv.).8.3 the Stoic god is defined as divinus spiritus per omnia maxima et minima aequali intentione diffusus. Here Seneca also hints at the Stoic conception of τόνος (intentio), though it is not true that this is the same everywhere. In this passage the non-Stoic conception of an incorporeal god (incorporalis ratio) is also envisaged. Cf., e.g., Lagrange 1928: 332 f. 20 Cf., e.g., Sen. benef. 4.7.1: quid enim aliud est natura quam deus et divina ratio toti mundo partibusque eius inserta?; 4.8.2, nat. 2.45.3, frg. 122 Haase = F 84 Vottero. See, e.g., Wildberger 2006: I, 35 f. 21 Cf., e.g., Sen. epist. 92.30: totum hoc quo continemur et unum est et deus; et socii sumus eius et membra. Cf. 95.52: membra sumus corporis magni; nat. 7.30.4: regnum suum, id est se, regit. 22 In agreement with Zeno and Chrysippus: SVF II 531: εἷς ἐστι [ὁ κόσµος]. Cf. Wildberger 2006: I, 17. For men as limbs of God, cf. SVF III 4: µέρη γάρ εἰσιν αἱ ἡµέτεραι φύσεις τῆς τοῦ ὅλου.

382

aldo setaioli

an immaterial “spirit”—in the Stoic conception only a body can indeed act on other bodies.23 Surely, there are a number of texts in which Seneca appears to assume a distinction between God and the world; the former can be seen both as the mind governing the universe and as the universe itself governed and ruled by that mind.24 This is why Seneca says that God is all we can see25 and, elsewhere, that he is both all we can see and all we cannot see.26 Throughout Epistle 65 to Lucilius he opposes God, conceived of as an active and acting power, to a totally passive matter,27 and often, elsewhere, he calls him rector, or artifex, or conditor of the universe.28 But to Seneca God remains “the greater and better part of his own work,”29 or, as he puts it with an expressive boldness that, though seemingly paradoxical, provides nevertheless a faithful formulation of the innermost meaning of the Stoic doctrine, “his own maker.”30 The whole cosmos can be called “God” as the word “man” denotes the human microcosm made up of body and soul. But just as Seneca agrees with Plato in considering the soul as each man’s real self,31 so, in the universe, God, the mens universi,32 corresponds to the soul in man.33 In his capacity of universal λόγος God can

23 Criticism of the incorporeal god (deum sine corpore) of Platonism is implied in Sen. frg. 32 Haase = F 66 Vottero. Cf. Vottero 1998: 305. For God in Stoicism, cf., e.g., Long and Sedley 1987: I, 277–279. For Seneca, e.g., Gentile 1932: 19–27. 24 Cf., e.g., Wildberger 2006: I, 22. Also Gersh 1986: I, 165f. 25 Sen. nat. 2.45.3: ipse enim est hoc quod vides totum, partibus suis inditus. 26 Sen. nat. 1 pr. 13: quid est deus? quod vides totum et quod non vides totum. Cf. Lagrange 1928: 332. 27 Clearly, Seneca has the Stoic distinction between τὸ ποιοῦν and τὸ πάσχον in mind: epist. 65.23: potentius autem ac pretiosius quod facit, quod est deus, quam materia patiens dei. 28 Cf., respectively, Sen. dial. 7 (= vit. beat.).8.4 and epist. 65.23, nat. 2.35.1 and epist. 65.19, 119.5. 29 Sen. nat. 7.30.3: maior […] pars sui operis et melior. 30 Sen. frg. 15 Haase = F 87 Vottero: nos aliunde pendemus […] deus ipse se fecit. Cf. Scarpat 19702: 153, Lausberg 1970: 93f., Mazzoli 1977: 29f., Mazzoli 1984: 961, Wildberger 2006: I, 14 f. God “makes himself” as active principle (ποιοῦν) acting upon passive matter (πάσχον), thus creating the world, which, broadly speaking, can itself be called “God.” 31 Cf. Sen. dial. 6 (= cons. Marc.).24.5, 25.1, and the Platonic texts quoted in Setaioli 2000: 295 and n. 116. For this idea in Platonism and in pagan and Christian Latin literature, see Setaioli 1995: 17 f., with the bibliography quoted and discussed. 32 Sen. nat. 1 pr. 13; cf. benef. 4.7.1. Seneca is obviously thinking of the λόγος of the Greek Stoics. 33 Sen. epist. 65.24: quem in hoc mundo locum deus obtinet, hunc in homine animus; quod est illic materia, id in nobis corpus est; nat. 2.45.1: animum et spiritum mundi; cf. 1 pr. 15. Cf. Bodson 1967: 32f., Rist 1969: 258f., Mazzoli 1984: 958. The idea is at least as old as Cleanthes: cf. Wildberger 2006: II, 492 f. n. 132.

