Unveiling Emotions II: Emotions in Greece and Rome: Texts, Images, Material Culture 3515106375, 9783515106375

The study of emotions has emerged as one of the most dynamic topics of research in Ancient History, Classics, and Archae

100 76 10MB

English Pages 387 [393] Year 2013

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
CONTENTS
PREFACE
APPROACHING EMOTIONS IN GREEK AND ROMAN HISTORY AND CULTURE
EMOTIONS AND HISTORICAL REPRESENTATION IN XENOPHON’S HELLENIKA
EMPATHY, EMOTIONAL DISPLAY, THEATRICALITY, AND ILLUSION IN HELLENISTIC HISTORIOGRAPHY
A SHORT HISTORY OF SHUDDERS
REFLECTIONS ON THE DISCOURSE OF FEAR IN GREEK SOURCES
EVOKING ANGER THROUGH PITY
‘NEGATIVE’ EMOTIONS AND GREEK NAMES
IS PISTIS/FIDES EXPERIENCED AS AN EMOTION IN THE LATE ROMAN REPUBLIC, EARLY PRINCIPATE, AND EARLY CHURCH?
MODELING ROMAN PRIDE
GRIEF AND MOURNING IN THE ROMAN CONTEXT
GALEN AND GRIEF
EMOTIONALITY IN GREEK ART
FEELING LOW
THE IMPRINT OF EMOTIONS SURROUNDING THE DEATH OF CHILDREN IN ANTIQUITY
AFFECTIVE POLITICS
ABBREVIATIONS
INDEX
CONTRIBUTORS
Recommend Papers

Unveiling Emotions II: Emotions in Greece and Rome: Texts, Images, Material Culture
 3515106375, 9783515106375

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Unveiling Emotions II Alte Geschichte Franz Steiner Verlag

Emotions in Greece and Rome: Texts, Images, Material Culture Edited by Angelos Chaniotis and Pierre Ducrey

HABES 55

Unveiling Emotions II Edited by Angelos Chaniotis and Pierre Ducrey

habes Heidelberger Althistorische Beiträge und Epigraphische Studien Herausgegeben von Angelos Chaniotis und Christian Witschel. Beirat: François Berard (Lyon), Anthony R. Birley (Vindolanda/ Friedberg), Kostas Buraselis (Athen), Lucas de Blois (Nijmegen), Ségolène Demougin (Paris), Elio Lo Cascio (Rom), Mischa Meier (Tübingen), Elizabeth Meyer (Charlottesville), Silvio Panciera (Rom), Michael Peachin (New York), Henk Versnel (Leiden) und Martin Zimmermann (München). Band 55

Unveiling Emotions II Emotions in Greece and Rome: Texts, Images, Material Culture Edited by Angelos Chaniotis and Pierre Ducrey

Franz Steiner Verlag

Umschlagabbildung: Theatrical masks from the Portico of Tiberius, Aphrodisias. Photo: Angelos Chaniotis

Bibliografische Information der Deutschen Nationalbibliothek: Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über abrufbar. Dieses Werk einschließlich aller seiner Teile ist urheberrechtlich geschützt. Jede Verwertung außerhalb der engen Grenzen des Urheberrechtsgesetzes ist unzulässig und strafbar. © Franz Steiner Verlag, Stuttgart 2013 Druck: Laupp & Göbel, Nehren Gedruckt auf säurefreiem, alterungsbeständigem Papier. Printed in Germany. ISBN 978-3-515-10637-5

CONTENTS Preface......................................................................................................................7 Approaching emotions in Greek and Roman history and culture: Introduction......9 ANGELOS CHANIOTIS and PIERRE DUCREY Emotions and historical representation in Xenophon’s Hellenika.........................15 MELINA TAMIOLAKI Empathy, emotional display, theatricality, and illusion in Hellenistic historiography..................................................................................................53 ANGELOS CHANIOTIS A short history of shudders....................................................................................85 DOUGLAS CAIRNS Reflections on the discourse of fear in Greek sources.........................................109 MARIA PATERA Evoking anger through pity: portraits of the vulnerable and defenceless in Attic oratory..............................................................................................135 LENE RUBINSTEIN ‘Negative’ emotions and Greek names................................................................167 NIKOLETTA KANAVOU Is pistis/fides experienced as an emotion in the Late Roman Republic, early Principate, and early Church?...............................................................191 TERESA MORGAN Pride in the Roman world....................................................................................215 YELENA BARAZ Grief and mourning in Roman context.................................................................237 KATARIINA MUSTAKALLIO Galen and grief: The construction of grief in Galen’s clinical work ..................251 DANIEL KING

6

Contents

Emotionality in Greek art.....................................................................................273 OLYMPIA BOBOU Feeling low: Social status and emotional display in Hellenistic art.....................313 JANE MASSÉGLIA The imprint of emotions surrounding the death of children in antiquity.............331 CHRYSSA BOURBOU Affective politics: the emotional regime in the Imperial Greek city..................351 ONNO M. VAN NIJF Abbreviations.......................................................................................................369 Indices..................................................................................................................371 List of Contributors..............................................................................................385

PREFACE Angelos Chaniotis and Pierre Ducrey The idea for this volume was born in Gare du Nord, in Paris, in 2006, when Pierre Ducrey, then Treasurer of the International Committee of Historical Sciences (CISH) and member of the Programme Committee of the 21st CISH International Congress of Historical Sciences, and Angelos Chaniotis, then Senior Research Fellow for Classics at All Souls College, were looking for a theme for a panel dedicated to ancient history in the forthcoming Congress – a congress that covers all historical disciplines, all periods, and all regions of the world. The subject of emotions immediately appealed to us as a subject that both reflects current trends in ancient history and classics and opens possibilities for a dialogue among the historical disciplines. The CISH Programme Committee first accepted it as ‘Round Table’, to upgrade it later to a ‘Specialised Theme’, due to the importance of the subject and the rich content of the proposed program. The proposal was finally accepted in the CISH General Assembly at Beijing, in September 2008. In the meantime (June 2008), Chaniotis received a grant from the European Research Council for the project ‘The Social and Cultural Construction of Emotions: the Greek Paradigm’ (University of Oxford, 2009–2013).1 This grant provided the funds for the organisation of the Table Ronde ‘Emotions as Historical Factor’, which took place during the 21st CISH International Congress of Historical Sciences in Amsterdam (26 August, 2010). Ten scholars from Greece, Finland, Switzerland, the United Kingdom, and the USA presented papers and contributed to the discussion. The majority of the chapters in the present volume were originally written for that Table Ronde (by Douglas Cairns, Angelos Chaniotis, Jane Masséglia, Teresa Morgan, Katariina Mustakallio, Maria Patera, and Onno van Nijf). In order to provide a representative sample of approaches to emotions in Greece and Rome by historians, philologists, archaeologists, and art historians, the organisers solicited more contributions – having no illusion that a single volume could ever comprehensively cover the subject. Several contributors to this volume were recipients of scholarships through the Oxford project or participated in workshops organised by the project (Olympia Bobou, Douglas Cairns, Nikoletta Kanavou, Daniel King, Jane Masséglia, Lene Rubinstein, and Melina Tamiolaki). As already said, this volume does not comprehensively cover all aspects of emotions in Greece, Rome, and the Roman Empire, although it comprises a representative sample of sources (Greek and Latin historiography, oratory, and poetry, 1

See http://emotions.classics.ox.ac.uk.

8

Preface

inscriptions, medical treatises, archaeological sources, and personal names), methods (analysis of texts and language, iconology, study of skeletal remains), emotions (anger, grief, pride, fear, and joy), and themes that range from emotional arousal, the study of emotional communities, the interdependence of status and emotional expression, the description of emotional episodes in historiography, and the physiological aspects of emotions to ancient discourses of emotion, the interdependence of funerary ritual and emotion, the difficulties in reconstructing the emotional background of child burials, and the question of whether trust was experienced as an emotion. This volume continues the problematique explained in the volume Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World (edited by Angelos Chaniotis in the same series). Four introductory essays in that volume describe the problems connected with the study of emotions in papyri, inscriptions, literary texts, and archaeological sources. Further essays discuss case studies. The chapters in this volume should be understood as approaches to further case studies and types of sources. The editorial work for this volume received valuable help from the research assistants of the Oxford project, Dr Harriet Archer, Emily Lord-Kambitsch, Dr Jonah Rosenberg, and Katharine Waterfield, who proofread the volume and corrected the English of the contributors who are not native speakers. Michael Anthony Fowler (Columbia University), Chaniotis’ research assistant at the Institute for Advanced Study, translated the chapter written by Maria Patera. The editors are also very grateful to Harald Schmitt (Steiner Verlag) for his assistance in technical aspects of the volume’s production. The editors gratefully acknowledge the contribution of the European Research Council, which funded the Oxford project. Neither the Table Ronde nor this volume would have been possible without the ERC’s generous funding.

APPROACHING EMOTIONS IN GREEK AND ROMAN HISTORY AND CULTURE An Introduction Angelos Chaniotis and Pierre Ducrey In Ismail Kadare’s novel Aksidenti (The Accident) an Albanian analyst employed by the European Council and an Albanian woman working in the Archaeological Institute in Vienna are killed in a car accident in Austria under unclear circumstances. The authorities are puzzled about this accident, which may not have been an accident after all. Above all they are puzzled about the relationship between the man and the woman. Despite the fact that they have access to their correspondence and other documents and that they interview friends, acquaintances, and the individuals who had observed them in their last days, they ultimately cannot say whether it was love that brought them together. An independent researcher continues the search for truth, connecting it with the question of whether love truly exists. In order to answer these questions, he reconstructs the last forty weeks of the couple’s lives and presents this reconstruction in a narrative, in which the man’s and the woman’s perspectives alternate. He finds out – or so he thinks – that nothing was as it seemed. As Akira Kurosawa’s Rashomon can be seen as an essay on the subjective nature of historical accounts, Kadare’s novel addresses the limited possibilities of an observer to understand the emotions of others. Therefore, it addresses the question of whether historians can study emotions. ‘Yes, we can!’ is the response given by an increasing number of historians in the last decades.1 The study of emotions has emerged as one of the most dynamic subjects of historical research. Emotions hold a strong position in ancient studies as well, for more than a decade.2 The question no longer is whether students of classical antiquity should consider emotions in their research. As there is hardly any ancient text or image that does not directly or indirectly originate in emotions, reveals emotions, or seeks to arouse emotions, classicists have no other choice but to attempt to reconstruct the emotional background of their sources. War is a case in point. War ranks high up among the factors that influenced political and social institutions, and left its imprint on art, literature, and culture,3 thus allowing us to measure the role and importance of feelings, both collective and individual. Euro1 2 3

Hitzer 2011; Plamper 2012. For an overview see Chaniotis 2012b. E.g., Ducrey 1985 [2009]; Raaflaub and Rosenstein (eds.) 1999; Chaniotis 2005.

10

Angelos Chaniotis and Pierre Ducrey

pean literature begins with emotional and emotive images of war in the Iliad and the poetry of Archilochos, and scenes of battles and the sack of cities inspired artists from the beginning of Greek visual art, in Mycenaean painting and stonecarving. The scene in the Iliad, in which Achilles kills innumerable Trojans thus provoking the anger of the river Skamandros, who threatens to drown him (21.214–273), could not have left an ancient audience unmoved, and a similar emotive power stems from a relief jar from Mykonos (c. 670 BCE).4 Achaean warriors are shown seizing infants from the arms of their mothers and cutting them in two with their huge swords. In scenes such as this – or as the representations of the Sack of Troy on an amphora by the Kleophrades Painter (c. 500 BCE; see p. 275 figure 2) – we can make plausible assumptions concerning the emotions expected to be aroused: fear and empathy. The same emotions later prevail in dramatic performances – for instance, in the way Euripides illustrated the fate of captured women and children after the fall of Troy in his Trojan Women, probably with the destruction of Melos by the Athenians in mind (416 BCE). References to audience responses are rare but, when they exist, they are telling. When Phrynichos presented the destruction of Miletos by the Persians (494 BCE) on the tragic stage, he moved the audience to tears. The poet was fined one thousand drachmas for having reminded the Athenians of their misfortunes, and the production of this drama was subsequently forbidden (Herodotos 6.21). War is connected with the whole range of emotional responses: fear of death and defeat; hope, joy, and pride for victory; contempt for the enemy, and gratitude for the successful leader; grief for the dead and affection among comrades; desire for booty and envy for the more fortunate or privileged. For this reason, Greek and Roman historical narratives of war cannot be dissociated from descriptions of emotional backgrounds and emotional responses.5 Analogous observations can be made with regard to most aspects of ancient public life, whether this is the popular assembly of a Greek city or the Roman senate, an Athenian or Roman court, the funeral of a public figure, the celebration of a festival, or the adventus of an emperor.6 The question, therefore, clearly cannot be whether ancient historians and classicists should approach emotions but with what questions they should do so. The nature of their sources7 sets certain limits to their quest. They cannot directly study neurobiological processes, and only in some well-documented cases they 4 5

6

7

Erwin 1963. E.g., Thucydides 3.36.4–6; 3.49.1; 5.84–116; 7.29.5; 7.86.5; Livy 31.24.3. Emotional aspects in the description of sieges: Chaniotis 2013, 451–454. See also, in this volume, pp. 15–52 (Xenophon), 53–84 (Hellenistic historians), and 273–311 (Greek art). Emotions in the interaction of the Roman elite: Kaster 2005. Emotional aspects of festivals: Chaniotis 2011. See also, in this volume, pp. 351–368 (the assembly as an emotional community), 215–235 (pride in the Roman elite interaction), and 237–250 (display of grief in Rome). Overviews of the sources and the methodological problems connected with their study: Kotsifou 2012 (papyri); Chaniotis 2012c (inscriptions); Sanders 2012 (literary sources); Masséglia 2012 (archaeological sources). For the variety and heterogeneity of the source material see also the contributions in Munteanu (ed.) 2011.

Approaching Emotions in Greek and Roman History and Culture: an Introduction

11

may have access to psychological reactions or to the physiological/somatic aspects of emotion.8 But they do have access to the external stimuli that generated emotions. They also have information concerning the various factors that determine the manifestation of emotions. In the context of Greek and Roman history and culture the study of emotions primarily means the study of representations and displays. How were emotions and feelings observed and described in literary sources? How were emotions displayed, concealed, or restrained?9 How were emotions evaluated in intellectual discourses?10 What means were used for the arousal of emotions and the creation of emotional communities?11 How were emotions exploited in persuasion strategies in political life, in the court, in diplomacy, and in interpersonal relations? How were emotional norms (or ‘emotional regimes’) constructed and transmitted?12 Were certain historical periods dominated by a particular emotion or specific attitudes towards emotions? Consequently, the study of emotions in Greek and Roman history and culture means first and foremost the study of contexts of communication and of emotional communities. But it also entails the study of those parameters that determined the arousal, manifestation, and representation of emotions in text, image, and material culture. Such parameters vary, ranging from gender, age, and education to hierarchical relations, religion, ideology, and values.13 The essays assembled in this volume address many of the above questions. The sequence in which the chapters are presented in the volume does not reflect manifold links among them. Such links concern types of sources (historiography: Chaniotis and Tamiolaki; the visual arts: Bobou, Masséglia, and Mustakallio), particular emotions (fear: Cairns and Patera; grief: Bourbou, King, and Mustakallio), approaches (linguistic expression and metaphors: Cairns, Morgan, and van Nijf), subjects (emotional arousal: Chaniotis and Rubinstein; emotional display: Chaniotis and Mustakallio; intellectual discourse: King and Patera; funerary rituals: Bourbou and Mustakallio; public oratory: Rubinstein and van Nijf; social relations and value systems: Baraz, Morgan, Masséglia, and Patera), and timeframes (the Hellenistic period: Boubou, Chaniotis, and Masséglia; the Late Republic and the early Imperial period: Baraz and Morgan). In the field of the literary representation of emotions, Melina Tamiolaki attributes to Xenophon a series of innovations, especially a strong interest in vivid descriptions and in the narration of emotional episodes combined with the use of a more diversified vocabulary (pp. 15–52). For Xenophon emotions served as a medium that enhanced his reader’s understanding of historical events. The part 8 9 10 11 12 13

See the study of shudder by Douglas Cairns in this volume (pp. 85–107). On the restraint of emotion see Harris 2001 and Kalimtzis 2012 (anger); cf. Maria Patera’s observations on the rationalisation of fear (pp. 108–112, 117–119, 123–131 in this volume). E.g., Sihvola and Engberg-Pedersen 1998; Knuuttila 2004; Konstan 2006; Zaborowski 2012. On the concept of emotional community see Rosenwein 2006 and Onno van Nijf in this volume (pp. 351–368); see also Chaniotis 2011 and forthcoming. On ‘emotional regime’ see Reddy 2001 and Onno van Nijf in this volume (pp. 351–368). Gender and emotion: Munteanu (ed.) 2011; cf. pp. 237–250 in this volume.

12

Angelos Chaniotis and Pierre Ducrey

played by emotions in the interaction between historians and their readers is also discussed by Angelos Chaniotis, who focuses on the arousal of empathy in Hellenistic historiography (pp. 53–84). A different aspect of emotional arousal is the subject of Lene Rubinstein (pp. 135–165). As she explains, the presentation of images of vulnerability in Attic court speeches aimed to provoke the pity of the jurors toward the victim and their anger against the defendant; emotional arousal was a persuasion strategy. Persuasion strategies are intrinsically connected with values. An instructive example for the connection between values and emotions is treated by Yelena Baraz (pp. 215–235), who places pride in the context of Roman society, a society obsessed with status. A mismatch between a man’s sense of self-worth and his assessment by his community leads to social tensions and excessive behaviours that may range from mockery to violence. These tensions were addressed by the discourse on pride in Roman society in the Late Republic and the early Imperial period. Social values are also reflected by personal names, a small group of which is examined by Nikoletta Kanavou (pp. 167–189). Kanavou observes that words denoting negative emotions were used for the composition of names that expressed the wish that an individual may remain free of a negative emotion or arouse it in his opponents. Communication is the primary aim of emotional display in the ritual of lament in Roman culture, the subject of Katariina Mustakallio’s contribution (pp. 237– 250). Changes in mourning rituals from the Republic to the early Principate, in particular the gradual expansion of public mourning and demonstrations of grief for members of the elite, are an instructive example for the dynamic character of emotional display and the impact of social, institutional, and cultural changes on the ‘emotional regimes’. A different ‘emotional regime’, that of the Greek city in the Imperial period, is approached by Onno van Nijf (pp. 351–368). Using the concept of the emotional community (see note 11) as a heuristic tool, he shows the political significance of emotional metaphors in the negotiations between citizens and elite, and in political life. The importance of metaphor and metonymy has often been emphasised in studies on emotions;14 it also is the focus of Douglas Cairns’ chapter (pp. 85–107). With the somatic symptoms of fear and the Greek term φρίκη as his starting points, Cairns explains how the designation of a symptom (shudder) comes to function as a name of the related emotion. Thus, he establishes a continuity between emotions as physical experiences and emotional concepts as linguistic and cultural categories. Linguistic expression is also the subject of Teresa Morgan, who explores the social value of trust (pistis, fides) in Greek and Roman culture and in early Christianity (pp. 191–214). Although trust is not an emotion, pistis and fides are invoked in ancient texts in terms indistinguishable from those used for emotions, sharing with emotions the same psychosomatic locations. In these passages pistis and fides function as emotive signals expected to arouse emotions.