physics iii: theology

383

be said to be “pure soul”34—a soul that, though unfolding toward the outside, constantly returns toward and into itself.35 Though Seneca’s god is one of the two principles recognized by Stoic ontology—the other being matter36—it should be emphasized that this god cannot be conceived of as an entity separated from his phenomenalization in nature and the cosmos.37 God is surely λόγος, ratio, but, as Seneca says in an epistle,38 he is a ratio faciens, a reason that prints its mark upon the cosmos. Seneca’s universe is not dualistic in any way, though, as we shall see, the influence of Platonism in his work goes beyond mere linguistic formulations.39 We can at most speak of an “immanent dualism,”40 or of a cosmos resembling one, boundless Aristotelian σύνολον, in which form and matter cannot be separated.41 Only at the end of each cosmic cycle, marked by a universal conflagration (ἐκπύρωσις), does God’s actualizing power rest for a moment, leaving matter in a state of chaos.42 Such a unitary cosmos entails no impassable gulf between God and man: it is indeed urbs dis hominibusque communis,43 as Seneca puts it, echoing well-known formulations by the Greek Stoics.44

Sen. nat. 1 pr. 14: in illo nulla pars extra animum est; totus est ratio. Cf. Wildberger 2006: I,

20. 35

Sen. dial. 7 (= vit. beat.).8.4: rectorque universi deus in exteriora quidem tendit, sed tamen introsum undique in se redit. 36 Sen. epist. 65.23: universa ex materia et ex deo constant. Cf. Wildberger 2006: I, 3–5. 37 Sen. benef. 4.8.2: nec natura sine deo est nec deus sine natura. Cf., e.g., Rozelaar 1976: 459, Ortega Muñoz 1983: 313, Wildberger 2006: I, 11, 16. Cf., e.g., SVF I 88. 38 Sen. epist. 65.12. Cf. Wildberger 2006: I, 39. 39 On this level Seneca’s predilection for the rhetorical mold of the antithesis can at times enhance the impression of dualism. 40 This felicitous oxymoron has been coined by Mazzoli 1984: 958 (“dualismo ‘immanente’”). Cf. Wildberger 2006: I, 18. For the unacceptable idea of Seneca smuggling Platonic conceptions under cover of Stoicism cf. Setaioli, supra, p. 200. 41 So Scarpat 19702: 156f. André 1983: 55 considers Seneca’s position to be “une sorte de mysticisme rationnel.” Cf. Attridge 1978: 68 f. 42 Sen. epist. 9.16: resoluto mundo et dis in unum confusis paulisper cessante natura [scil. Iuppiter] acquiescit sibi cogitationibus suis traditus. Cf. SVF II 1064, Epict. diatr. 3.13.7. For the idiosyncratic use of the reflexive pronoun (acquiescit sibi), anticipating turns found in the Romance languages and at the same time emphasizing through language and style the inextricable unity of the active and passive aspect of Seneca’s universe, with God’s acting upon matter amounting to acting upon himself, cf. Setaioli 2006–2007: 339–341. Cf. also nat. 7.30.4: regnum suum, id est se, regit and frg. 15 Haase = F 87 Vottero, quoted in notes 21 and 30, respectively. 43 Sen. dial. 6 (= cons. Marc.).18.2. 44 SVF II 528: πόλις ἐκ θεῶν καὶ ἀνθρώπων συνεστῶσα. Cf. II 527. We shall come back to this correlation and affinity between gods and men. For now we shall be content with pointing out

384

aldo setaioli 3.

One might think that a dualistic element did find its way into Seneca’s basically monistic conception by way of his ideas on the hereafter. Though at times he considers death to mark the total annihilation of man, not rarely he does in fact present it as a passage to a new and better life.45 His emotional longing for the immortality of the soul, however, never quite turns into a philosophical position advocating a status of the soul distinct from, and opposed to, the physical world.46 It is actually in this connection that we find a passage proving, on the one hand, that Seneca was not insensitive to the ideas current in contemporary Platonism, but permitting us, on the other, to appreciate how he adapted them to Stoicism47 by transforming them to fit the “immanent dualism” we have been hinting at. When Seneca compares the soul to the sun, whose beams, though reaching down to earth, remain nevertheless attached to their heavenly origin,48 there can be no doubt that his language is influenced by Platonic conceptions. Exactly the same image appears, a few centuries after Seneca, in a Latin Neoplatonist: Macrobius.49 The very different meaning an important corollary of the idea: man is a sacred thing for every other man, and should be treated as such (Sen. epist. 95.33, cf. dial. 4 [= de ira 2].31.7). Cf. Bellincioni 1979: 274, Mazzoli 1984: 970. 45 In other words, Seneca fluctuates between the two prospects proposed by Socrates in Plat. apol. 40c–41c (the so-called “Socratic alternative”). 46 A detailed study of this problem and the related bibliography may be found in Setaioli 1997a, collected and updated in Setaioli 2000: 275–323, 411. It is hardly necessary to retrace my argument here. 47 Seneca does not ‘Platonize’ Stoicism; he rather ‘Stoicizes’ Platonism. Cf., e.g., Wildberger 2006: II, 457 n. 36. Neither Wildberger nor, to my knowledge, other scholars (cf., e.g., Rozelaar 1976: 466f.), however, connect Sen. epist. 41.5 with the Platonizing texts (Porphyry, Macrobius) we shall presently mention. Andreoni Fontecedro 2000: 190, who also ignores Porphyry and Macrobius, considers Sen. epist. 41.5 to be an unaccomplished anticipation of the Platonists’ concept of emanation, but in view of the parallels we shall point out, and in spite of Macrobius’s linguistic echoes of Seneca, it seems to me much more probable that Seneca is adapting a Platonic idea rather than that Porphyry may have taken up Seneca’s image in order to give it a Platonizing twist. 48 Sen. epist.41.5: [animus] maiore sui parte illic est unde descendit. quemadmodum radii solis contingunt quidem terram sed ibi sunt unde mittuntur, sic animus magnus et sacer […] conversatur quidem nobiscum, sed haeret origini suae; illinc pendet, illuc spectat et nititur, nostris tamquam melior interest. Similar ideas, without the image of the sun, are found elsewhere in Seneca: e.g., epist. 65.18, benef. 3.20.1. For this passage and its parallels in later Platonism, cf. Setaioli 2000: 309 f., Setaioli 2006–2007: 342 f. 49 Macr. somn. 1.21.34: sicut solem in terris esse dicimus, cuius radius advenit et recedit, ita animorum origo caelestis est sed lege temporalis hospitalitatis hic exulat. Macrobius’s formulation is very probably influenced by Seneca (in terris ~ terram; radius ~ radii).