14

See Theodoropoulou 2012, with further bibliography.

Approaching Emotions in Greek and Roman History and Culture: an Introduction

13

Maria Patera surveys diverse attitudes towards emotions in Greek sources (pp. 109–133): the rationalisation and control of fear, the instrumentalisation of fear in the political life, for instance in Sparta and Rome, and the association of fear with age (childhood), gender, culture, and education. The perception of emotions has traditionally been studied with the help of philosophical treatises, especially the work of Aristotle. Daniel King’s study of the construction of grief in the work of Galen (pp. 251–272) shows the necessity to combine the study of philosophical treatises with that of other sources, in order to understand how perceptions of emotion developed in intellectual circles and were diffused beyond them. The historical context of the representation of emotions in Greek art is discussed in three essays. Olympia Bobou (pp. 273–311) identifies specific markers for the representation of emotions (body language, facial expressions) and studies their evolution from the late Archaic to the late Hellenistic period. The display of emotions and emotionality were determined by a variety of factors, such as the identity of the figures (mythological, divine, human), age, and gender. The importance of status is stressed by Jane Masséglia (pp. 313–330), who explores how emotional expression in Hellenistic art correlated with the status of the individual depicted but also with the place in which works of sculpture were set up. While emotional restraint was connected with elite status and education, especially in honorific statues, low-status figures, whose statues did not have this honorary function, visibly show emotions. The study of emotions in antiquity requires the full exploitation of the source material – beyond the ‘usual suspects’, that is, philosophy and drama (see note 7). In this volume, the range of sources – literary sources (historiography, medical authors, Greek and Latin poetry and oratory), inscriptions, and the visual arts – is complemented with a study that approaches the difficult question of whether the emotional background of material culture can be reconstructed. With skeletal remains of infants and children as her starting point, Chrysa Bourbou explains both the difficulties in studying the emotional context of children burials but also the possibilities opened by new developments in bioarchaeology (pp. 331–350). To understand emotions as a historical phenomenon means to study the impact of specific historical conditions on the manifestations of emotions in texts, images, and in material culture. This be done both with diachronic surveys of sources (pp. 85–133, 191–214), and with the study of particular contexts, such as those of the Hellenistic world, the late Roman Republic, and the Imperial period. We hope that the variety of approaches, sources, and subjects assembled in this volume endorses a statement made in the volume Unveiling Emotions:15 Historians have to study emotions, because emotions have shaped all the source material that they have at their disposal. Therefore, the ancient historian does not only – perhaps not even primarily – study texts in order to understand emotions. It is far more urgent for an ancient historian to study emotions in order to understand texts.

15

Chaniotis 2012b, 23.

14

Angelos Chaniotis and Pierre Ducrey

BIBLIOGRAPHY Chaniotis, A. (2005) War in the Hellenistic World: a Social and Cultural History, Malden/Oxford. ––– (2011) Emotional Community through Ritual. Initiates, Citizens, and Pilgrims as Emotional Communities in the Greek World, in A. Chaniotis (ed.), Ritual Dynamics in the Ancient Mediterranean: Agency, Emotion, Gender, Representation, Stuttgart, 264–290. ––– (ed.) (2012a) Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, Stuttgart. ––– (2012b) Unveiling Emotions in the Greek World. Introduction, in Chaniotis (ed.) 2012a, 11– 36. ––– (2012c) Moving Stones: The Study of Emotions in Greek Inscriptions”, in Chaniotis (ed.) 2012a, 91–129. ––– (2013) Under Siege: Challenges, Experiences, and Emotions, in B. Campbell and L. Tritle (eds.), The Oxford Handbook of Warfare in the Classical World, Oxford, 438–456. ––– (forthcoming) Creating Emotional Community Through Text, in E. Sanders and M. Johncock (eds.), Emotion and Persuasion in Classical Antiquity. Ducrey, P. (1985) Guerre et guerriers dans la Grèce antique, Fribourg [revised edition, Paris 2009]. Ervin, M. (1963) A Relief Pithos from Mykonos, Archaiologikon Deltion 18, 37–75. Harris, W. V. (2001) Restraining Rage: the Ideology of Anger Control in Classical Antiquity, Cambridge, Ma./London. Hitzer, B. (2011) Emotionsgeschichte – ein Anfang mit Folgen, Humanities – Sozial- ud Kulturgeschichte 23 November 2011 (http://hsozkult.geschichte.hu-berlin.de/forum/2011-11-001>). Kalimtzis, K. (2012) Taming Anger: The Hellenic Approach to the Limitations of Reason, London. Kaster, R. A. (2005) Emotion, Restraint, and Community in Ancient Rome, Oxford. Knuuttila, S. (2004) Emotions in Ancient and Medieval Philosophy, Oxford/New York. Konstan, D. (2006) The Emotions of the Ancient Greeks, Toronto. Kotsifou, C. (2012) Emotions and Papyri: Insights into the Theatre of Human Experience in Antiquity, in Chaniotis (ed.) 2012a, 39–90. Masséglia, J. (2012) Emotions and Archaeological Sources: a Methodological Introduction, in Chaniotis (ed.) 2012a, 131–152. Munteanu, D. L. (ed.) (2011) Emotion, Genre, and Gender in Classical Antiquity, London. Plamper, J. (2012) Geschichte und Gefühl. Grundlagen der Emotionsgeschichte, Munich. Raaflaub, K. and N. Rosenstein (eds.) (1999) War and Society in the Ancient and Medieval Worlds. Asia, The Mediterranean, Europe and Mesoamerica, Cambridge, Mass./London. Reddy, W. (2001) Navigation of Feeling, Cambridge. Rosenwein, B. H. (2006) Emotional Communities in the Early Middle Ages, Ithaca. Sanders, E. (2012) Beyond the Usual Suspects: Literary Sources and the Historian of Emotions”, in Chaniotis (ed.) 2012a, 151–173. Sihvola, J. and T. Engberg-Pedersen (1998) The Emotions in Hellenistic Philosophy, Dordrecht. Theodoropoulou, M. (2012) The Emotion Seeks to be Expressed: Thoughts from a Linguist’s Point of View, in Chaniotis (ed.) 2012a, 433–468. Zaborowski, R. (2012) Some Remarks on Plato on Emotions, Mirabilia 15/2, 141–170.

EMOTIONS AND HISTORICAL REPRESENTATION IN XENOPHON’S HELLENIKA Melina Tamiolaki 1 INTRODUCTION1 The place of emotions in history is a controversial topic, linked both with the history of History (in the sense of the methods, principles and priorities of a scientific discipline) and with what could be tentatively called the ‘history of emotions’. 2 The representatives of the Annales School (Marc Bloch and Lucien Febvre) were among the first to shift attention from the study of politics to the history of mentalities (histoire des mentalités) as a central object of historical studies.3 Emotions played an important role in this new direction of historical research: Lucien Febvre, in a well-known study, emphasised the importance of studying ‘la vie affective d’autrefois’.4 The connotations and impact of this suggestion have triggered an ongoing debate about the place of emotions in history; the theory of the Annales School has been expanded, modified, refined, or even contested.5 A significant trait of this debate is that it departs from or is mainly concentrated upon the period of the Middle Ages. Attempts have been made either to trace the beginning of the history of emotions in that period or to connect this pe-

1

2

3 4 5

The present project was supported by a scholarship from the programme The Social and Cultural Construction of Emotions: the Greek Paradigm (Oxford University). I am grateful to Prof. Angelos Chaniotis for his meticulous reading of my work and his suggestions. I also thank Michael Flower and Tim Rood for further comments on the revised version. Τranslations of Xenophon’s Hellenika are from Strassler and Marincola 2009, sometimes adapted. For the topic of the history of emotions, see the illuminating overview by Hitzer 2011. I subscribe, however, to the reserve expressed by Chaniotis 2012b, 15: ‘the term “history of emotions” ... might be somewhat misleading ... But even if emotions might not have a history, history certainly has emotions.’ Cf. the interviews of W. Reddy, B. Rosenwein, and P. Stearns in Plamper 2010, esp. pp. 249, 260. See also below. For the Annales School, see MacMullen 2004, 107–127; Prochasson 2008, 53–57; Frevert 2011, 28–30. Febvre 1939–41. Rosenwein 2002 offers an excellent survey of the place of emotions in historical studies; see esp. 821–823, her criticism of the approach of the Annales. Rosenwein 2006 is a fine example of how historians should deal with emotions of a certain period. Cf. also the comparative review of Reddy 2001 and Rosenwein 2006 by Nagy 2007.

16

Melina Tamiolaki

riod with one particular emotional attitude.6 Antiquity plays a minor (or zero) role in this debate,7 which is explained by the relative paucity of evidence concerning the ancient world.8 Ancient texts can hardly answer questions such as ‘how did people of that period really feel’,9 or ‘how could a certain era be characterised on emotional terms’. This fact, however, essentially undermines the whole conception and edifice of writing a history of emotions: if we cannot write a history of emotions for the ancient world, then we are not legitimised to compensate for this inadequacy by an exclusive focus on the Middle Ages or even the modern era (claiming, for example, that the Middle Ages correspond to the ‘childlike period of humanity’ is like claiming that the history of humanity begins in the Middle Age).10 What about studying emotions in the ancient world, then? What kind of questions do we expect to answer? At least four issues arise from the study of emotions in ancient texts: — the emotions (described or inferred) of the agents of the ancient texts, — the emotions ancient writers intend to arouse in the recipients of their texts (be it readers or hearers), — the emotions of the writers themselves,11 and 6

7

8

9 10

11

For example, Febvre 1939–41 is inspired by Johan Huizinga’s book Herbst des Mittelalters (Huizinga 1924), in which the author initiates the idea of a childlike nature of the medieval emotional life. Norbert Elias’ influential book Über den Prozess der Zivilisation (Elias 1976), also follows a similar line of thought, by claiming that humanity underwent an evolution from a vibrant expression of emotions (represented, for instance, in the Middle Ages), towards a greater restraint. Cf. also more recently, Reddy 2001, who starts his discussion on ‘emotions in history’, this time not from the Middle Ages, but from France of the eighteenth century. See again Rosenwein 2002, 826-828; Frevert 2011, 27–29. Rosenwein 2002, 827 note 30, cites some works about emotions in Greek and Roman antiquity. Interestingly, her criticism of Harris 2001 for contributing to the bracketing of the Middle Ages in his study of anger in ancient Greece, ‘arguing that emotional control existed in the ancient world and then again in the sixteenth century’, epitomises the complexity and the limitations of the project of writing a history of emotions. It seems more feasible and more productive to examine emotions in a certain period and place. Prochasson 2008, 27–31, borrows an expression inaugurated by Charles Péguy, concerning the distinction of ancient and modern historians: what characterises the modern historian, according to Péguy, is ‘le manque du manque’, in the sense that there is an abundance of evidence on which he/she can rely. Exceptions are inscriptions and papyri, which can indeed provide more ‘lively’ evidence about emotions in specific occasions. See Chaniotis 2012c; Kotsifou 2012. The same problem is revealed in Frevert 2009, who makes a passing reference to the ancient world, in order to claim that trust is a modern invention. Yet, her explanation is far from satisfactory, since she seems to admit that trust in the modern context and with modern connotations is indeed a modern phenomenon (which is self-evident), without really addressing the problem of whether it is linked in some ways to ancient trust. On Greek and Roman concepts of trust, see T. Morgan’s study in this volume (pp. 191–212). Cf. also Harris 2010, 19 note 87: ‘we should not return, and have no need to return, to the sort of broad undifferentiated emotional characterization of whole epochs…’ Latin historiography provides more evidence about this. See for example, Tacitus, Historiae, 1.1.1: sed incorruptam fidem professis neque amore quisquam et sine odio dicendus est. Cf.

Emotions and Historical Representation in Xenophon’s Hellenika

17

— the interaction (interdependence or divergence) of the previous three factors. For example, the suffering of an unjust or hubristic person depicted in a Greek tragedy may arouse the pleasure of the reader; or an attempt on the part of a Greek historian to praise unconditionally an actor of his text may inspire a variety of responses in the readers, ranging from admiration to suspicion or even annoyance.12 A separate and important issue regarding the study of emotions in the ancient world concerns what I would call ‘the shadow of Aristotle’. Aristotle provided a penetrating analysis of certain fundamental emotions in the second book of his Rhetoric. More importantly, his analysis made a ground-breaking connection between emotion and cognition and also stressed the importance of social context and status for their definition and circumscription; he thus foreshadowed not only modern developments of psychology, but also (what is now labeled as) ‘social constructionism’ in the study of emotions.13 It is no wonder then that modern scholarship has often dealt with the topic of defining to what extent ancient texts confirm, adapt, or even challenge Aristotle’s analysis of the emotions.14 Although this line of approach has borne important results, it is reductionist for a number of reasons. Firstly, Aristotle does not include all the emotions in his analysis and this, if applied strictly to the ancient texts, may cause bewilderment: for example, concerning grief and pleasure, which according to Aristotle are sensations rather than emotions (πάθη), should we include or exclude them in our analysis of texts? Secondly, Aristotle’s Rhetoric is designed to help public speakers to persuade for their causes. Nevertheless, ancient texts show emotions in private contexts as well. Moreover, as Christoph Rapp has emphasised in his fine commentary, emotions are not the most important part in Aristotle’s analysis of persuasion, in comparison, for instance, with judgments (ἐνθυµήµατα).15 Therefore, focusing exclusively on the emotions runs the risk of missing the overall purpose and content of Aristotle’s Rhetoric. Finally, Aristotle’s Rhetoric has a clear moral dimension (1355a31: οὐ γὰρ δεῖ τὰ φαῦλα πείθειν); on the contrary, ancient texts show that speakers may want to persuade ‘for evil’ as well. For these reasons, in my opinion, the study of the emotions in the ancient world should view Aristotle’s Rheto-

12 13

14

15

also MacMullen 2004, 53: ‘on conviendra sans peine que l’historiographie latine conservée accorde plus de place que l’historiographie grecque aux émotions.’ For illuminating discussions about the issues raised by the study of emotions in the ancient world, see Chaniotis 2012b; Sanders 2012. See Fortenbaugh 2002, for the importance of cognition in Aristotle, and the essays collected in Harré (ed.) 1986 for social constructionism. The bibliography on emotions from a psychological perspective is abundant: see, for instance, Oatley 1992; Elster 1999; Ben-Ze’ev 2000; Borod 2000; Frijda, Manstead, and Bem 2000; Dixon 2003; Franks 2010. Konstan 2006 is valuable for being the first to offer a systematic analysis of emotions in Aristotle, in combination with ancient Greek literature, although he often strives to prove that ‘Aristotle got it right’ in his descriptions of emotions. Cairns 2003, more convincingly, talks about Aristotle providing a ‘prototypical scenario’ about the emotions, the examples from Greek literature being its variations. Rapp 2002, 343–345.

18

Melina Tamiolaki

ric with greater caution.16 In the analysis which follows, although I will take Aristotelian definitions of emotions into account, I will not limit myself only to the Aristotelian emotions.17 The above issues obviously concern all ancient literary genres (epic, tragedy, philosophy, rhetoric). Yet the case of historiography seems thornier, and Polybios may be held responsible for making matters more complex: by severely criticising the Hellenistic historian Phylarchos for the inclusion of emotions in his history, and by negatively comparing history with tragedy,18 he created a standard according to which the inclusion of emotions in historiography should be considered suspect. Ironically, however, historians of all periods included emotional descriptions and scenes of pathos in their works, not least Thucydides, who was Polybios’ model.19 Polybios himself assigns an important role to the emotions in his history as well.20 It seems, then, that the whole question is a matter of degree: historians should not exaggerate in the depiction of emotions, to the extent that this depiction ends up obstructing the ultimate purpose of history, which is to serve the truth. But there is no doubt that the definition of this limit is a subjective and ambivalent matter, as is the topic of ‘truth’ in historiography generally.21 Consequently, it would be more fruitful to dissociate the study of the emotions in ancient historiography from the topic of ‘truth’, and rather focus on the analysis of emotions as bearers, indicators, or signposts of historical explanation. This is not to say that I am eschewing discussion of whether emotions should or should not be included in the study of history.22 It only means that, given the recent development of emotion studies, I consider this discussion rather outdated. What needs to be further studied and explored, in my opinion, is in what ways ancient historians employ and on what occasions have recourse to emotions (and to which emotions, more specifically, and for which reasons), in order to convey historical meaning.23 16

17 18 19

20 21 22

23

The role Aristotle’s Rhetoric should (or should not) play in the study of emotions in the ancient world has greater connotations than those briefly outlined here. I intend to return to this issue in more detail on another occasion. For an illuminating analysis of how we should study emotions in the ancient world see Cairns 2008. Polybios 2.56. Cf. Walbank 1972, 34–40 and 1990; Gray 1987; Walker 1993; Marincola 2003; see also the chapter by Angelos Chaniotis in this volume (pp. 53–84). Already Cornford 1907 shows the connections of Thucydides with contemporary tragedy, concerning topics of vividness and descriptions of pathos, an idea which has been endorsed and further elaborated by more recent Thucydidean scholarship. Cf. Erskine 2014. For the topic of ‘truth’ in ancient historiography, see Tamiolaki forthcoming b, which contains references to previous bibliography. For example, MacMullen 2004 often adopts a polemical attitude in order to support his view that emotions should play an important role in the study of history. On the other hand, Prochasson 2008 examines the problems and complexities of this approach. I offered some preliminary suggestions on this topic in Tamiolaki 2013b, concerning motivation in Thucydides. But the topic of the emotions in historiography is generally underexplored (it is absent, for instance, from Munteanu, ed., 2011).