physics iii: theology

385

expressed by this image in Platonism is apparent in Macrobius’s context, and is spelled out in a passage of Porphyry’s, in which we find, appended to the same image, the specific remark that the soul is a transcendent entity that cannot mix with corporeal objects and is radically different in essence from all sources of material light, including the sun.50 Seneca, by contrast, often states that the soul is akin to the cosmos or even its most magnificent part.51 I have hinted at contemporary Platonism, rather than at Plato, quite on purpose. In Seneca we find repeated allusions to the resistance of matter to God’s action in the universe, which may appear to introduce a further element of at least seeming dualism. Though some scholars deny any Platonic influence in this connection,52 the fact cannot be denied that these allusions recur, among other texts, in Epistles 58 and 65, where Platonic, or rather Platonizing ideas are discussed; that they are accompanied by hints at undoubtedly Platonizing conceptions; and—last but not least—that they are closely matched by doctrines found in Middle Platonic and Neoplatonic texts that have come down to us. In my opinion, the presence of Platonizing suggestions cannot be denied in Seneca; what we should be concerned with is to ascertain whether they introduce an element that cannot be re-absorbed into Stoic monism. In the preface to the first book of the Natural Questions Seneca asks himself whether God accomplishes whatever he wants or whether matter sometimes prevents this great craftsman from achieving perfection53—and in several passages he answers his own question by accepting the second alternative. He has God himself say in the De providentia that it was not in his power

50 Porph. F 261F, p. 288f. Smith: ὡς γὰρ ὁ ἥλιος τῇ παρουσίᾳ τὸν ἀέρα εἰς φῶς µεταβάλλει ποιῶν αὐτὸν φωτοειδῆ, καὶ ἑνοῦται τῷ ἀέρι τὸ φῶς ἀσυγχύτως ἅµα καὶ αὐτῷ κεχυµένον, τὸν αὐτὸν τρόπον καὶ ἡ ψυχὴ ἑνουµένη τῷ σώµατι µένει πάντως ἀσύγχυτος, κατὰ τοῦτο µόνον διαλλάττουσα, ὅτι ὁ µὲν ἥλιος σῶµα ὢν καὶ τόπῳ περιγραφόµενος οὐκ ἔστι πανταχοῦ, ἔνθα καὶ τὸ φῶς αὐτοῦ, ὡς οὐδὲ τὸ πῦρ· µένει γὰρ καὶ αὐτὸ ἐν τοῖς ξύλοις ἢ ἐν θρυαλλίδι δεδεµένον ὡς ἐν τόπω. ἡ δε ψυχή, ἀσώµατος οὖσα καὶ µὴ περιγραφοµένη τόπῳ, ὅλη δι’ ὅλου χωρεῖ καὶ τοῦ φωτὸς ἑαυτῆς καὶ τοῦ σώµατος, καὶ οὐκ ἔστι µέρος φωτιζόµενον ὑπ’ αὐτῆς ἐν ᾧ µὴ ὅλη πάρεστιν. The image and the idea had probably been current for a long time in Platonism. 51 E.g., Sen. dial. 12 (= cons. Helv.).8.4: pars eius magnificentissima. This is why it is naturally bent on contemplating the cosmos: dial. 12 (= cons. Helv.).8.6: animum ad cognatarum rerum conspectum tendentem; nat. 1 pr. 12: in originem redit […] scit illa ad se pertinere. 52 As done, e.g., by Wildberger 2006: I, 53–55, 276 f. The problem is merely hinted at in Grammatico 1987: 140. 53 Sen. nat. 1 pr. 16: deus quicquid vult efficiat an multis tractanda destituant et a magno artifice prave multa formentur non quia cessat ars, sed quia id in quo exercetur saepe inobsequens arti est. What is at stake is God’s power: quantum deus possit (ibid.).

386

aldo setaioli

to exempt men from misfortunes,54 the reason being that “the craftsman cannot change matter,” as Seneca has said a few lines before.55 This concept is developed at greater length in Epistles 58 and 65—two letters in which in all probability Seneca drew upon a commentary on Plato’s Timaeus, as I have argued elsewhere.56 In the first one Seneca pictures God roaming in the midst of Plato’s ideas and seeing to the permanence of his creatures,57 which he could not endow with inborn immortality because of matter’s resistance.58 Though in all probability referring to the sensible world, Seneca’s mention of a permanence guaranteed by God’s assistance in overcoming the creatures’ own perishability is strongly reminiscent of the words addressed in Plato’s Timaeus by the demiurge to the lesser gods, who have been created by him, and because of this owe immortality not to their own nature but to his will and care.59 In Seneca too the supreme god pervading the cosmos creates lesser deities to act as his ministers.60 The idea of lesser gods is anything but foreign to Stoicism,61 and the concept of an immanent god is as orthodox as can be; but the supreme god generating62 lesser deities at the beginning of creation closely resembles the picture sketched by Plato in the

54 Sen. dial. 1 (= prov.).6.6: non poteram vos istis subducere. Cf. Epict. diatr. 1.1.12: οὐκ ἐδυνάµην. See supra, p. 298, n. 179. 55 Sen. dial. 1 (= prov.).5.9: non potest artifex mutare materiam. Cf., e.g., Riesco Terrero 1966: 62f. Here Seneca appears to have solved in accordance with the “immanent dualism” of the Stoa the problem he poses at nat. 1 pr. 16: ipse [scil. deus] materiam sibi formet an data utatur. 56 Setaioli 1988: 126–140. 57 If, as I believe, Seneca is still moving within the frame of Stoic cosmology, this permanence is perforce limited to the duration of the cosmic cycle. Foraeternus used in this sense, cf. Setaioli 2000: 304 and n. 160. 58 Sen. epist. 58.27: miremur in sublimi volitantes rerum omnium formas deumque inter illas versantem et hoc providentem, quemadmodum quae inmortalia facere non potuit, quia materia prohibebat, defendat a morte ac ratione vitium corporis vincat. 28: manent enim cuncta, non quia aeterna sunt, sed quia defenduntur cura regentis: inmortalia tutore non egerent. Wildberger 2006: I, 54 f. is surely right in remarking that the way Seneca pictures Plato’s ideas has not a few unplatonic traits. It can hardly be denied, however, that the ideas are undoubtedly propria Platonis suppellex, to borrow Seneca’s own words (epist. 58.18); besides, as already remarked, Seneca is not following Plato, but the Platonism of his own time. 59 Plat. Tim. 41b: ἐπείπερ γεγένησθε, ἀθάνατοι µὲν οὐκ ἐστὲ οὐδ’ ἄλυτοι τὸ πάµπαν, οὔτε µὲν δὴ λυθήσεσθέ γε οὐδὲ τεύξεσθε θανάτου µοίρας, τῆς ἐµῆς βουλήσεως µείζονος ἔτι δεσµοῦ λαχόντες ἐκείνων οἷς ὅτ’ ἐγίγνεσθε συνεδεῖσθε. 60 Sen. frg. 16 Haase = F 86a Vottero: cum prima fundamenta molis pulcherrimae iaceret […], quamvis ipse per totum se corpus intenderat, tamen ministros regni sui deos genuit. 61 We have already found them at Sen. epist. 9.16 (supra, note 42), and will return to them below. Lesser deities as ministers to the supreme god appear, e.g., at SVF II 1178. 62 Γένεσις and γίγνεσθαι recur very frequently at Plat. Tim. 40d–f.

physics iii: theology

387

Timaeus.63 In the Timaeus passage, however, it is the fact of being generated that, by itself, rules out immortality, whereas in Epistle 58 Seneca, as we have seen, ascribes this characteristic to the resistance of matter limiting God’s power. In Epistle 65 Seneca attributes to Plato an outline of philosophical causes that has no parallel in the master himself but is closely matched in later Platonic tradition from Alcinous down to Porphyry and to Proclus.64 In this letter Seneca quotes a passage from the Timaeus in his own translation.65 Plato’s text does indeed present the idea of the limitation and imperfection of things created, which God wishes to make “as much like him as poss