Emotions and Historical Representation in Xenophon’s Hellenika

19

We can now turn to Xenophon, a prolific writer, with a great inclination towards innovation: he inaugurated new ideas, while at the same time building upon previous tradition, and he created new genres, while at the same time renewing older ones. Concerning the topic of emotions, his work provides ample material: Cyropaedia could be considered his most ‘emotional’ work, not only because of the abundance of terms used there to describe emotions, but also because of a love story which is inserted in it, the famous romance of Panthea.24 If the presence of this romance, (which, as has often been suggested, foreshadows the ancient novel),25 reflects a kind of ‘emotional turn’ in the fourth century BCE,26 the study of emotions in Xenophon’s historical works could open up interesting perspectives as well: Xenophon was constantly experimenting with genres (historiography included); analysing the role of emotions in the Hellenika, which is the focus of the present study, essentially means studying emotions as a vehicle of historiographical innovation. Obviously Xenophon was not the first to employ emotions in his history. Already Herodotos and Thucydides, his most important predecessors, often had recourse to emotions, in order to explain the motivation of the agents of their histories, or provide characterisation. A central concern of this study is to examine how Xenophon acquired and expanded the heritage of his predecessors regarding the depiction of emotions.27 My paper is divided into three parts. The first part focuses on the vocabulary of emotions employed by Xenophon. In the second part I analyse certain episodes, in which emotions play an important role and are used to promote a specific historical representation. The third part is devoted to Xenophon’s innovations concerning the emotions. Finally, in a concluding section I attempt to sketch out a comparison between Xenophon and Hellenistic historiography. In the course of my study, I occasionally employ terms deriving from the modern study of emotions, such as ‘emotional episode’, ‘emotional community’, ‘emotives’, and ‘emotionology’.28 Most of these terms are based on experimental 24 25 26 27

28

For the story of Pantheia in the Cyropaedia, see Romilly 1988; cf. Tamiolaki 2010, 305–309 (for a political interpretation of this episode). Tatum 1989, ‘Introduction’. For the ‘emotional turn’ in the Hellenistic period, opposed to the fifth century, see Konstan 2006, 29–31. I limit my comparison to Herodotos and Thucydides, because Xenophon’s intertextual dialogue with them is more apparent. See Rood 2004; Nicolai 2006; Tamiolaki 2008 and forthcoming a; Baragwanath 2012. An ‘emotional episode’ is an event in which emotions are involved. See Sanders 2012, 157 and cf. also below. The term ‘emotional community’ was coined by Rosenwein 2002; cf. Rosenwein 2006. See, for example, the definition in Rosenwein 2002, 842: ‘these are precisely the same as social communities-families, neighborhoods, parliaments ... – but the researcher looking at them seeks above all to uncover systems of feeling: what these communities (and the individuals within them) define and assess as valuable or harmful to them. ... I further propose that people move (and moved) continually from one such community to another ... adjusting their emotional displays ... to these different environments.’ For qualifications of this theory, see Chaniotis forthcoming, who highlights also ‘emotional disunities’. The term ‘emotive’ was inaugurated by Reddy 2001, 96–111. It is based on the theory of

20

Melina Tamiolaki

psychology and should, therefore, have limited application in literary studies. I think, however, that their cautious use might lead to a better definition and delimitation of topics concerning the emotions. More importantly, if history imitates real life, as it has often been observed,29 then using real life terms in the study of a historical text ultimately contributes to a better understanding of the effect of historical mimesis.30 2 TRANSLATING EMOTIONS IN XENOPHON’S HELLENIKA31 In her stimulating study about emotional communities in the Middle Ages, Barbara Rosenwein talks about generic emotions;32 that is, emotions which are more suitable for and dictated by the norms of a particular genre. John Marincola, in a study devoted to the emotions in ancient historiography, initially focuses on pity and fear (the most important emotions tragedy should arouse and subsequently the basic emotions of tragic historiography), but he suggests that the study of emotions in historical texts should not be limited to these two emotions.33 Indeed, it is difficult – if not impossible – to distinguish ‘historical emotions’ in ancient Greek literature: ancient historiography interacted fruitfully with other literary genres (mainly epic poetry and tragedy), and was also influenced by overarching rhetorical principles which affected all genres,34 so the distinction between ‘tragic’ and ‘historical’ emotions is blurred. Besides, the focus on fear and pity is misleading: although fear could indeed be considered a ‘historical’ emotion,35 it is remarkable that pity plays a relatively minor role in Greek historiography,36 whereas other emotions (such as, for example, anger) definitely occupy a more prominent place.

29

30 31 32 33 34 35

36

speech act (performatives) and refers to people’s statements about their emotions. Finally, the term ‘emotionology’ was inaugurated by Peter and Carol Stearns (Stearns and Stearns 1985) and describes what people think about the emotions. It has also to be noted here, that the term ‘emotion’ used throughout this study is a convenience. There are subtle distinctions between terms such as emotion, passion, and feeling in all Western languages. See Febvre 1939–41, 5–8; Rosenwein 2006, 3–5; Schwarz-Friesel 2007, 138–173. See Gray 1987, for historical mimesis, and Walker 1993 for enargeia in Greek historiography. Cf. Allan 2013 and Grethlein 2013 for sophisticated analyses of these techniques in Thucydides. Cf. Kuzmics 2009; Vendrell Ferran 2010. I borrow the expression ‘translating emotions’ from Konstan 2003a. See, however, the important qualifications proposed by Cairns 2008. Rosenwein 2006, 27-28. Marincola 2003, 287: ‘the nearly exclusive attention given to pity and fear practically blinded scholars to the whole world of emotion that is delineated in ancient historians.’ See Woodman 1988, for the influence of rhetoric in historiography and Sauge 1993, for the influence of epic poetry, and Walbank 1960, for history and tragedy. Although Konstan 2006 does not often analyse historical texts, he devotes roughly ten pages (135–144) to fear in Greek historiography. For fear in Thucydides cf. also Romilly 1956 and Huart 1968, 337–346. See Konstan 2001, 75–104; Lateiner 2005; Sternberg 2005b, 25.

Emotions and Historical Representation in Xenophon’s Hellenika

21

Given this framework, in this part of my paper I will provide an overview of the most significant emotions attested in Xenophon’s Hellenika, expressed either by the relevant vocabulary or by facial expressions and gestures.37 As I mentioned above, my aim is to examine whether Xenophon differentiates from his main predecessors, Herodotos and Thucydides, in the depiction of emotions. Consequently, I will not deal with emotions, which are invariably present in the three historians;38 I will focus instead on two aspects: first, emotions which are common (among the three historians or between Xenophon and one of them) and in which Xenophon presents a variation/differentiation, and second, emotions which are Xenophon’s own emphasis. This direction of research should produce interesting results concerning Xenophon’s priorities and innovations. The Hellenika contain a great variety of terms expressing dissatisfaction/discomfort/distress: ἀγανακτῶ, ἀθυµία, ἀνιῶµαι, ἄχθοµαι, δυσχεραίνω, χαλεπαίνω, χαλεπῶς φέρω.39 Among these terms, δυσχεραίνω is Xenophon’s peculiarity; ἀνιῶµαι is absent in Thucydides, whereas ἀγανακτῶ and ἀθυµῶ are not attested in Herodotos.40 Moreover, Xenophon employs the adverb ἀχθεινῶς, which is a hapax legomenon in his work and not attested in his predecessors.41 The same tendency towards variation can be observed in the vocabulary of joy as well.42 Xenophon not only has recourse to a great variety of terms expressing joy (ἀγάλλοµαι, εὐθυµοῦµαι, εὐφραίνοµαι, ἥδοµαι, χαίρω), but he also uses the prefix ὑπερ- twice in the relevant vocabulary: ὑπερήδοµαι, ὑπερχαίρω.43 Of 37

38

39

40 41

42 43

In this part of my paper, I am interested in the vocabulary of emotions, so passages from the Hellenika will be presented out of their context. For gestures in ancient Greece, see Neumann 1965, Cairns (ed.) 2005. Cf. Halliwell 1990, 38, for depictions of psychological experience in ancient texts. Fear is such an example. Xenophon amply employs verbs related to fear (φοβοῦµαι, δέδοικα, ὀκνῶ, ἐκπλήττω), but with less sophistication than Thucydides: for example, Thucydides additionally employs the verb ὀρρωδῶ and uses fascinating expressions such as δεδιὼς τὴν ὀρρωδίαν (2.88.1). Other examples include hope (ἐλπίς) – although references to hope in the Hellenika are remarkably few (3.5.1, 4.6.14, 5.4.42, 6.3.20, 6.5.35, 7.2.10), in comparison with their abundance in his main predecessors (see Powell 1966 and Bétant 1969, s.v.) – as well as shame (αἰδώς, αἰσχύνη), mainly invoked in the speeches of the historians. But this concept does not have a prominent role in the historians, as Cairns 1993, viii, notes. Here and below, terms are put in alphabetical order and not according to their frequency in Xenophon’s text. For their differentiation in meaning, see LSJ, s.v., although the lexicon does not render all the nuances. Schematically, if we replace Russel’s circumplex (Russel 1980, 1989) by a pyramid, the milder terms ἀνιῶµαι, ἀθυµῶ, and δυσχεραίνω would occupy the base of the pyramid, ἄχθοµαι and χαλεπαίνω its middle part, and the more intense terms ὀργίζοµαι and ἀγανακτῶ the summit. Ἀθυµίη in Herodotos means cowardice (1.37.2), it does not express an emotion. Hellenika 4.8.27. Τhere are only two other occurrences of this adverb, in Flavius Josephus, Antiquitates Judaicae 18.218 and in Pollux, Onomasticon 3.99. A similar example is the noun ἀχθηδών employed by Thucydides (2.37.3) and found again only in post-classical literature. For variation as an essential trait of Xenophontic language, see Gautier 1911, 117-129. If we used the above pyramid (note 39) for terms of joy, the base would be covered by the terms χαίρω, εὐθυµοῦµαι, the middle by the terms εὐφραίνοµαι, ἀγάλλοµαι and the summit by the verb ἥδοµαι. Moreover, it is interesting that the word ὑπερχαρής also appears in Hellenistic inscriptions. See Chaniotis 2012b, 93 note 11.

22

Melina Tamiolaki

these terms, εὐθυµοῦµαι is unattested in Herodotos and Thucydides, whereas εὐφραίνοµαι is attested only once in Herodotos.44 Concerning the verbs beginning with ὑπερ-, it seems that the inclination towards emotional exaggeration has a Herodotean flavor: Herodotos uses the terms ὑπερήδετο (1.90.1, 1.54.1, 3.22.3), ὑπερδείσας (8.94.1), ὑπερδειµαίνων (5.19.1), ὑπεραλγήσας (2.129.3), ὑπεραρρωδέω (8.72), ὑπεραχθεσθέντες (6.21.2), ὑπερλυπέοµαι (8.90.3). Other linguistic variations can be observed in the depiction of anger and hatred. Concerning the former, Xenophon often employs the verb ὀργίζοµαι and the noun ὀργή, but he also uses the rare verb ἔµηνε (aorist of the verb µαίνω) once, again a hapax in his work and unattested in his historical predecessors.45 The passage describes Agesilaos’ anger for Lysander’s growth of power: ... a huge mob was always paying court to Lysander and following him around, so much that it seemed as if Agesilaos was but a private citizen and Lysander the king. Agesilaos later made it very clear that this state of affairs had caused his rage (ὅτι µὲν οὖν ἔµηνε καὶ τὸν 46 Ἀγησίλαον ταῦτα ἐδήλωσεν ὕστερον).

The verb ἔµηνε is a poetic word, attested in tragedy and comedy,47 and obviously linked with the Homeric µῆνις. It has been observed that µῆνις refers to divine wrath, but in a careful analysis Douglas Cairns has shown that this word expresses rather a higher intensity of the emotion.48 Why does Xenophon use this word in this context? I would like to suggest that this choice is intentional. According to Aristotle, anger is aroused by an infliction on one’s honour and is accompanied by a desire for retaliation.49 It is no wonder that Aristotle considers Achilles a characteristic example of anger: Achilles feels insulted by Agamemnon and desires to take revenge for his attitude.50 Xenophon did not know Aristotle’s analysis, but by using the verb ἔµηνε, he wishes to distance himself from the more usual term ὀργή (which, moreover, in historiography, does not always fit with the Aristotelian definition)51 and establish a connection between Agesilaos and Achilles.52 44 45 46 47 48 49 50 51

52

Herodotos 4.9.5. Herodotos uses the verb µαίνοµαι, which means ‘go mad’ and is not always linked with anger. Hellenika 3.4.8. Cf. also below p. 25 for this episode. Euripides, Ion 520; Aristophanes, Thesmophoriazousai 561. Cf. Sophocles, Antigone 790: µέµηνεν. Cairns 2003. Aristotle, Rhetoric 1378a2. See Grimaldi 1988, ad loc.; Rapp 2002, ad loc.; Konstan 2006, 45–56. Aristotle, Rhetoric 1378b2. It has to be noted on this point that ὀργή is often irrational in Greek historiography. In Thucydides, it is constantly opposed to γνώµη. See Huart 1968, 50–57. Given this tradition (or historiographical convention), Xenophon might have more reasons to use a different word here that could denote the offence and the retaliation. It is not the only occurrence in which Agesilaos is compared with a Homeric hero. Before his expedition in Asia Minor, Xenophon reports that he sacrificed in Aulis, ‘as it was the very place where Agamemnon had sacrificed before sailing to Troy’ (Hellenika 3.4.3). See Dillery 1995, 23-24, 107, 116, who offers further suggestions why Xenophon included this information about Agesilaos in the Hellenika and not in Agesilaos.

Emotions and Historical Representation in Xenophon’s Hellenika

23

The similarities implied concern the following topics: in both cases (the Iliad and the Hellenika), two individuals are involved; in both cases there is a perception of insult, and feelings of inferiority/superiority and status come into play;53 in both cases, anger is intense, but not expressed immediately and instead it evolves (in the Iliad, throughout the poem, in the Hellenika, throughout the episode); in both cases, retaliation follows: Achilles abstains from battle, whereas Agesilaos refuses to pay services to the people Lysander sends him. In sum, the word ἔµηνε in the Hellenika creates an ambiance of strong offence and justifies Agesilaos’ subsequent actions. Concerning the emotion of hatred, along with the usual terms µισέω and µῖσος Xenophon also employs a powerful metaphor in the Hellenika,54 describing the emotion of helots and other underprivileged groups, which revolted at Sparta with Kinadon: For whenever they conversed with such men about the Spartiates, they said that not even one of the men in those groups could hide the fact that they would even eat them raw with pleasure (ὅπου γὰρ ἐν τούτοις τις λόγος γένοιτο περὶ Σπαρτιατῶν, οὐδένα δύνασθαι κρύπτειν 55 τὸ µὴ οὐχ ἡδέως ἂν καὶ ὠµῶν ἐσθίειν αὐτῶν).

If we follow the definition of Aristotle, according to whom hatred is felt towards types rather than individuals,56 this is certainly a radical expression of hatred: the underprivileged groups of Sparta form an emotional community bound by their common hatred (as a desire to inflict harm), an emotion directed against the whole group of the Spartan homoioi. Moreover, this expression, the only emotional expression in the whole narrative of the event, which is embedded in the speech of the denunciator of the conspiracy,57 functions paradigmatically and vividly underlines the dramatic situation in Sparta. Linguistic innovations can also be observed in the terms related to tears and laughter. Tears and laughter are both present in the three historians,58 but Xenophon additionally uses the term κλαυσίγελως, another hapax in his work.59 Moreover, laughter in Xenophon really derives from good temper, while in Herodotos it foreshadows disaster and in Thucydides it expresses sarcasm.60 Further53

54 55 56 57 58 59 60

Agesilaos feels insulted because Lysander behaves more ‘regally’ than he deserves. In this case, the cause of anger seems more justified, since Lysander’s actions suggest a transgression of his status. On the contrary, Achilles seems to challenge the privileges assigned to Agamemnon as a king. See also below. For metaphors as powerful means of expressing emotions, see Chaniotis 2012b, 112 (and the examples given there from inscriptions). Hellenika 3.3.6. Cf. Xenophon, Anabasis 4.8.14. Aristotle, Rhetoric 1382a1–14. For the conspiracy of Kinadon and the historical and historiographical issues raised by it, see Tamiolaki 2010, 276–279, 411–415 (with previous bibliography). See Lateiner 2009, for tears in Greek historiography. Hellenika 7.2.9. Hellenika 4.5.9; Thucydides 4.28.5, 6.35. The references are more numerous in Herodotos, see Powell 1966, s.v. Cf. Radermacher 1947 for laughter in ancient Greece, who, however, does not deal with historiography.

24

Melina Tamiolaki

more, Xenophon employs the verb πενθέω and the noun πένθος, attested only in Herodotos and not in Thucydides.61 Finally, Xenophon uses the words φαιδρός and σκυθρωπός, alluding to facial expressions, both of which are absent in his predecessors.62 He also uses the expression λαµβάνω τῆς χειρός,63 which indicates a gesture of friendship again not attested in his predecessors. I turn now to two terms which are used in new contexts by Xenophon. The first is ἔρως. Ἔρως denotes a strong sexual desire (for a man or a woman),64 but in historiographical texts it often expresses the imperialistic impulse.65 In Xenophon ἔρως is used twice in new contexts. Firstly, in the description of Agesipolis’ death: … while he was engaged in these activities ... he was seized by a burning fever. And since he has previously visited the temple of Dionysos at Aphytis, he now experienced a strong desire for its shady buildings and its bright cold waters (ὡς δὲ πρόσθεν ἑορακότα τὸ ἐν Ἀφύτει τοῦ ∆ιονύσου ἱερὸν ἔρως αὐτὸν τότ' ἔσχε τῶν τε σκιερῶν σκηνηµάτων καὶ τῶν λαµπρῶν καὶ ψυχρῶν ὑδάτων). He was carried there still alive, but six days after he contracted the fever, he died outside the temple.66

The detail devoted to the scene of Agesipolis’ death is remarkable and unfamiliar in conventional historiography (rather being invested with novelistic overtones). It is also interesting that, contrary to its use in previous historiographical texts, in this context the word ἔρως has peaceful and nostalgic connotations. Secondly, Xenophon uses a poetic syntax of the verb ἐρῶµαι (with an infinitive as its object).67 He describes the Athenians’ fight against the Mantineans in 362 BCE as follows: Who could fail to admire their bravery? For though they saw that the enemy coming against them were numerous, and though the cavalry had already suffered misfortune in Corinth, they gave no thought to this, nor to the fact that they were about to fight Thebans and Thessalians, men who had the highest reputation for horsemanship. Instead, they felt a sense of shame at the thought of being on the spot but failing to assist their allies. So as soon as they caught sight of the enemy, they charged them, motivated by a deep desire to win back their ancestral reputation (ἐρῶντες ἀνασώσασθαι τὴν πατρῴαν δόξαν).68

In this passage the participle ἐρῶντες obviously describes a strong desire, but again one can observe how Xenophon builds upon and renews previous historiographical tradition. Eros characterises Athenians par excellence: in the Funeral 61 62 63

64 65 66 67 68

Hellenika 2.2.3, 4.5.10. For the passages in Herodotos, see Powell 1966, s.v. Hellenika 3.4.11, 4.5.7. Darwin 1872 triggered the debate about the connection of facial expressions and emotions. Hellenika 4.1.15, 4.1.30, 4.1.37. Xenophon has a great interest in gestures. My student Mariyianna Lampadaridou (Lampadaridou 2013) has offered a useful collection of the gestures attested in the Cyropaedia. Eros does not figure in Aristotle’s analysis, but it certainly involves emotions. The most famous example is Thucydides 6.24.3, concerning the Sicilian expedition (καὶ ἔρως ἐνέπεσε τοῖς πᾶσιν ὁµοίως ἐκπλεῦσαι). Cf. also Herodotos 1.73.1: γῆς ἱµέρῳ. Hellenika 5.3.19. See Aristophanes, Frogs 44; Acharneis 146; Sophocles, Antigone 220. Hellenika 7.5.16.

Emotions and Historical Representation in Xenophon’s Hellenika

25

Oration, Pericles urges them to become ἐρασταί (lovers) of their city, and they are again driven by ἔρως in their expedition against Sicily.69 By using this word in a context of praise of the Athenians, Xenophon alludes to their imperialistic past, while expressing at the same time (perhaps with some irony) a sort of nostalgia about it. The second term which is used in a new context in the Hellenika is φθόνος. Its most famous use in Herodotos concerns divine envy (φθονερὸν θεῖον).70 In Thucydides the term φθόνος and its compounds are often used by the orators and refer to the envy aroused by the Athenian empire and power.71 In this context, arousing envy is considered something positive and a sign of power. Alkibiades also claims that his fellow citizens envy him,72 but this is not corroborated by Thucydides himself, who prefers to use more ‘neutral’ terms about the emotions of the Athenians towards Alkibiades in his narrative, such as φοβηθέντες, ἀχθεσθέντες.73 Xenophon, on the contrary, uses φθόνος in contexts which are closer to the Aristotelian definition of this emotion. According to Aristotle, φθόνος is a disturbing pain arising from the well-being of another, not because the other person is undeserving, but simply because he is our equal or similar.74 In the Hellenika, φθόνος occurs in hierarchical relationships between individuals and only twice describing a collective emotion.75 More interestingly, these occurrences appear in the narrative and not in the speeches, as is usually the case in Thucydides. To summarise, Xenophon’s innovations concerning the vocabulary of emotions point to the peculiar character of the Hellenika, if compared with its literary predecessors (the histories of Herodotos and Thucydides). It seems that Xenophon inherits from both, but does not hesitate to proceed to important adaptations. He shows a greater concern for vivid descriptions (what is defined in literary criticism as enargeia), as is proved by the great linguistic variety in the depiction of emotions, as well as by the presence of facial expressions and gestures. The descriptions of positive emotions, moreover, particularly joy, laughter, and relief testify to the less war-centred character of the Hellenika (in the sense of a greater emphasis on individual scenes and emotional episodes).

69 70 71

72 73 74 75

Thucydides 2.43.1, 6.24.3. For envy in Herodotos, see Harrison 2003. Thucydides 2.35.2, 2.45, 2.64.4–5. There are a few references to personal envy (3.43, 6.78.2– 3, in the speeches of Diodotos and Hermokrates respectively), one reference to divine envy (in the speech of Nikias, 7.77.4), and again only one reference in Thucydides’ narrative (4.108.7), concerning envy towards Brasidas. Cf. also Huart 1968, 397-399. Thucydides 6.16.3. Thucydides 6.15.4. Aristotle, Rhetoric 1386b18–20, with Konstan 2006, 112-114. Hellenika 2.4.29 (βασιλεὺς φθονήσας Λυσάνδρῳ), 3.2.13 (ὑπεφθόνει τοῖς στρατηγοῖς), 3.4.8 (ὑπὸ τοῦ φθόνου ἐσίγων). The Mantineans are presented envying the Spartans’ prosperity (5.2.2), a passage which echoes the envy aroused by Athenian power, while for the Thebans Xenophon states that ὑποφθόνως καὶ οὐκέτι φιλικῶς εἶχον πρὸς τοὺς Ἀρκάδας (7.1.26).

26

Melina Tamiolaki

3 EMOTIONS AND HISTORICAL REPRESENTATION: CONTINUING THUCYDIDES The investigation of vocabulary presented above is the first step, which helps us grasp Xenophon’s emphasis on the emotions and subsequently the particular character of the Hellenika. The second step is to examine in what contexts emotions are employed, in what ways, and for what purposes. We touch thus upon an important topic of Greek historiography: emotions as motives for actions, or motivation more broadly. Ramsay MacMullen has proposed an equivalence of emotion with motivation,76 but the issue is admittedly more complex, since the way historians (ancient and modern) present motivation is not always associated with emotions; the distinction between the intellectual and emotional element is difficult (since emotions also involve processes of evaluation),77 yet historical texts often stress intellectual rather than emotional factors as causes of decisions and actions. In the course of my analysis, I will not deal with intellectual motives. I will examine instead contexts in which one or more emotions constitute a factor of motivation. The topic of motivation in ancient historiography is old, but we can observe an important renewal of interest in recent years.78 An essential component of this recent debate is that it does not have as its starting point the positivist question ‘did ancient historians really know the emotions of the agents of their histories’, but rather attempts to examine the function and role of motivation as a means of historical explanation. Studies have so far focused on Herodotos and Thucydides: a basic technique of the former is that he often presents alternative motives for an action (by using the formula εἴτε … εἴτε),79 whereas the latter avoids giving many options to his readers about the motives of the agents of his history.80 Xenophon has attracted less scholarly attention and this is the gap that the present study aims to fill. Xenophon seems to follow Thucydides more closely on this topic. The only distinctively Herodotean motivation element that we find in the Hellenika is the phrase καὶ ὁ θεὸς δέ, ὡς ἔοικε, πολλάκις χαίρει τοὺς µὲν µικροὺς µεγάλους ποιῶν, τοὺς δὲ µεγάλους µικρούς (‘god, it seems, often delights in making the small great and the great small’),81 which recalls the famous φθονερὸν θεῖον. Yet there is a subtle differentiation from Herodotos: this phrase is uttered in a speech and not in the main narrative; in fact, divine envy does not pervade the Hellenika as an essential cause of events, as is the case in Herodotos’ history.82 On the con76 77 78 79 80

81 82

MacMullen 2004, 23-24, 70-71. See Nussbaum 2001 and more recently von Scheve 2011 (with previous bibliography). Montgomery 1965; Westlake 1968; Schneider 1974; Baragwanath 2008; Tamiolaki 2013b. For motivation in Herodotos, see Baragwanath 2008. The history of the Thucydides is characterised by a tension between awareness of the difficulty to reconstruct emotions and sophisticated, truthful presentations of emotions. See in detail, Tamiolaki 2013b. Hellenika 6.4.23. For Herodotos, see Harrison 2003. For Xenophon’s religious views, see Pownall 1998.

Emotions and Historical Representation in Xenophon’s Hellenika

27

trary, Xenophon, borrowing rather from the Thucydidean tradition, extensively presents thoughts and emotions as causes of actions. Although his descriptions of emotions as motivation are not as sophisticated as Thucydides’,83 there is no doubt that his presentation aims to convey historical meaning and to promote specific historical interpretations. In what follows, I will present three of Xenophon’s techniques concerning the presentation of emotions as factors of motivation: a) repetition, b) perceived or focalised emotion, and c) appeal to emotions in speeches. 3.1 Repetition of emotions I begin with repetition of emotions.84 Repetition can be observed both in the speeches and in the narrative. Its function in the speeches is more obvious, since it considerably increases the weight of emotional appeal,85 but its presence in the narrative deserves more careful exploration. The most conspicuous examples in the Hellenika, in which the repetition of emotions is attested, concern two episodes: the attack of the Lacedaemonians against the Eleans around 400 BCE86 and the raid of Sphodrias against Attica in 378 BCE. For the first episode, Xenophon reports:87 The Spartans had long been angry (πάλαι ὀργιζόµενοι) with the Eleans because the latter had made an alliance with the Athenians, Argives, and Mantineans. In addition, the Eleans, alleging that they had obtained a judgment against the Spartans, had forbidden them from taking part in the horse races and the athletic contests. Not content with this, they offered further insult: when Lichas handed over his chariot to the Thebans and they were announced as the winners of the race, he entered the stadium intending to crown the victors. But the Eleans set upon him and whipped him – although he was an old man – and drove him from the games. 83

84 85 86 87

For example, a trait of the Thucydidean motivation is the interweaving of multiple elements and explanations: e.g., Thucydides 7.86.5, for the death of Nikias: ἀλλὰ τῶν Συρακοσίων τινές, ὡς ἐλέγετο, οἱ µὲν δείσαντες, ὅτι πρὸς αὐτὸν ἐκεκοινολόγηντο, µὴ βασανιζόµενος διὰ τὸ τοιοῦτο ταραχὴν σφίσιν ἐν εὐπραγίᾳ ποιήσῃ, ἄλλοι δέ, καὶ οὐχ ἥκιστα οἱ Κορίνθιοι, µὴ χρήµασι δὴ πείσας τινάς, ὅτι πλούσιος ἦν, ἀποδρᾷ καὶ αὖθις σφίσι νεώτερόν τι ἀπ' αὐτοῦ γένηται, πείσαντες τοὺς ξυµµάχους ἀπέκτειναν αὐτόν (‘but since some of the Syracusans had been in communication with him, they were afraid, it is said, that if questioned under torture on such grounds, he would confound them in the midst of their success, while others, especially the Corinthians, were afraid that by bribing people, since he was indeed wealthy, he would escape, and there would be trouble for him once again, and they persuaded their allies and put him to death;’ translated by Lattimore 1998). Here, Thucydides not only gives the motivation but also offers additional explanations for this motivation (see the phrases in italics). We do not find such complex sets of motivation in Xenophon. For the explanatory power of emotions in history, see also Harris 2010, 16–19. Cf. Stueber 2008, for the explanatory power of ‘re-enactive empathy’ in the writing of history. For this technique in Thucydides, cf. Tamiolaki 2013b, 54–57. For repetition and redundancy as techniques of emotional arousal in the inscriptions, see Chaniotis 2012c. See for instance Hellenika 2.3.39–40, 6.3.12, repetition of the motive of fear. For the controversies surrounding the chronology of the Elean War, see Tuplin 1993, 201– 205. Hellenika 3.2.21–23.

28

Melina Tamiolaki Then, in a later incident, when King Agis had been sent to sacrifice to Zeus in accordance with an oracle, the Eleans prohibited him from praying for a victory in war, saying that it had long been established that Greeks should not consult oracles about a war against other Greeks. Angered by all these affronts (ἐκ τούτων οὖν πάντων ὀργιζοµένοις), the ephors and the assembly resolved to teach the Eleans a lesson in how to behave moderately (σωφρονίσαι αὐτούς).

In this episode, Lacedaemonian anger towards the Eleans is mentioned twice, at the beginning and at the end, in a form of ring composition/recapitulation about the cause of the invasion. It is illuminating to compare Diodoros’ presentation of the same event:88 While these events were taking place, the Lacedaemonians brought a number of charges against the Eleans (καὶ ἄλλα µὲν πλείονα τοῖς Ἠλείοις ἐνεκάλουν), the most serious being that they had prevented Agis, their king, from offering sacrifices to the god and that they had not allowed the Lacedaemonians to compete in the Olympic Games. Consequently, having decided to wage war on the Eleans, they dispatched ten ambassadors to them, ordering them, in the first place, to allow their subject cities to be independent, and after that they demanded of them their quota of the cost of the war against the Athenians. This they did in quest of specious pretexts for themselves and of plausible openings for war (ταῦτα δ᾿ ἔπραττον προφάσεις αὑτοῖς εὐλόγους καὶ πιθανὰς ἀρχὰς ζητοῦντες πολέµου).

In Diodoros (who draws from Ephoros) the core of the event is the same, but Diodoros does not explicitly mention anger as a motive of the Lacedaemonians and, more importantly, he states openly that the Lacedaemonians’ alleged complaints were just a pretext to start the war against the Eleans. Obviously the chronological distance allowed a greater detachment for Diodoros. But the comparison of his account with that of Xenophon reveals that Xenophon’s treatment is doubly biased: firstly, the repetition of the motive of anger (and this without the use of the formula λέγεται) gives an impression of truthfulness and objectivity. Secondly, the detailed explanation for this motive, which involves feelings of insult and dishonour, in combination with the use of the word σωφρονίσαι, contributes to the justification of the Lacedaemonians’ action. Overall, Xenophon’s repetition of ὀργιζόµενοι in this context serves to minimise (if not conceal) the Lacedaemonians’ imperialistic plans.89 Concerning the second episode, the raid of Sphodrias, it is the emotion of fear which is predominant. Yet we have a variation of the technique: it is not the same fear which is repeated; rather it is the combination of two different fears (that of the Athenians and that of the Thebans) which creates a specific effect on the narrative. Xenophon’s account goes as follows:90 Now the Athenians saw the power of the Spartans, and they saw, too, that the war was no longer being waged in Corinth but that the Spartans were going right past Attica and invading Theban territory. All this made them so fearful (οὕτως ἐφοβοῦντο) that they put on trial the two generals who had been accomplices of Melon against Leontiades and his followers; one of them they put to death, while the other, who did not wait around to stand trial, they exiled. 88 89 90

Diodoros 14.17.4–6. Cf. also Pausanias 3.8.3. Cf. Tuplin 1993, 54f., for further suggestions about the whole episode of the Elean War. Hellenika 5.4.19–20.

Emotions and Historical Representation in Xenophon’s Hellenika

29

The Thebans in their turn also became frightened (καὶ αὐτοὶ φοβούµενοι) that no one besides themselves would be at war with the Spartans, and so they contrived the following ploy (τοιόνδε εὑρίσκουσι µηχάνηµα). They persuaded Sphodrias, the Spartan governor of Thespiaia (by giving him money, it was suspected), to invade Attica, so that he might induce the Athenians to go to war against the Spartans.

It is obvious that Xenophon’s intention is to give a picture of generalised fear as the main cause of the events he reports: the Athenians are afraid of the growth of the Spartans’ power, a phrase that recalls Thucydides’ famous description of the reverse situation before the outbreak of the Peloponnesian War (1.23.6), whereas the Thebans seem to experience the same kind of fear: they urge the Spartan harmostes Sphodrias to invade Attica, in order to provoke hostilities between Athens and Sparta, because they are afraid of being alone fighting the Lacedaemonians. The phrase καὶ αὐτοὶ φοβούµενοι suggests that Xenophon intends the readers to connect the two events and situations. Again, however, Diodoros’ account is different: according to him (or his source), it was the Spartan king Kleombrotos who influenced Sphodrias to invade Attica (15.29.5). Although Diodoros is not contemporary with the events he narrates, his explanation should be taken into serious consideration, precisely because of the flaws that one can detect in Xenophon’s account: firstly, the relations between Athens and Sparta were not ideal;91 therefore, Theban fear is not really ‘needed’ as a catalyst. Secondly, in the course of the narrative in the Hellenika, Sphodrias is presented as developing his own initiative against Attica; that’s why he is finally summoned to Sparta by the ephors, in order to be put on trial. Again, this fact does not sit easily with Theban fear. Thirdly, after Sphodrias’ trial and acquittal, Xenophon reports that the Lacedaemonians sent Agesilaos to fight the Thebans, ‘because they thought that he would be a more sensible leader (φρονιµώτερον ἡγεῖσθαι) than Kleombrotos’.92 This explanation would not take us very far without Diodoros’ account. It thus becomes evident that, by delaying this information about Kleombrotos,93 Xenophon carefully avoids direct blame for the Spartan king. Finally, the subsequent course of the events, described in the Hellenika (5.4.34–57), does not confirm the Theban motive and plan initially presented, since we observe fights between Thebans and Spartans and not between the allied Thebans and Athenians against the Spartans. Overall, three different explanations of the event emerge: first, Sphodrias influenced by the Theban fear; second, Sphodrias influenced by Kleombrotos; and third, Sphodrias acting on his own initiative. The second and third explanations can be reconciled, which is not necessarily the case with the first and third explanations. The third explanation could also stand alone, but this would be less plausible, in my opinion: Xenophon here seems to exploit the tradition about Spartans’

91 92 93

Cf. Hellenika 4.4.18 (the Athenians are again presented as fearful of the Spartan growth of power). Hellenika 5.4.35. For the technique of narrative delay, see Hornblower 1994, 142, 147.

30

Melina Tamiolaki

indignation against people acting independently of their city and the ephors,94 but Agesilaos himself has cynically claimed that to the extent these actions are successful, they should not arouse complaints.95 Had Sphodrias’ raid succeeded, perhaps it would not have been necessary for Xenophon to come up with this explanation of the event. Then why did Xenophon invent the motive of the Theban fear?96 Most probably, this invention is linked to his well-known prejudice against the Thebans:97 by stressing Theban fear, Xenophon both exempted the Spartans from the failure of Sphodrias’ raid and made an additional comment on the malicious character of the Thebans (note also the word µηχάνηµα, which has negative connotations of fraud and deceit). 3.2 ‘Perceived’ or ‘focalised’ emotion The second technique on which it is worth commenting is what could be called ‘perceived’ or ‘focalised emotion’, that is the occurrences in which one or more agents of Xenophon’s Hellenika are presented knowing or finding out the emotion(s) of others. This technique is absent from his predecessors: the closest parallel can be found in Thucydides, who combines, however, two intellectual processes in the case of Hermokrates (οἰόµενος εἰδέναι, 6.3.23).98 In the Hellenika, it is applied both to individuals and to communities. Concerning individuals, it may not be coincidental that two of these occurrences refer to the knowledge of a satrap about an emotion experienced by Agesilaos: a) Tissaphernes’ strategy was to exploit his superiority in cavalry, as Agesilaos had none; and since Κaria was unsuitable terrain for cavalry and he believed also that Agesilaos would be furious at his deceit (ὅτι ἡγεῖτο αὐτὸν ὀργίζεσθαι αὐτῷ διὰ τὴν ἀπάτην) and also thought (νοµίσας) that he would, in fact, attempt to take Tissaphernes’ own estate in Κaria, he marched his entire infantry to Κaria ... because he thought the plain there would permit his cavalry to 99 overrun the Greeks before they could reach terrain that was unsuitable for horses. b) Tithraustes, for his part, seemed to believe that Agesilaos disdained the King’s power (καταµαθεῖν δοκῶν τὸν Ἀγησίλαον καταφρονοῦντα τῶν βασιλέως πραγµάτων) and that Agesilaos was not at all intending to leave Asia, but, rather, had great hopes that he could subjugate the King (ἐλπίδας ἔχοντα µεγάλας αἱρήσειν βασιλέα); as he was unsure what to do in 94 95 96 97

98

99

Cf. the case of Brasidas during the Peloponnesian War (Thucydides 4.82–83, with Hornblower 1996, 50–56). Hellenika 5.2.32. Lévy 2003, 236 n. 51, also thinks that Xenophon invented the Theban initiative in this event. Riedinger 1991, 174; Buckler 1982, 203. In Tamiolaki 2007, 209–211, I offer further suggestions about Xenophon’s narrative in Hellenika 5.4.19–20, in connection also with the formation of the second Athenian Confederacy. That said, I do not mean that the Hellenika contains more sophisticated presentations of emotions than the history of Thucydides. But the specific technique (that is, a participle or verb denoting an intellectual process combined with an emotion) is more typical of Xenophon; Xenophon seems to be more confident (or more naïve?) about the emotions of the agents of his history. Hellenika 3.4.12.

Emotions and Historical Representation in Xenophon’s Hellenika

31

the current state of affairs, Tithraustes sent Timokrates of Rhodos to Greece, giving him gold. 100 ...

At first glance these passages convey the impression that Xenophon seems careless about the validity of the motivation he reports, since it is highly unlikely that Tissaphernes or Tithraustes really knew how Agesilaos felt. More specifically, Tissaphernes could indeed have speculated that his having cheated Agesilaos101 may have aroused his anger; on the contrary, Tithraustes’ knowledge about Agesilaos’ hope ‘to conquer the land of the king’ can only be loosely inferred but not confirmed by Xenophon’s narrative.102 Yet, as I suggested above, it would be more productive not to dwell on the positivist question, but rather examine the function of this technique in its context. I would suggest that its function is threefold: firstly, it shows that emotions can become the object of rational consideration and hence are viewed as a factor which influences the course of the events. Secondly, it highlights especially the importance of Agesilaos’ emotions and is thus inscribed into the general portrayal of Agesilaos in the Hellenika. Finally, and more broadly, it is worth noticing that the mediation of the emotion in this case is double: Xenophon the narrator reports that an agent of his history knows the emotions of another agent. Why would Xenophon want to mediate the depiction of emotions in this way? It seems that this mediation serves to show the limits of speculation on the basis of emotions: Tithraustes’ speculations, on the one hand, proved rather successful, since Persian gold worked and Agesilaos’ plans in Asia were finally obstructed.103 But in the case of Tissaphernes, although Agesilaos might indeed have been angry with his deceit, he did not act as Tissaphernes thought and did not march towards Karia.104 Let us now examine the collective perceptions of emotions; these are two passages which concern the Lacedaemonians and their enemies. In both cases there is a rational and an emotional consideration involved: a) The Spartans now attacked Argos, because they saw the Argives enjoying the fruits of their lands and taking pleasure in the war (ἐνθυµηθέντες τοὺς Ἀργείους τὰ µὲν οἴκοι καρπουµένους, 105 ἡδοµένους δὲ τῷ πολέµῳ). b) The Spartans sent to the Mantineans and ordered them to tear down their wall, saying that otherwise they could not be confident that they would not take the enemy’s side. They brought the following charges against them: first, that they had learned (αἰσθάνεσθαι γὰρ ἔφασαν) that the Mantineans had been sending grain to the Argives even while the Spartans were at war with them; second, that the Mantineans had occasionally failed to send their military forces to serve with those of Sparta on the pretext that a sacred truce was then in effect; and finally they charged that whenever the Mantineans did accompany the Spartan forces, they served with them badly. The Spartans also said that they were aware that the Mantineans envied them, whenever anything good 100 101 102 103 104

Hellenika 3.5.1. Hellenika 3.4.6. Agesilaos’ Panhellenism is ambivalent in the Hellenika. See Dillery 1995, 41–58. Agesilaos is summoned back to Greece a bit later (Hellenika 4.2.1-5). Xenophon writes characteristically (Hellenika 3.4.12): ὁ δ’ Ἀγησίλαος ἀντὶ τοῦ ἐπὶ Καρίαν ἱέναι … 105 Hellenika 4.4.19.

32

Melina Tamiolaki

happened to the Spartans and, conversely, were delighted when some misfortune overtook them (ἔτι δὲ γιγνώσκειν ἔφασαν φθονοῦντας µὲν αὐτούς, εἴ τι σφίσιν ἀγαθὸν γίγνοιτο, ἐφηδοµένους δ᾿, εἴ τις συµφορὰ προσπίπτοι). It was also said that the Thirty Years peace between Sparta and Mantineia, which had been made after the battle at Mantineia, expired in this year.106

I will focus on the second passage, which offers more ground for close analysis. In this passage the phrase concerning the emotions of the Mantineans is placed (and perhaps lost) among arguments of rational (military and political) consideration: at the beginning of the motivation description, the Lacedaemonians are presented as displeased with the Mantineans, because the latter helped their enemies, the Argives, and because they were not good military followers; at the end, there follows again a rational argument, about the Thirty Years Peace. Interestingly, the mediation of the Mantineans’ emotions is threefold in this case: Xenophon reports that the Lacedaemonians say that they know the Mantineans’ emotions towards them (ἔφασαν, γιγνώσκειν). The Mantineans are perceived by the Lacedaemonians as φθονεροί and ἐπιχαιρέκακοι.107 By mediating the description of emotion in this way and by surrounding it with rational considerations, Xenophon seems to show awareness that this motive is not sufficient to justify an invasion. Now, he could then have omitted it; nevertheless, the presence of this motive is compatible with a constant trait of the Hellenika, which is the moral presentation of cities.108 Consequently, the emotional element as a comment on the Mantineans’ character, though marginalised, reflects Xenophon’s tendency to assign individual qualities to cities.109 3.3 Appeal to emotions in speeches We can now examine the third technique, the appeal to emotions in speeches. The presence of emotions in speeches is obviously not Xenophon’s innovation: his predecessors (Herodotos, Thucydides, but also Homer and the tragedians) also present leaders trying to inspire hope and courage in their followers, orators wanting to arouse pity or demagogues attempting to manipulate the emotions of their audience. Xenophon’s innovations lie elsewhere: firstly, besides conventional use of emotions in speeches, Xenophon resorts extensively to dialogue in the Hellenika, which may be a Socratic legacy.110 Interestingly, emotions play an important role in the dialogues, since the protagonists of the Hellenika often utter first-

106 Hellenika 5.2.2. 107 In this passage, Xenophon’s description exactly matches the Aristotelian definition of φθόνος (envy) as ‘a disturbing pain arising from the well-being of another’ (Rhetoric 1386b18–19), as well as of ἐπιχαιρεκακία, as the pleasure experienced by another’s misfortune (Nikomachean Ethics 1108b1–10); see Konstan 2006, 112–116. 108 The most characteristic example is Xenophon’s praise for the endurance of the Phleiasians (Hellenika 7.2.16–23). 109 The same tendency is observed in Herodotos, see Tamiolaki 2006, 32–35. 110 For the presence of dialogue in the Hellenika, see Gray 1981.

Emotions and Historical Representation in Xenophon’s Hellenika

33

person statements about their emotions (emotives, according to William Reddy’s terminology). A characteristic example is Agesilaos, who states:111 ... I am exceedingly happy when I avenge myself on an enemy, I think I am much happier still when I can devise something good for my friends.

Another example is Kinadon. When he is arrested, before being lead to his death, he explains his initiative for the conspiracy as follows: ‘he had done it so that he would not be inferior to anyone (µηδενὸς ἥττων εἶναι) at Sparta.’112 This statement is complex and can involve more than one emotion: anger (as a desire for retaliation because of a perceived insult), envy (for the privileges of the homoioi),113 and hatred (towards the emotional community of the Spartan homoioi). Another distinctively Xenophontic trait which derives from the abundant presence of dialogue in the Hellenika is that in this work there is no systematic interconnection between speech and narrative, as is the case in Thucydides’ history.114 From this perspective, occurrences in which this connection is observed deserve close attention. Finally, and more surprisingly perhaps, Xenophon’s Hellenika contains few emotional speeches (in the sense of an accumulation of emotions, emotional exaggeration etc.); rather, orators in the Hellenika tend to focus on particular emotions, so that one could detect specific patterns concerning the topic of emotional appeal. I will begin my analysis with (the limited presence of) emotional speeches in the Hellenika. An episode which had a strong impact in Athenian political life and was also emotionally exploited was the restoration of democracy and the amnesty of 401 BCE.115 This event not only united all Athenians under the democracy, a reunion that was further enhanced by foreigners’ and even slaves’ participation in the restoration of democracy,116 but also fostered an image of forgiving and indulgent democrats, eager to forget the evils suffered under the oligarchs. This was the meaning of amnesty.117 One would expect a great emotional exploitation of this event by Xenophon. Yet Xenophon’s treatment seems more balanced. This episode covers ten pages in the Oxford Classical Texts edition. In these pages, we read two speeches by Thrasyboulos, the main protagonist of the restoration, and one by the herald of the Mystery Initiates, Kleokritos. It is interesting to observe the different emotional register of the two speakers. In his first speech, Thrasyboulos does not employ

111 Hellenika 4.1.10: ἐγὼ µέντοι, καίπερ ὑπερχαίρων, ὅταν ἐχθρὸν τιµωρῶµαι, πολὺ µᾶλλόν µοι δοκῶ ἥδεσθαι, ὅταν τι τοῖς φίλοις ἀγαθὸν ἐξευρίσκω. 112 Hellenika 3.3.11. 113 In fact, if Kinadon was a hypomeion (‘inferior’), as it has been suggested, that is a degraded citizen, these emotions would seem to be justified. 114 For the interconnection of speech and narrative in Thucydides, the classic study is Romilly 1966; see also, more recently, Morrison 2006 and Pausch 2010. 115 For this event, see Loraux 1997; Wolpert 2002. 116 See the famous inscription IG II2 10, with Rhodes and Osborne 2003, 20–26. 117 Cf. Chaniotis 2013c, for the emotional contexts of amnesty regulations.

34

Melina Tamiolaki

terms related to emotions, but it is obvious that he tries to arouse the anger of the Athenians towards the Thirty Tyrants: Those stationed at the end of their line on the left are the very Thirty who have deprived us of our city, although we had committed no wrong. These are the men who drove us from our houses and who condemned our dearest relatives and friends. ... If god wills, this victory will restore to us our country, our homes, our freedom, and honors, and, if we have them, our wives and children. Those of us are truly blessed who as victors will see this sweetest of days. Even one who dies fighting here will be fortunate, for no one, however wealthy, could procure for himself so fine a monument (εὐδαίµων δὲ καὶ ἄν τις ἀποθάνῃ· µνηµείου γὰρ οὐδεὶς οὕτω πλούσιος ὢν καλοῦ τεύξεται). When it is the right moment, I shall begin the paean. When we call upon Enyalios, then let us all with one and the same spirit have our vengeance upon these men, in requital for all the wanton outrages we have endured at their hands (τότε πάντες ὁµοθυµαδὸν ἀνθ᾿ ὧν ὑβρίσθηµεν τιµωρώµεθα τοὺς ἄνδρας).118

Τhe expression εὐδαίµων δὲ καὶ ἄν τις ἀποθάνῃ famously recalls the Thucydidean Epitaphios;119 Thrasyboulos aims at inspiring pride in his followers, by establishing a connection between the glorious Athenians praised by Pericles and the democrats who fight against the Thirty Tyrants. After the democrats’ victory, the episode concludes with a second speech by Thrasyboulos, pronounced this time before the supporters of oligarchy. This speech again contains no emotional terms. It begins with the phrase ‘I advise you to know yourselves (γνῶναι ὑµᾶς αὐτούς)’ which has Socratic connotations. Yet, the rhetorical questions which follow are emotionally loaded; they serve to underline the justice of the democrats’ cause and also arouse feelings of shame in the audience:120 Do you believe that you are more righteous than we are? Well, the common people are poorer than you, but they have never acted unjustly against you for the sake of money. Whereas you, on the other hand, who are the wealthiest of us all, have committed many shameful crimes just for the sake of personal gain. But since you lack all concern for justice, perhaps you are so confident because of your bravery. And yet what better way could there be of determining who is braver than by considering what happened when we fought against each other? ... Do you base your pride on the friendship of the Spartans? How can that be now, when they have handed you over, just as people collar and hand over vicious dogs, to the common people here, whom you have wronged – and having done that, they then turned and went home?

The comparison of the Lacedaemonians’ behavior towards the oligarchs with the handing over of dogs is a strong metaphor, which aims at challenging the bond between the oligarchs and the Spartans. Overall, it is remarkable that neither speech of Thrasyboulos has a reconciliatory tone; rather both tend to underline the superiority of the democrats (in terms of morality, courage, philopatria) over the oligarchs.

118 Hellenika 2.4.13 and 17. 119 Thucydides 2.43.4–5. For further intertextual allusions with Thucydides in the speeches of the Hellenika, see Tamiolaki forthcoming a. 120 Hellenika 2.4.40–42.

Emotions and Historical Representation in Xenophon’s Hellenika

35

On the contrary, the speech of Kleokritos, pronounced before the supporters of oligarchy, contains emotional elements which more openly support the reconciliation:121 Fellow citizens, why do you drive us out? Why do you wish to put us to death? For we have done you no wrong ever, and we have shared with you (µετεσχήκαµεν) the most solemn sacred rites and sacrifices and the most beautiful festivals; we men from both sides have joined in dances together (συγχορευταί), gone to school together (συµφοιτηταί), served as soldiers together (συστρατιῶται); we have endured many dangers in common with you (µεθ’ ὑµῶν κεκινδυνεύκαµεν) by land and by sea, for our common safety and our common freedom (ὑπὲρ τῆς κοινῆς ἀµφοτέρων ἡµῶν σωτηρίας τε καὶ ἐλευθερίας). In the name of the gods of our fathers and our mothers, in the name of our common ancestry, our links through marriage and our bonds of friendship – in the name of all these things, which so many of us share with one another (πάντων γὰρ τούτων πολλοὶ κοινωνοῦµεν ἀλλήλοις) – respect the gods and men and cease from doing wrong to your country. ... Know well, however, that even for these men who have just been killed by us, not only you but we, too, have wept many tears (πολλὰ κατεδακρύσαµεν).

First of all, the emphasis on common activities – we notice the repetition of compounds with συν- and µετ-, as well as the antonym ἀλλήλοις – stresses the bonds of the emotional community and their adherence to common values; secondly, the appeal to the shame before gods and humans (αἰδούµενοι) creates a contrast between the pious Athenians and the impious Thirty Tyrants, thus concealing the connection between oligarchs and the tyrants. Finally, and more importantly, the emphatic word κατεδακρύσαµεν, which he employs at the end in order to describe the evils of the war, points to the common emotional response of both democrats and oligarchs towards the war, since both deplore its evils. Why does Xenophon present these divergent perceptions of reconciliation? It is also relevant that in his narrative he states plainly that the Athenians agreed not to remember past wrongdoings (µὴ µνησικακήσειν),122 without elaborating on it emotionally. I suggest that the explanation of this presentation is twofold: first of all, Xenophon’s narrative reflects something real: the tensions and difficulties of this reconciliation. A bit later the very essence of the amnesty is violated, since the democrats send away to Asia Minor those who had served in the cavalry under the Thirty Tyrants in order to get rid of them.123 Overall then, it would be reasonable to assume that Kleokritos, as a religious representative, expresses the idealised picture of reconciliation, whereas Thrasyboulos represents the more pragmatic one.124 Secondly, Xenophon’s oligarchical preferences125 must have cer-

121 Hellenika 2.4.20–22. 122 Hellenika 2.4.43: ‘both parties then swore oaths not to remember past wrongdoings (µὴ µνησικακήσειν), and to this day they live as fellow citizens and the people abide by their oaths.’ 123 Hellenika 3.1.4. 124 Of course, context also plays a role here: when Kleokritos delivers his speech, the democrats have not yet won, so he needs to make a more idealised appeal in order to persuade the oligarchs to give up fight; on the contrary, Thrasyboulos in his second speech uses the rhetoric of a winner addressing the losers.

36

Melina Tamiolaki

tainly played a role in the presentation of this narrative: Xenophon would not intend to give an idealised picture of the democrats and would probably prefer to minimise the achievement of the amnesty.126 Let us turn now to the joint presence of emotions in speech and narrative. The presence of emotions in these contexts usually serves to promote and/or confirm a historical explanation. For example, concerning the Arginousai trial, Xenophon states in his narrative that a significant motive for the attitude of the prytaneis was fear:127 This so frightened the presiding committee (οἱ δὲ πρυτάνεις φοβηθέντες) that they now agreed to put the matter to a vote, all of them except one – Socrates son of Sophroniskos, who said he would do nothing except in accordance with the law.

It is the emotion of fear that Euryptolemos attempts to ban in his speech:128 For what is it that you fear that leads you to act in such a haste (τί δὲ καὶ δεδιότας σφόδρα οὕτως ἐπείγεσθε)? Is it that you fear you might not be able to punish or acquit someone as you wish if you judge according to the law, but that you would be able to do so if you act contrary to the law? That is precisely the way that Kallixenos persuaded the Council to introduce the decree about the single vote to the Assembly.

Interestingly, when Xenophon gives the result of the voting, he does not involve emotions, but states prosaically:129 When they voted for the second time, they chose the motion of the Council. After this they found the eight generals who took part in the sea battle guilty. The generals who were present, six in number, were put to death.

Yet, it has become obvious from the previous discussion that fear must have motivated the decision to execute the generals. The same goes for Theramenes and the Thirty. The Thirty Tyrants are dominated by fear, an emotion that dictates their political activities:130 Kritias and the rest of the Thirty, who already feared (φοβούµενοι) that opposition might form, especially around Theramenes, drew up a list of three thousand men, who, they said, would share in governing the city.

But this fear is connected with the fear experienced by the citizens because of the tyrants’ policy, as becomes obvious from Theramenes’ speech:131

125 Although the traditional view according to which Xenophon was openly hostile to democracy has been nuanced (see Gray 2004; Kroeker 2009), it would be far-fetched to maintain that Xenophon was not at all influenced by his aristocratic preferences and background; cf. Tamiolaki 2013a. 126 See for a contrasting presentation, [Aristotle], Athenaion Politeia 40.2–3, who places stronger emphasis on the element of homonoia. 127 Hellenika 1.7.15. 128 Hellenika 1.7.26. 129 Hellenika 1.7.34. 130 Hellenika 2.3.18. 131 Hellenika 2.3.39.

Emotions and Historical Representation in Xenophon’s Hellenika

37

For when Leon of Salamis – a man who was both thought to be and actually was meritorious, and one who was completely innocent of any wrongdoing – was put to death, I knew that those like him would grow fearful (οἱ ὅµοιοι τούτῳ φοβήσοιντο), and because of their fear (φοβούµενοι δὲ) they would become hostile to this government.

The repetition of the motive of fear in the narrative of the Thirty Tyrants presents an ambiance of generalised terror and suspicion and thus enhances the dramatic dimension of this period of Athenian history. Another example concerns hatred towards the Lacedaemonians. Xenophon states in his narrative that Tithraustes gave money to some people in Greek cities, in order to wage war against the Lacedaemonians and that consequently those people ‘took the money and began to slander the Spartans, each in his own city, and when they had brought their individual cities to hate the Spartans (εἰς µῖσος αὐτῶν προήγαγον), they then allied the greatest cities with one another’.132 The same motif of hatred towards the Lacedaemonians is also invoked in the speech of the Thebans before the Athenians:133 This is what will now happen to the Spartans if both of us fight side by side against them in battle: the many who hate them will reveal themselves (ἀναφανήσονται πολλοί οἱ µισοῦντες αὐτούς).

Here Xenophon builds upon previous tradition: the Athenians had also aroused the enmity of their allies due to their tyrannical behavior, but Thucydides’ description was milder than Xenophon’s, since he did not speak openly about hatred in his narrative.134 Moreover, the Athenians themselves stated in the Melian Dialogue that arousing hatred is a sign of power.135 It seems then that Xenophon, by describing the negative results of hatred towards the Lacadaemonians, further elaborates and adapts the Thucydidean motif of hatred towards an imperial power. Another trait of speeches in Xenophon’s Hellenika is what could be called ‘patterned emotional appeal’. This mainly concerns the emotion of fear. Speeches of the Hellenika could be roughly divided into two categories: those dominated by the ‘do not be afraid’ motif and those pervaded by the ‘be afraid’ motif. To the first category belong speeches pronounced before battles: the speeches of Derkylidas and Teleutias are the most characteristic examples. Derkylidas concludes his speech as follows:136 And if someone among you is afraid that we will be besieged here by land and sea, let him know that there is not yet a Greek fleet at sea, and if the barbarians should attempt to rule the sea, Greece will not allow it; and so by helping herself, Greece will also be your ally.

Teleutias does not refer directly to fear, but his speech evidently aims at arousing hope and courage in his soldiers:137 132 Hellenika 3.5.1. 133 Hellenika 3.5.11. 134 Thucydides’ focalisation is rather towards the eunoia that the cities showed towards the Lacedaemonians (Thucydides 2.8). 135 Thucydides 5.95. 136 Hellenika 4.8.4, 4.8.5. 137 Hellenika 5.1.16–17.

38

Melina Tamiolaki I know well that in former time you were brave men; but you must now strive to become even braver so that, just as we gladly toil together, so, too, we will gladly enjoy prosperity together. ...

It is not by chance that both of these speeches are successful and have a positive impact on their audience. For the first one, Xenophon writes:138 ‘When they had heard this, the people of Abydos obeyed him, not grudgingly but enthusiastically.’ The reaction to Teleutias’ speech is similar:139 So he spoke, and all the men shouted out that he should command them to do whatever was necessary, since they were ready to obey him.

This emotional rhetoric is a constant feature of Greek historiography and there is a great controversy over whether pre-battle speeches reflect reality or represent conventions of the genre. 140 The patterned emotional appeal in Xenophon’s speeches rather reinforces the latter interpretation (but dwelling on this topic is out of the scope of this paper). Besides pre-battle speeches, the ‘don’t be afraid’ motif is also present in the speeches of Kallistratos and Prokles the Phleiasian. The former urges the Spartans not to be afraid of the Persian King,141 whereas Prokles urges the Athenians not to be afraid of the Spartans:142 If, nevertheless, you are afraid that the Spartans might again cause you trouble in the future should they escape harm this time, bear in mind that one does not normally fear the rise to power of those whom one has benefited, but, rather, those whom one has injured.

On the other hand, the ‘be afraid’ motif is related to the growth of power of certain cities or individuals. The speakers who have recourse to this motif usually aim at attracting an alliance. For example, Kleigenes of Akanthos, in order to achieve an alliance with the Lacedaemonians, warns them about the growth of power of Olynthos:143 Consider this, too: how is it reasonable for you to take great care to prevent Boeotia from uniting and yet ignore a much greater power that is being consolidated, one that is becoming strong not only by land, but also by sea?

Similarly, Polydamas of Pharsalos warns the Lacedaemonians about the growth of power of the tyrant Iason of Pherai:144 If you send assistance that will be suitable not only for the Pharsalians but for all the Thessalians to wage war on Iason, then his cities will abandon their alliance with him, for everyone is afraid of where this man’s power might lead.

138 Hellenika 4.8.6. 139 Hellenika 5.1.18. 140 The topic of pre-battle speeches is vast and complex. See a recent assessment in the essays collected by Abbamonte, Miletti, and Spina (eds.) 2009. 141 Hellenika 6.3.13. 142 Hellenika 6.5.40. 143 Hellenika 5.2.15–16. 144 Hellenika 6.1.14.

Emotions and Historical Representation in Xenophon’s Hellenika

39

The motif of fear about the growth of a power evokes the Thucydidean explanation for the outbreak of the Peloponnesian War.145 Xenophon seems again to adapt this motif in the political circumstances of the fourth century BCE. This motif has a great currency during this period, since many cities struggle for hegemony in Greece,146 so fear is evidently more widespread. However, it is noteworthy that Xenophon also shows the limits of this appeal: the speech of Kleigenes is successful, but this is not the case with Polydamas, who, despite his long speech, does not manage to obtain military assistance from the Lacedaemonians.147 In sum, Xenophon has recourse to emotions on a wide range of occasions. He seems to build upon Thucydidean techniques of ascribing motivation, but he develops them further. Xenophon puts greater emphasis on emotions in his narrative than in the speeches, in which we can observe elements of standardisation. 4 XENOPHON’S INNOVATIONS IN THE TREATMENT OF EMOTIONS Although Xenophon follows Thucydides closely in his historical representation, he departs from him in significant ways concerning the depiction of emotions, and inaugurates certain innovations. The first innovation is a greater tendency towards dramatisation, as is visible in the description of collective emotions. Herodotos too described collective reactions, such as that of the Athenians to Phrynichos’ drama,148 and Thucydides presents scenes of pathos in his narrative of the Sicilian expedition, but these scenes are the exception rather than the rule.149 On the contrary, Xenophon’s exploitation of these scenes is highly theatrical and seems to form a more prevalent pattern. More specifically, there is a famous scene concerning the Athenians after their destruction at the Aigos Potamoi, whereas all the other descriptions concern Sparta. The passage about the Athenians goes as follows:150 The Paralos arrived at Athens during the night, bringing news of the disaster (συµφορά) at Aigos Potamoi, and a cry (οἰµωγή) arose in the Piraeus and ran up through the Long Walls and into the city itself as one man imparted the calamitous news to the next. As a result, no one slept that night as they mourned (πενθοῦντες) not only for the men destroyed but even more for themselves, thinking they would suffer (πείσεσθαι νοµίζοντες) the same catastrophes they had inflicted on others – the Melians ..., Histiaians, Skionaians, Toronaians, Aiginitans, and many other Greeks.

The words συµφορά, οἰµωγή, πενθοῦντες, in combination with the vivid scene of people unable to sleep, reinforce the dramatic color of this event. The intertextual dialogue with Thucydides has already been discussed by Tim Rood: Xenophon 145 Thucydides 1.23.6. 146 For the political situation in the fourth century, see Beck 1997; cf. the essays collected by Funke and Luraghi (eds.) 2009, concerning the Peloponnesian League. 147 For a similar function of motivation in Thucydides, see Tamiolaki 2013b, 64–69. 148 Herodotos 6.21. 149 Cf. Stahl 2013. 150 Hellenika 2.2.3.

40

Melina Tamiolaki

mentions events (the destruction of Melos, Histiaia, Skione, Torone, and Aigina), which are well-known from Thucydides’ history.151 I would argue, furthermore, that Xenophon not only exploits the Thucydidean description, but also makes an implicit ‘correction’ to it: Thucydides described the suffering of smaller cities by the Athenians plainly and prosaically, without trying to arouse the pity of his audience/readership. 152 By presenting the Athenians as victims, fearing being found in the place of the smaller cities they had oppressed, it appears as if Xenophon complements the emotions missing in the Thucydidean narrative. The passages about the Spartans describe their reactions after battles, the first two after their defeat at Lechaion (390 BCE) and Leuktra (371 BCE) respectively, and the third after their victory over the Arkadians under the leadership of Archidamos (the famous ‘Tearless Battle’ in 368 BCE): a) Now inasmuch as such a defeat was most unusual for the Spartans, there was much grieving (πολὺ πένθος) throughout the army – except for those whose sons, fathers, or brothers had died in that battle: these men went around with beaming faces, delighting in their personal misfortune, as if they had been victors (οὗτοι δ᾿ ὥσπερ νικηφόροι λαµπροὶ καὶ 153 ἀγαλλόµενοι τῷ οἰκείῳ πάθει περιῇσαν). b) Some Spartans thought their misfortune unendurable (ἀφόρητον τὴν συµφορὰν ἡγούµενοι); they said that they had to prevent the enemy from setting up a trophy and should try to take up the corpses by fighting for them rather than getting them back under truce. ... When the battle was over, a messenger was sent to Sparta to announce the disaster. He arrived on the last day of the festival of the Gymnopaidiai, during the performance of the men’s chorus in the theatre. When the ephors heard about the disaster, they were greatly distressed (ἐλυποῦντο µέν), as one would expect them to be, but they did not have the chorus leave the theater. Instead they allowed them to finish their performance, and then they revealed the names of the dead to each of the families. They ordered the women not to cry out but to bear their suffering in silence, and on the next day, one could see the relatives of those who had died going around with bright and happy faces, while you would have seen on the street only a few of those whose relatives had been reported as still alive, and these few were making their way with gloomy expressions and downcast faces (προεῖπαν δὲ ταῖς γυναιξὶ µὴ ποιεῖν κραυγήν, ἀλλὰ σιγῇ τὸ πάθος φέρειν. τῇ δ᾿ ὑστεραίᾳ ἦν ὁρᾶν, ὧν µὲν ἐτέθνασαν οἱ προσήκοντες, λιπαροὺς καὶ φαιδροὺς ἐν τῷ φανερῷ ἀναστρεφοµένους, ὧν δὲ ζῶντες ἠγγελµένοι ἦσαν, ὀλίγους ἂν εἶδες, τούτους δὲ σκυθρωποὺς καὶ ταπεινοὺς 154 περιιόντας). c) When the battle was over, Archidamos set up a trophy and immediately dispatched Demoteles to Sparta to announce there the magnitude of the victory and the fact that not a single Spartan had died in the battle, while very many of the enemy had been slain. It is said that when those back in Sparta heard the news, they all began to weep, beginning with Agesilaos, and then on to the members of the gerousia and the ephors, and finally everyone. So it is in155 deed the case that tears are common to both joy and grief.

151 Rood 2004. 152 Thucydides 5.116.4 (Melos), 1.114.3 (Histiaia), 5.32.1 (Skione), 5.3.4 (Torone), 2.27.1–2 (Aigina). 153 Hellenika 4.5.10. 154 Hellenika 6.4.15–16. 155 Hellenika 7.1.32.

Emotions and Historical Representation in Xenophon’s Hellenika

41

Concerning the defeats, Xenophon wishes to praise the Spartan character and ethos, according to which death in battle is something glorious, which should not arouse grief but joy.156 He devotes great detail to the Spartan reactions after the important defeat at Leuktra and establishes a distinction between grief for political purposes and personal grief. The psychology of the Spartans seems thus torn between grief for their political loss and joy (?) for those who died in the battle. Yet, Xenophon’s narrative suggests that the joy is only apparent: it concerns only their facial expression (λιπαροὺς καὶ φαιδροὺς ἐν τῷ φανερῷ ἀναστρεφοµένους). It is interesting to compare similar reactions of Spartans in the Hellenika. In another episode, Agesilaos adopts a similar attitude: while his soldiers are distressed with Tissaphernes, he approaches him with a bright face (Ἀγησίλαος δὲ µάλα φαιδρῷ τῷ προσώπῳ).157 Again Agesilaos hides his negative emotions when he is summoned back to Sparta from Asia Minor:158 Agesilaos was quite distressed (χαλεπῶς µὲν ἤνεγκεν) at these orders, thinking of how many honors and hopes he would now be deprived of. Nevertheless, he called his allies together and revealed to them what the city had ordered, explaining that it was necessary for him to bring help to his country. ...

These passages suggest that Xenophon might have wanted to show that a trait of the Spartan character was the ability to hide emotions. In this way, the psychology of the Spartans is presented in accordance with (and perhaps even derives from) their politeia.159 The reaction after the victory over the Arkadians completes the depiction of the Spartan psychology. In this case, Xenophon reports tears of joy, which is again an innovation not found in his predecessors (perhaps this is the reason why he adds the authorial comment that tears are a response common to both grief and joy). Contrary to the previous passages, the Spartans are glad here that nobody died. It seems then that the motif of the Spartan ‘bright face’ over the dead was a means to compensate for their real (inner) mourning and this attitude presumably formed a part of the famous Spartan mirage.160 Another trait, which further differentiates Xenophon from his predecessors, is the emphasis on individual emotions and subsequently a greater interest in the process of psychological transformation. Terms such as ὀργίζοµαι, ἄχθοµαι etc. are often used in historiography to describe collective emotions.161 On the con156 Cf. the note by Marincola 2009, 246, about the Spartan law of disenfranchisement for survivors of a defeat in battle, which was suspended by Agesilaos (Plutarch, Agesilaos 30.4). 157 Hellenika 3.4.11. 158 Hellenika 4.2.3. 159 Cf. the succinct comment of Thucydides (5.68) about the κρυπτόν (secret character) of the Spartan politeia. Reddy 2001, 60f., mentions anthropological studies about the Balinese, for whom ‘maintenance of cheerful moods at all costs was a matter of “public health”. ... To put on a “bright face” when feeling grief stricken, for the Balinese, is not dissembling, because the face can help the heart to change.’ 160 For the ‘mirage spartiate’, see Ollier 1943; Cartledge 2003; Powell and Hodkinson (eds.) 1994. 161 See the many references of the verb ὀργίζοµαι in Bétant 1969, ad loc.

42

Melina Tamiolaki

trary, in the Hellenika, there is a balance between collective and individual emotions and on many occasions a greater emphasis on the latter.162 The psychological transformation can be immediate – that is, a quick and almost automatic change of emotional response – or may evolve in the course of an emotional episode. The most characteristic examples of immediate psychological transformation concern Agesilaos and the Mantineans: a) When Agesilaos learned of the defeat (that is, of Knidos), he was at first very distressed (χαλεπῶς ἤνεγκεν). He considered, however, the character of his army: the majority, he knew, would share happily in good news, but there was no need to share adverse tidings with them; so in the end, he changed mood (µεταβαλών) and ordered that an announcement be made that Peisandros had died, but that he had been victorious in the naval battle.163 b) Αfter this, the city wall was taken down and the Mantineans were settled into four villages, dwelling just as they had in ancient times. At first they were distressed (τὸ µὲν πρῶτον ἤχθοντο) that they had been forced to take down the houses they already had and build others in their place; but later the wealthy were pleased with what had been done (ἥδοντο τοῖς πεπραγµένοις), because they now dwelt nearer to their estates ... they now had an aristocratic government and they were freed from the burden of demagogues. ... the Mantineans now served much more enthusiastically (πολὺ προθυµότερον) than when they had lived under a democratic government.164

In order to better delineate the role of individual emotions in the Hellenika, we can now examine certain ‘emotional episodes’ in this work more closely. An emotional episode is an event between two or more individuals, which involves emotions. It involves the following aspects: antecedent conditions, emotions, verbal expressions, and resolution.165 Emotional episodes are often attested in Herodotos, but are rare in Thucydides.166 Xenophon thus seems to follow and further develop a Herodotean legacy. Two emotional episodes in the Hellenika are worth commenting on. The first one concerns the tension in the relationship between Agesilaos and Lysander. This tension is caused (antecedent conditions) by a transgression in Lysander’s behavior, who acts like a king although he is an idiotes (τῆς βασι-

162 For example, ὀργή is a markedly individual emotion, experienced by Kallikratidas (1.6.6), Pausanias (2.4.35), Derkylidas (3.1.17–19), Agesilaos (3.4.4, 3.4.8, 3.4.12, 5.3.24), Pharnabazos (4.8.7–8), Teleutias (5.3.5). On the contrary, references to collective anger concern the anger of the Lacedaemonians towards the Thebans (3.5.5), of the Athenians towards the Lacedaemonians (3.5.9, 5.4.63), and of the Argives towards the Phleiasians. The same goes for hatred: the Hellenika contains examples in which, contrary to Aristotle’s definitions, it is directed against an individual: Kallixenos (1.7.35), Ismenias (5.2.23), and Mnasippos (6.2.20). 163 Hellenika 4.3.13. 164 Hellenika 5.2.7. 165 See Sanders 2012, 157, who cites the relevant bibliography. 166 E.g., see the role of fear in the story of Gyges and Kandaules (1.6–14). In general, all short stories in Herodotos could be considered emotional episodes. On the contrary, Thucydides tends to emphasise the collective or even subject the individual to the collective (as is the case with the eros between Harmodios and Aristogeiton; cf. also below p. 44).

Emotions and Historical Representation in Xenophon’s Hellenika

43

λείας ὀγκηρότερον διάγων).167 Agesilaos is angry with Lysander and decides to take his revenge: he does not help people who are sent to him by Lysander. The following dialogue between Lysander and Agesilaos is revealing:168 Lysander was distressed by this dishonour (βαρέως φέρων τῇ ἀτιµίᾳ) and so he went to the king and said, ‘Now I see, Agesilaos, that you know very well how to diminish your friends.’ ‘Yes, by Zeus,’ Agesilaos replied, ‘at least those who wish to appear greater than me. As for those who increase my honor, well, I would be ashamed of myself (αἰσχυνοίµην ἄν) if I did not know how to honor them in return.. ‘Well,’ said Lysander, ‘perhaps you are behaving now more sensibly than I did in the past. Do me one favor at the very least – send me away from here so that I might not be ashamed (ὅπως ἂν µήτ’ αἰσχύνωµαι) by being here and having no influence with you, and also so that I might not be in your way. Whenever I go, I shall always attempt to always conduct myself so as to be of some advantage to you.’

This episode is important because it raises issues of status: it is noteworthy that although both individuals feel insult, only Agesilaos is presented as angry (which is in accordance with the Aristotelian definition of anger, according to which only superiors experience this emotion).169 We can detect again here certain ‘emotives’ (first-person statements about emotions) and also observe their ‘protean transformation’, according to Reddy’s terminology: from the moment these emotions are expressed, they are subject to change.170 For instance, the allusion to shame (αἰσχύνη) in this passage is interesting: Agesilaos states that he would feel ashamed if he didn’t help those who enhance his power, whereas Lysander urges him to assign him something, so as to cease to feel ashamed by not being able to serve the king.171 It is obvious that Agesilaos’ reference to his hypothetical shame urges Lysander to express his own shame; in fact, in the course of the dialogue, Lysander’s initial annoyance/distress (βαρέως φέρων) has been transformed into shame. Indeed, Agesilaos sends Lysander to the Hellespont, where the latter undertakes successful military action, by attracting the alliance of the Persian Spithridates. Xenophon states in the end that Agesilaos was glad (ἥσθη): again, Agesilaos’ initial anger has been transformed into joy and this constitutes the resolution of the episode. The second emotional episode, which is worth commenting on, concerns the eros of Archidamos, son of Agesilaos, for Kleonymos, son of Sphodrias.172 The latter asks for the former to influence his father, the king, so as to acquit Sphodrias (antecedent situation). Several emotions are evolved in this episode: first of all, eros; secondly, pity of Archidamos for his friend: ‘when Archidamos saw Kleonymos weeping, he stood beside him and likewise shed tears (συνεδάκρυε παρεστηκώς);’173 thirdly, fear and shame of Archidamos towards 167 168 169 170 171

Hellenika 3.4.8; cf. also Plutarch, Lysandros 23-24. Hellenika 3.4.9. Aristotle, Rhetoric 1379a, with Konstan 2003b; cf. Chaniotis 2012c, 116–118. See Reddy 2001, 101–103, for the ‘self-altering’ effect of emotives. It seems to me that this reference to shame is innovative, because the standard for this emotion in this context is not society or culture, but the self. 172 Hellenika 5.4.25–33. 173 Hellenika 5.4.27.

44

Melina Tamiolaki

his father, Agesilaos: ‘Kleonymos, you should know that I am unable even to look my father in the face (οὐδ’ ἀντιβλέπειν δύναµαι)’.174 When Archidamos talks to his father and fails to persuade him, Xenophon describes again his psychological situation: δύσελπις (‘desponding’). Finally, there is a sudden reversal: Agesilaos decides to acquit Sphodrias and this prompts the joy (ἡσθείς) of Kleonymos (resolution of the episode).175 The two emotional episodes have some traits in common: firstly, Agesilaos is a prominent figure in both of them; secondly, they both underline psychological transformations: initial negative emotions are gradually eliminated and give way to joy; thirdly, and perhaps more importantly, both of these episodes constitute pauses in the narrative. More specifically, the episode about the eros between Archidamos and Kleonymos has the traits of a Herodotean digression (the episode is framed by the introductory phrase ἐγένετο δὲ τοῦτο τὸ αἴτιον and the conclusive τοιούτῳ µὲν δὴ τρόπῳ Σφοδρίας ἀπέφυγε, both of which point to its digressive character). I would argue, however, that there is a subtle intertextual dialogue with Thucydides as well: one of the rare digressions in Thucydides concerns the eros between Harmodios and Aristogeiton, an (emotional?) episode, which has a political dimension, since it plays a role in the murder of the tyrant’s brother.176 Thucydides’ digression may have provided further legitimisation for Xenophon, in order to present another example of how a political action (the acquittal of Sphodrias) was motivated by an emotion (eros). But Xenophon’s adaptation and innovation lies of course in that he provides much greater detail about emotions than his predecessor. The third innovation of Xenophon is that he theorises about emotions in his main narrative (what could be called, in modern terms, ‘emotionology’, how people think about emotions).177 This can be observed in his other works as well,178 but in the Hellenika we find one example of ‘emotionology’, Xenophon’s opinion about anger:179 Now I claim that men can learn from such experiences, and they can learn especially that it is not right to punish anyone in anger – even a slave, since masters who are angry often them174 Hellenika 5.4.27. For veiling as an expression of aidos, see Cairns 2009. 175 Hellenika 5.4.33. See Gray 1989, 59–63, who focuses on Xenophon’s interest in sentiments in this episode. 176 Thucydides 6.54–59 (δι’ ἐρωτικὴν ξυντυχίαν). 177 In Thucydides we also find theorisation about emotions, but only in the speeches, e.g., 2.11.7 (Archidamos theorises about anger); 3.45.5 (Diodotos theorises about hope and eros). 178 I plan to analyse these occurrences in a bigger project devoted to the function of emotions in Xenophon’s leadership theory. 179 Hellenika 5.3.7: ἐκ µέντοι γε τῶν τοιούτων παθῶν [ὡς] ἐγώ φηµι ἀνθρώπους παιδεύεσθαι µάλιστα µὲν οὖν οὐδ οἰκέτας χρὴ ὀργῇ κολάζειν· πολλάκις γὰρ καὶ δεσπόται ὀργιζόµενοι µείζω κακὰ ἔπαθον ἢ ἐποίησαν· ἀτὰρ ἀντιπάλοις τὸ µετ᾿ ὀργῆς ἀλλὰ µὴ γνώµῃ προσφέρεσθαι ὅλον ἁµάρτηµα. ἡ µὲν γὰρ ὀργὴ ἀπρονόητον, ἡ δὲ γνώµη σκοπεῖ οὐδὲν ἧττον µή τι πάθῃ ἢ ὅπως βλάψῃ τι τοὺς πολεµίους. Harris 2001, 319, comments on this passage from the perspective of slavery, but the issue is broader. See Harris 2001, 229– 263, for restraining anger about leaders, and more recently, Kalimtzis 2012, for the topic of taming anger.

Emotions and Historical Representation in Xenophon’s Hellenika

45

selves suffer greater evils than they inflict on their servants. And it is a complete and utter mistake to attack an enemy with anger rather than with judgment. For anger acts without foresight, whereas judgment has in view a way to harm one’s enemy without suffering any hurt from him in return.

The theorisation concerns the anger (ὀργή) of a leader and forms part of a historical explanation, according to which Teleutias’ defeat at Olynthos is due to his angry reactions. Xenophon’s comment aims thus at providing advice towards leaders, so that they achieve better political and military results. Moreover, it is interesting that Xenophon follows the (so-called) ‘hydraulic model’ of the emotions here: he describes ὀργή as something totally irrational, which does not involve any evaluation and contrasts it with γνώµη.180 This contrast is already attested in Thucydides.181 Given the presence in the Hellenika of Aristotelian examples of ὀργή as well, it would be reasonable to assume that the two models/ways of thinking about the emotions (hydraulic and cognitive) were not contradictory in ancient culture, but coexisted. Finally, Xenophon’s comment that the masters (δεσπόται) should not punish even their slaves in a state of anger reflects his theory about successful leadership, according to which leaders should gain the willing obedience of their followers.182 A final significant innovation of Xenophon in the Hellenika is that he more than once reveals his own emotions. This concerns mainly the emotion of awe or admiration: for the wit of Theramenes at the moment of his death,183 for the military preparations of Agesilaos in Ephesos,184 for the disciplined dioecism in Mantineia,185 for the endurance of the Phleiasians,186 and for the bravery of the Athenians.187 In this way, Xenophon intends to make his readers share his positive evaluations and emotions. Moreover, he seems further to develop a Herodotean technique, the emphasis on θῶµα. Herodotos often states that he will report something ‘marvellous’,188 but this concerns a place, a custom or an exotic and unusual activity.189 Xenophon’s adaptation lies in his assignation of a moral dimension to this emotion: what he considers worthy of admiration is in most cases a character trait.

180 For the ‘hydraulic model’ of the emotions, see Rosenwein 2006, 33–37. 181 Cf. Huart 1968, 50–57. 182 Cf. also Xenophon’s critical comments about Klearchos’ harsh attitude towards his soldiers (Xenophon, Anabasis 2.6.9–10). For the idea of caring about slaves in Xenophon, see Tamiolaki 2010, 193–201. 183 Hellenika 2.3.56: ἐκεῖνο δὲ κρίνω τοῦ ἀνδρὸς ἀγαστόν… 184 Hellenika 3.4.17–18: πῶς οὐκ εἰκὸς ἐνταῦθα πάντα µεστὰ ἐλπίδων ἀγαθῶν εἶναι; 185 Hellenika 5.2.6–7: καὶ τοῦτο µὲν εἰρήσθω µέγα τεκµήριον πειθαρχίας. 186 Hellenika 7.2.16: γενναίους δὲ καὶ ἀλκίµους πῶς οὐκ ἄν τις φαίη εἶναι τοὺς τοιαῦτα διαπραττοµένους; 187 Hellenika 7.5.15–17: ἐνταῦθα δὴ τούτων αὖ τὴν ἀρετὴν τίς οὐκ ἄν ἀγασθείη; 188 See Herodotos 4.129 and 6.43 for the expression θῶµα µέγιστον ἐρέω. 189 For wonders in Herodotos, see Hunzinger 1995; Munson 2001.

46

Melina Tamiolaki

5 CONCLUSION: TOWARDS AN EMOTIONALISATION OF HISTORY? The foregoing discussion shows that Xenophon’s Hellenika places a greater emphasis on the emotions than the histories of his predecessors: emotional vocabulary, detail and enargeia are more prominent in his work. An intriguing question that arises from this investigation is whether and to what extent Xenophon’s work foreshadows Hellenistic historiography. Given that Xenophon’s ideas about leadership had currency during the Hellenistic period,190 it would be worth exploring possible connections with Hellenistic historiography as well. Yet this step of our inquiry is significantly hampered by the fact that Hellenistic historiography, with the exception of Polybios (who does not, however, represent mainstream Hellenistic historiography)191 is preserved only in fragments,192 so the comparison cannot be complete or systematic. Hence, what I will present below will rather be based on the general view about Hellenistic historiography; I will also have recourse to a few selected fragments as a test-case for my question. The basic traits of Hellenistic historiography are detail and enargeia, the emphasis on wonders and paradoxa (in the sense of sudden reversals of fortune)193 and the aim of arousing the sympathy of the readers.194 Now, Xenophon is interested in enargeia, certainly more so than his predecessors. Moreover, a quick look at the fragments of Douris and Phylarchos collected by Felix Jacoby reveals other convergences as well: the description of everyday scenes, the presence of dialogue, the emphasis on individual emotions, and the detailed descriptions of pathos.195 On the other hand, we do not find in Xenophon descriptions of wonders and, although his narrative contains reversals of fortunes,196 these are not dramatised very much and can hardly be characterised as sensational narratives. This leads us to an issue that I deliberately left for the end of this investigation: what was Xenophon’s aim concerning his readership? Did he aim to arouse their emotions, pity or sympathy, like the Hellenistic historians, for instance? Two observations can be made here: firstly, we cannot be certain about Xenophon’s intentions, except for what we can infer from his authorial comments; these suggest that he obviously wanted his readers to share his emotion in these instances (most of the time, admiration). Secondly, and more importantly, it has hopefully emerged from the previous discussion that Xenophon’s use of the emotions in his 190 Farber 1979; Azoulay 2004, 433–447. 191 The bibliography on Polybios has been recently increasing: see the dissertation of Herchenroeder 2010 (with previous bibliography), and now Miltsios (2013). 192 See Breisach 1994, 27–34 and Marincola 2007. Cf. Chaniotis 2013a and 2013b. 193 This trait is a Herodotean legacy. See Murray 1972. 194 See Chaniotis 2013b, for the tracing of these traits in Hellenistic decrees and oratory as well; see also his remarks in this volume, pp. 53–82. 195 Description of everyday scenes: Douris, FGrH 76 F 3; Phylarchos, FGrH 81 F 12 (6). Dialogue: Phylarchos, FGrH 21 (18). Emphasis on individual emotions: Phylarchos, FGrH 32 (33). Descriptions of pathos: Douris, FGrH 76 F 67; Phylarchos, FGrH 81 F53 (51). 196 Hellenika 5.4.1 is considered a characteristic passage which announces the reversal of fortune for the Lacedaemonians.

Emotions and Historical Representation in Xenophon’s Hellenika

47

historical narrative is subtle and aims at something more than the mere pleasure or pity of the reader; it aims to promote a specific historical explanation; in sum, it aims to persuade.197 We do not have enough evidence from the sensational Hellenistic historians to be able to determine whether this significantly differentiates Xenophon from them, as Polybios’ criticism on ‘tragic historiography’ might lead us to believe. It is quite possible that Xenophon’s connection between emotion and persuasion exercised an influence on later historiography. Furthermore, it might be relevant that there are events for which Xenophon could have provided sensational narratives, but he carefully avoids doing so: we already observed the avoidance of emotional excesses and the balanced narrative about the amnesty. Other examples include the Lacedaemonians’ sparing of the Athenians after their defeat at Aigos Potamoi, which is described as follows:198 The Spartans, however, said they would not enslave a Greek city (Λακεδαιµόνιοι δὲ οὐκ ἔφασαν πόλιν Ἑλληνίδα ἀνδραποδιεῖν) that had accomplished so much good for Greece during the time of its greatest dangers; they preferred, rather, to offer peace to Athens upon the following conditions. ...

This event could have also given the opportunity for emotional elaboration based on pity (a word, which is not invoked at all). Another example is the speech against the tyrant Euphron, pronounced a little before his condemnation. Although this event would again allow a great emotional exploitation, based for instance on the well-known hatred for the tyrants, this speech is very pragmatic and does not have recourse to emotions.199 The examples can be multiplied. Overall, our investigation underlines the transitional character of Xenophon’s Hellenika concerning the topic of the emotions: on the one hand, Xenophon introduces innovations in terms of vocabulary, emotional narrative and theorisation about the emotions. On the other hand, he follows his predecessors (and specifically Thucydides) in that his aim is not so much to arouse the emotions of his readers, but rather to make them think about a historical interpretation. Finally, his greater interest in individual emotions contributes to a renewal of the genre of historiography, creating a certain mixture with the neighboring genres of biography and enkomion. Emotions thus emerge as another important vehicle for Xenophon’s experimentation. BIBLIOGRAPHY Abbamonte, G., L. Miletti, and L. Spina (eds.) (2009) Discorsi alla prova, Naples. Allan, R. J. (2013) History as Presence. Time, Tense and Narrative Modes in Thucydides, in Tsakmakis and Tamiolaki (eds.) 2013, 371–389. Azoulay, V. (2004) Xénophon et les grâces du pouvoir. De la charis au charisme, Paris. Baragwanath, E. (2008) Motivation and Narrative in Herodotus, Oxford.

197 Cf. Nünning 2008. 198 Hellenika 2.2.20. 199 Hellenika 7.3.7.

48

Melina Tamiolaki

––– (2012) A Noble Alliance: Herodotus, Thucydides and Xenophon’s Procles, in Foster and Lateiner (eds.), Thucydides and Herodotus, Oxford, 316–344. Beck, H. (1997) Polis und Koinon. Untersuchungen zur Geschichte und Struktur der griechischen Bundesstaaten im 4. Jahrhundert v. Chr., Stuttgart. Ben-Ze’ev, A. (2000) The Subtlety of Emotions, Cambridge, Mass./London. Bétant, E. A. (1969) Lexicon Thucydideum, Hildesheim/New York. Borod, J. C. (2000) The Neuropsychology of Emotion, Oxford. Braund, S. and G. W. Most (2003) (eds.) Ancient Anger: Perspectives from Homer to Galen, Cambridge. Breisach, E. (1994) Historiography Ancient, Mediaeval, and Modern, Chicago. Buckler, J. (1982) Alliance and Hegemony: The Case of the Theban Hegemony, Ancient World 5, 79–89. Cairns, D. L. (1993) Aidôs: The Psychology and Ethics of Honour and Shame in Ancient Greek Literature, Oxford. ––– (2003) Ethics, Ethology, Terminology: Iliadic Anger and the Cross-Cultural Study of Emotion, in Braund and Most (eds.) 2003, 11–49. ––– (2005) (ed.) Body Language in the Greek and Roman Worlds, Swansea. ––– (2008) Look Both Ways: Studying Emotion in Ancient Greek, Critical Quarterly 50, 43–63. ––– (2009) Weeping and Veiling: Grief, Display and Concealment in Ancient Greek Culture, in Fögen (ed.) 2009, 37–57. Cartledge, P. (2003) The Spartans, Oxford. Chaniotis, A. (ed.) (2012a) Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, Stuttgart. ––– (2012b) Unveiling Emotions in the Greek World: Introduction, in Chaniotis (ed.) 2012a, 11– 31. ––– (2012c) Moving Stones: The Study of Emotions in Greek Inscriptions, in Chaniotis (ed.) 2012a, 91–129. ––– (2013a) Emotional Language in Hellenistic Decrees and Hellenistic Histories in M. Mari and J. Thornton (eds.), Parole in movimento. Linguaggio politico e lessico storiografico nel mondo ellenistico, Pisa, 339–352. ––– (2013b) Paradoxon, Enargeia, Empathy: Hellenistic Decrees and Hellenistic Oratory, in K. Kremmydas and K. Tempest (eds.), Hellenistic Oratory: Continuity and Change, Oxford, 201–216. ––– (2013c) Normen stärker als Emotionen? Der kulturhistorische Kontext der griechischen Amnestie, in K. Harter-Uibopuu and F. Mitthof (eds.), Vergeben und Vergessen? Amnestie in der Antike. Akten des ersten Wiener Kolloquiums zur Antiken Rechtsgeschichte, Wien, 27.– 28.10.2008, Vienna, 47–70. ––– (forthcoming) Creating Emotional Community Through Text, in E. Sanders and M. Johncock (eds.), Emotion and Persuasion in Classical Antiquity. Cornford, F. M. (1907) Thucydides Mythistoricus, London. Darwin, C. (1872) The Expression of the Emotions in Man and Animals, London. Dillery, J. (1995) Xenophon and the History of His Times, London. Dixon, T. (2003) From Passions to Emotions, Cambridge. Elias, N. (1976) Über den Prozess der Zivilisation. Soziogenetische und psychogenetische Untersuchungen, Frankfurt. Elster, J. (1999) Alchemies of the Mind: Rationality and the Emotions, Cambridge. Erskine, A. (2014) Roman Anger in Polybius, in D. Cairns and L. Fulkerson (eds), Emotions Between Greece and Rome, London (forthcoming). Farber, J. J. (1979), The Cyropaedia and Hellenistic Kingship, American Journal of Philology 100, 479–514. Febvre, L. (1939–41) La sensibilité et l’histoire : Comment reconstituer la vie affective d’autrefois? Annales d’histoire sociale 3.1/2, 5–20.

Emotions and Historical Representation in Xenophon’s Hellenika

49

Fögen, T. (ed.) (2009) Tears in the Graeco-Roman World, Berlin. Fortenbaugh, W. W. (2002) Aristotle on Emotion, London. Franks, D. D. (2010) Neurosociology: the Nexus Between Neuroscience and Social Psychology, New York. Frevert, U. (2009) Does Trust Have a History? Max Weber Lecture 1, 1–16. ––– (2011) Emotions in History – Lost and Found, Budapest. Fridja, N. H., A. S. R. Manstead, and S. Bem (eds.) (2000) Emotions and Beliefs: How Feelings Influence Thoughts, Cambridge. Funke, P. and N. Luraghi (eds.) (2009) The Politics of Ethnicity and the Crisis of the Peloponnesian League, Washington DC/Cambridge Mass. Gautier, L. (1911) La langue de Xénophon, Dissertation, Geneva. Gray, V. (1981) Dialogue in Xenophon’s Hellenica, Classical Quarterly 31, 321–334. ––– (1987) Mimesis in Greek Historical Theory, American Journal of Philology 118, 467–486. ––– (1989) The Character of Xenophon’s Hellenica, London. ––– (2004) Le Socrate de Xénophon et la démocratie, Etudes Philosophiques 2, 141–176. Grethlein, J. (2013) The Presence of the Past in Thucydides, in Tsakmakis and Tamiolaki (eds.) 2013, 91–118. Grimaldi, W. M. A. (1988) Aristotle, Rhetoric I: a Commentary, New York. Halliwell, S. (1990) Traditional Greek Conceptions of Character, in C. B. R. Pelling (ed.), Characterization and Individuality in Greek Literature, Oxford. Harré, R. (ed.) (1986) The Social Construction of Emotions, London. Harris, W. V. (2001) Restraining Rage: The Ideology of Anger Control in Classical Antiquity, Cambridge, Mass./London. Harris, W. V. (2010) History, Empathy and Emotions, Antike und Abendland 56, 1–23. Harrison, T. (2003) The Cause of Things: Envy and the Emotions in Herodotos’ Histories, in D. Konstan and N. K. Rutter (eds.), Envy, Spite and Jealousy: the Rivalrous Emotions in Ancient Greece, Edinburgh, 143–163. Herchenroeder, L. (2010) Hellenistic Historiography and the Sciences. Practices and Concepts in Polybius’ Histories, Dissertation, University of South California. Hitzer, B. (2011), Emotionsgeschichte – ein Anfang mit Folgen, Humanities-Sozial-und Kulturgeschichte 23 November 2011: http://hsozkult.geschichte.hu-berlin.de/forum/2011-11-001. Hornblower, S. (1994) Narratology and Narrative Techniques in Thucydides, in S. Hornblower (ed.), Greek Historiography, Oxford, 131–166. ––– (1996) A Commentary on Thucydides, vol. 2, Oxford. Huart, P. (1968) Le vocabulaire de l’analyse psychologique chez Thucydide, Paris. Huizinga, J. (1924) Herbst des Mittelalters: Studien über Lebens- und Geistesformen des 14. und 15. Jahrhunderts in Frankreich und in den Niederlanden, Munich. Hunzinger, C. (1995) La notion de thôma chez Hérodote, Ktema 20, 47–70. Kalimtzis, K. (2012) Taming Anger: The Hellenic Approach to the Limitations of Reason, London. Konstan, D. (2001) Pity Transformed, London. ––– (2003a) Translating Ancient Emotions, Acta Classica 46, 5–19. ––– (2003b) Aristotle on Anger and the Emotions: the Strategies of Status, in Braund and Most (eds.), 99–120. ––– (2006) The Emotions of the Ancient Greeks: Studies in Aristotle and Classical Literature, Toronto/Buffalo/London. Kotsifou, C. (2012) Emotions and Papyri: Insights into the Theatre of Human Experience in Antiquity, in Chaniotis (ed.) 2012a, 39–90. Kroeker, F. (2009) Xenophon as a Critic of the Athenian Democracy, History of Political Thought 30, 197–228. Kuzmics, H. (2009) Sociology as Narrative: Examples of Sociological Language in ‘Classic’ Texts, in D. Hopkins, J. Kleres, H. Flam, and H. Kuzmics (eds.), Theorizing Emotions. Sociological Explorations and Applications, Frankfurt, 95–119.

50

Melina Tamiolaki

Lampadaridou, M. (2013) Οι χειρονοµίες στην Κύρου Παιδεία του Ξενοφώντα [Gestures in Xenophon’s Cyropaedia] Seminar Paper, University of Crete. Lateiner, D. (2005) The Pitiers and the Pitied in Herodotus and Thucydides, in Sternberg (ed.) 2005a, 67–97. ––– (2009) Tears and Crying in Hellenistic Historiography: Dacryology from Herodotus to Polybius, in Fögen (ed.) 2009, 105–134. Lattimore, S. (1998) The Peloponnesian War. Thucydides Translated with Introduction and Notes, Indianapolis/Cambridge. Lévy, E. (2003) Sparte, Paris. Loraux, N. (1997) La cité divisée, Paris. MacMullen, R. (2004) Les émotions dans l’histoire, ancienne et moderne, Paris. Marincola, J. (2003) Beyond Pity and Fear: The Emotions of History, Ancient Society 33, 285– 315. ––– (2007) Universal Histories from Ephorus to Diodorus, in J. Marincola (ed.) 2007, Blackwell Companion to Greek and Roman Historiography, London, 171–179. Miltsios, N. (2013) The Shaping of Narrative in Polybius, Berlin. Montgomery, H. (1965) Gedanke und Tat: zur Erzählungstechnik bei Herodot, Thukydides, Xenophon und Arrian, Lund. Morrison, J. (2006) Interaction of Speech and Narrative in Thucydides, in Rengakos and Tsakmakis (eds.) 2006, 251–278. Munson, R. V. (2001) Telling Wonders: Ethnographic and Political Discourse in the Work of Herodotus, Ann Arbor. Munteanu, D. L. (ed.) (2011) Emotion, Genre and Gender in Classical Antiquity, London. Murray, O. (1972) Herodotus and Hellenistic Culture, Classical Quarterly 22/2, 200–213. Nagy, P. (2007) Les émotions et l’historien: de nouveaux paradigmes, in P. Nagy (ed.), Critique (émotions médiévales), Paris, 10–22. Neumann, G. (1965) Gesten und Gebärden in der griechischen Kunst (Berlin) Nicolai, R. (2006) Thucydides Continued, in Rengakos and Tsakmakis (eds.) 2006, 691–719. Nünning, V. (2008) Wie überzeugt Literatur ? Eine kleine Rhetorik des Erzählens, in A. Chaniotis, A. Kropp, and C. Steinhoff (eds.), Überzeugungstrategien, Heidelberg, 93–104. Nussbaum, M. C. (2001) Upheavals of Thought. The Intelligence of Emotions, Cambridge. Oatley, K. (1992) Best Laid Schemes: The Psychology of Emotions, Cambridge. Ollier, F. (1943) Le mirage spartiate, Paris. Pausch, D. (2010) ‘Und seine Worte waren ungefähr die folgenden’… Reden in der antiken Geschichtsschreibung zwischen Verwendung und Problematisierung, in U. Tischer and A. Binternagel (eds.), Fremde Rede – Eigene Rede. Zitieren und verwandte Strategien in antiker Prosa, Frankfurt am Main/Berlin/Bern, 35–57. Plamper, J. (2010) The History of Emotions: An Interview with William Reddy, Barbara Rosenwein and Peter Stearns, History and Theory 49, 237–265. Powell, A. and S. Hodkinson (eds.) (1994) The Shadow of Sparta, London/New York. Powell, J. E. (1966) A Lexicon to Herodotus, Hildesheim. Pownall, F. (1998) Condemnation of the Impious in Xenophon’s Hellenica, Harvard Theological Review 91, 251–277. Prochasson, C. (2008) L’empire des émotions. Les historiens dans la mêlée, Paris. Radermacher, L. (1947) Weinen und Lachen. Studien über antikes Lebensgefühl, Vienna. Rapp, C. (2002) Rhetorik. Aristoteles, Berlin. Reddy, W. (2001) Navigation of Feeling, Cambridge. Rengakos, A. and A. Tsakmakis (2006) (eds.) Brill’s Companion to Thucydides, Leiden/Boston. Rhodes, P. and R. Osborne (2003) Greek Historical Inscriptions 404–323 BC, Oxford. Riedinger, J.-C. (1991) Etude sur les Helléniques de Xénophon. Xénophon et l’histoire, Paris. Romilly, J. de (1956) La crainte dans l’œuvre de Thucydide, Classica e Mediaevalia 17, 119–127. ––– (1966) Histoire et raison chez Thucydide, Paris.

Emotions and Historical Representation in Xenophon’s Hellenika

51

––– (1988) Le conquérant et la belle captive, Bulletin de l’Association Guillaume Budé 1, 1–15. Rood, T. (2004) Xenophon and Diodorus: Continuing Thucydides, in C. Tuplin (ed.), Xenophon and His World, Stuttgart, 341–395. Rosenwein, B. (2002) Worrying about Emotions in History, Americal Historical Review 107/3 821–845. ––– (2006) Emotional Communities in the Early Middle Ages, Ithaca. Russel, J. A. (1980) A Circumplex Model of Affect, Journal of Personality and Social Psychology 39, 1161–1178. ––– (1989) Measures of Emotion, in R. Plutchik and H. Kellerman (eds.), The Measurement of Emotions, San Diego, 83–111. Sanders, E. (2012) Beyond the Usual Suspects: Literary Sources and the Historian of Emotions, in Chaniotis (ed.) (2012a), 151–173. Sauge, A. (1992) De l’épopée à l’histoire: fondements de la notion d’historié, Frankfurt/Bern/New York/Paris. Scheve, C. von (2011) Die soziale Konstitution und Funktion von Emotion: Akteur, Gruppe, normative Ordnung, Zeitschrift für die Erziehungswissenschaft 14, 207–222. Schneider, C. (1974) Information und Absicht bei Thukydides. Untersuchung zur Motivation des Handelns, Göttingen. Schwarz-Friesel, M. (2007) Sprache und Emotion, Tübingen. Stahl, H.-P. (2013) The Dot on the ‘i’: Thucydidean Epilogues, in Tsakmakis and Tamiolaki (eds.) 2013, 309–328. Stearns, P. N. and C. Z. Stearns (1985) Emotionology. Clarifying the History of Emotions and Emotional Standards, American Historical Review 90/4, 813–830. Sternberg, R. H. (ed.) (2005a) Pity and Power in Ancient Athens, Cambridge. ––– (2005b) The Nature of Pity, in Sternberg (ed.) 2005a, 15–47. Strassler, R. B. and J. Marincola (2009) The Landmark Xenophon’s Hellenika, New York. Stueber, K. R. (2008) Reasons, Generalizations, Empathy, and Narratives: the Epistemic Structure of Action Explanation, History and Theory 47, 31–43. Tamiolaki, H. (2006) Modèles individuels et collectifs chez Hérodote : un exemple de la formation de l’identité grecque, Bulletin de l’Association Guillaume Budé 65, 17–39. ––– (2007) La réflexion sur la liberté et l’esclavage chez Hérodote, Thucydide et Xénophon, Dissertation, Paris-Sorbonne. ––– (2008) Les Helléniques entre tradition et innovation. Aspects de la relation intertextuelle de Xénophon avec Hérodote et Thucydide, Cahiers des Études Anciennes 45, 15–52. ––– (2010) Liberté et esclavage chez les historiens grecs classiques, Paris. ––– (2013a) A Citizen as a Slave of the State? Oligarchic Perceptions of Democracy in Xenophon, Greek Roman and Byzantine Studies 53, 31–50. ––– (2013b) Ascribing Motivation in Thucydides. Between Historical Research and Literary Representation, in Tsakmakis and Tamiolaki (eds.) 2013, 41–72. ––– (forthcoming a) A l’ombre de Thucydide? Les discours des Helléniques et leur influence thucydidéenne, in P. Pontier (ed.), Xénophon et la rhétorique, Paris. ––– (forthcoming b) Lucian on Truth and Lies in Ancient Historiography. The Theory and its Limits, in L. Hau and I. Ruffell (eds.), Pluralising the Past, Swansea. Tatum, J. (1989) Xenophon’s Imperial Fiction. On the Education of Cyrus, Princeton. Tsakmakis, A. and M. Tamiolaki (eds.) (2013) Thucydides Between History and Literature, Berlin. Tuplin, C. (1993) The Failings of Empire: A Reading of Xenophon, Hellenica 2.3.11–7.5.27, Stuttgart. Vendrell Ferran, Í. (2010) Literarische Fiktion und Fiktionale Gefühl, in G. Koch, M. Vöhler, and C. Voß (eds.), Mimesis und ihre Künste, Munich, 91–108. Walbank, F. W. (1960) History and Tragedy, Historia 9, 216–234. ––– (1972) Polybius, Berkley/Los Angeles.

52

Melina Tamiolaki

––– (1990) Profit or Amusement. Some Thoughts on the Motives of Hellenistic Historians, in H. Verdin, G. Schepens and E. de Keyser (eds.), Purposes of History. Studies in Greek Historiography from the 4th to the 2nd Centuries B.C., Leuven, 253–266. Walker, A. D. (1993) Enargeia and the Spectator in Greek Historiography, Transactions of the American Philological Association 123, 353–377. Westlake, H. D. (1968), Individuals in Thucydides, Cambridge. Wolpert, A. (2002) Remembering Defeat. Civil War and Civic Memory in Ancient Athens, Baltimore/London. Woodman, A. J. (1988) Rhetoric in Classical Historiography: Four Studies, London.

EMPATHY, EMOTIONAL DISPLAY, THEATRICALITY, AND ILLUSION IN HELLENISTIC HISTORIOGRAPHY Angelos Chaniotis 1 IN THE MOOD Few books of Hellenistic history capture the mood of the Hellenistic world as some of Cavafy’s poems, inspired by the poet’s readings of ancient sources. The mood of a festive day is the subject of his Alexandrian Kings written in 1912 and referring to the ceremony in which Kleopatra’s children were acclaimed kings. The Alexandrians turned out in force to see Cleopatra’s children, Kaisarion and his little brothers, Alexander and Ptolemy, who for the first time had been taken out to the Gymnasium, to be proclaimed kings there before a brilliant array of soldiers. Alexander: they declared him king of Armenia, Media, and the Parthians. Ptolemy: they declared him king of Cilicia, Syria, and Phoenicia. Kaisarion was standing in front of the others, dressed in pink silk, on his chest a bunch of hyacinths, his belt a double row of amethysts and sapphires, his shoes tied with white ribbons prinked with rose-colored pearls. They declared him greater than his little brothers, they declared him King of Kings. The Alexandrians knew of course that this was all mere words, all theatre. But the day was warm and poetic, the sky a pale blue, the Alexandrian Gymnasium a complete artistic triumph, the courtiers wonderfully sumptuous, Kaisarion all grace and beauty (Cleopatra’s son, blood of the Lagids); and the Alexandrians thronged to the festival full of enthusiasm, and shouted acclamations in Greek, and Egyptian, and some in Hebrew, charmed by the lovely spectacle –

54

Angelos Chaniotis though they knew of course what all this was worth, what empty words they really were, these kingships. Translated by Edmund Keeley and Phillip Sherrard.

Through his vivid description of the setting and the appeal to our senses – the warmth of the day, the colourful spectacle, the voices of the crowd – Cavafy makes his reader an eyewitness to this ceremony. But unlike the Alexandrian audience, who is aware of the theatrical display of power but unaware of what the future shall bring, Cavafy’s reader not only knows too ‘what empty words they really were, these kingships’, but he also knows how short-lived Anthony and Kleopatra’s fortune will be and what tragedies lie ahead. If this poem has a tragic quality, it is because despite our safe distance from the showdown at Alexandria we are indirectly warned that our world may also be subject to similar disasters. Realising that there is a shared human fate, we are invited to feel compassion for the Alexandrians and fear for ourselves. Several other ‘Hellenistic’ poems in Cavafy’s work share the same themes of dramatic peripeteia, theatrical display, and illusion, such as King Demetrios, Return from Greece, The Seleucid’s displeasure, and Philhellen. Cavafy received his inspiration from his readings of sources,1 especially Plutarch, whose Hellenistic lives derive directly or indirectly from Hellenistic narratives. As a perceptive reader he has captured the essence of Hellenistic historiography, which has often – and, in my view, correctly – been characterised as ‘tragic’.2 My aim in this study has a certain affinity to that of Cavafy: I shall also attempt to capture some of the mood of Hellenistic historiography. Fortunately for the reader I will not do this through poetry but through the discussion of elements of emotional display and arousal of emotion in several selected passages. I shall argue that – Hellenistic historians showed great interest in the description of emotions in historical narratives; – they attempted to arouse emotions and empathy through their narratives; – they described the techniques through which the ‘protagonists’ of their narratives tried to control the emotions of their contemporary audiences – especially theatrical behaviour and illusion; – all this corresponds to contemporary trends that can be observed in Hellenistic documentary sources (decrees).

1

2

The ‘Hellenistic’ poems are one of the most important groups among Cavafy’s ‘historical’ works. On Cavafy’s historical poetry, see Bowra 1967; Dallas 1974; Beaton 1983; Jusdanis 1987; Keeley 1995, 75–131; Hirst 1995. Still readable: Walbank 1938, 1955, and 1960. For some criticism on the concept of ‘tragic historiography’, see Marincola 2010. See also Schepens 2004 and van der Stockt 2004.

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

55

2 FEAR AND ILLUSION IN PELLENE In the Life of Aratos, Plutarch narrates the sack of Pellene in the Peloponnese around 241 BCE. The Aitolians suddenly attacked and seized the small Achaian town, immediately beginning to plunder it. While the soldiers were fighting with one another over the booty, the officers seized the women. By putting their own helmets on the women’s heads, they showed to whom each woman belonged. Aratos, the Achaian general, exploited precisely this moment of the enemy’s victory. Taking advantage of the disorder he attacked the town:3 In the midst of this confusion (ταράχῳ), one of the captive women, it occurred by chance (ἔτυχε) that the daughter of Epigethes, a distinguished man, and herself conspicuous for her beauty and the stateliness of her body, was sitting in Artemis’ sanctuary, where she had been placed by the officer who had seized her for himself and had placed his three crested helmet upon her head. But suddenly (ἄφνω) she ran forth towards the tumult, and as she stood in front of the gate of the sanctuary and looked down upon the combatants from the high place (ἄνωθεν), with the three crested helmet on her head, she appeared to the citizens themselves as a vision of more than human majesty (θέαµα σεµνότερον ἢ κατ᾿ ἄνθρωπον ἐφάνη), while the enemy thought they saw an apparition from heaven and were struck with amazement and terror, so that no one among them thought of defending himself.

This text exemplifies both the problems we face in the study of Hellenistic historiography and some of the features that I would like to discuss. Plutarch does not reveal his source. He mentions, however, that this narrative was contradicted by a version given by the inhabitants of Pellene: according to them, the priestess of Artemis drove the enemies away by carrying the goddess’ statue. None of this was mentioned in a third source, Aratos’ memoirs, which attributed the victory to his attack; it was Aratos’ version that was commemorated in a painting. We can assume with some certainty that Plutarch was contrasting different sources, without merging different narratives into one. Yet we do not know to what extent he adopted the narrative structure and the formulations of the Hellenistic author, who narrated the story of Epigethes’ daughter.4 We may not know this, but if we compare this narrative with Hellenistic narratives, we can argue that there is hardly anything in it that could not have been included in the narrative of a Hellenistic author. As for the name of the Hellenistic author in whom Plutarch’s story ultimately originates, Phylarchos is a good suspect. He is quoted by Plutarch in Ara-

3

4

Plutarch, Aratos 31f. (after Phylarchos?): ἐν τούτῳ δὲ τῷ ταράχῳ µία τῶν αἰχµαλώτων, Ἐπιγήθους ἀνδρὸς ἐνδόξου θυγάτηρ, αὐτὴ δὲ κάλλει καὶ µεγέθει σώµατος εὐπρεπής, ἔτυχε µὲν ἐν τῷ ἱερῷ καθεζοµένη τῆς Ἀρτέµιδος, οὗ κατέστησεν αὐτὴν ὁ ἐπιλεκτάρχης ἑλὼν ἑαυτῷ καὶ περιθεὶς τὴν τριλοφίαν, ἄφνω δὲ ἐκδραµοῦσα πρὸς τὸν θόρυβον, ὡς ἔστη πρὸ τῶν θυρῶν τοῦ ἱεροῦ καὶ κατέβλεψεν εἰς τοὺς µαχοµένους ἄνωθεν ἔχουσα τὴν τριλοφίαν, αὐτοῖς τε τοῖς πολίταις θέαµα σεµνότερον ἢ κατ᾿ ἄνθρωπον ἐφάνη, καὶ τοῖς πολεµίοις φάσµα θεῖον ὁρᾶν δοκοῦσι φρίκην ἐνέβαλε καὶ θάµβος, ὥστε µηδένα τρέπεσθαι πρὸς ἀλκήν. Discussion of this passage in Chaniotis 2005a, 208f.; see also the remarks of Douglas Cairns in this volume (pp. 85–88 and 91). For the historical context see Scholten 2000, 123–127. On Plutarch’s adaptation of his sources see Pelling 1995 and 2002, 91–115.

56

Angelos Chaniotis

tos’ Life and elsewhere;5 and this is the manner in which he would have described such a dramatic event, writing history for the purpose of effect, and seeking to harrow up the emotions of his readers by the narrative of deeds of violence and horror – these are Polybios’ accusations against him.6 None of this can be regarded as proof that we have a fragment of Phylarchos. It is because of such problems that much of the lost Hellenistic historiography, which has only indirectly left its trace in later sources (especially in Plutarch and Pausanias), is often disregarded in studies of Hellenistic historiography. Research on the works of Hellenistic historians focuses on Polybios and on the more extensive fragments of the historians of Alexander and the later periods. Unavoidably, I shall do the same, but I will occasionally discuss passages in Plutarch that in my view reflect the mood of Hellenistic historiography. The second reason for the selection of this particular passage at the beginning of this study is the part played by emotions, both directly and indirectly. There is a direct reference to emotions: the terror felt by the Aitolians for what they thought was divine epiphany (φρίκη).7 Far more interesting than the description of emotions in this narrative is the arousal of emotions through the narrative. Our anonymous historian treated a historical event – the battle at Pallene. But through the subtle use of language and images, he presented this historical event as a drama staged by Fortune (ἔτυχε). By highlighting a seemingly unimportant detail – the placing of a helmet on the captive women –, he shows how the insolence (hybris) of the victorious enemies prepared their destruction. In the manner in which the author chose to narrate the story, the arrogant and wanton officer made the defenceless woman his own ruin, by placing his helmet on her head and bringing her to the sanctuary. The author constructs a subtle contrast between the confusion of the battle (τάραχος) and the behaviour of an individual figure – Epigethes’ daughter. This woman stands out and appears in a costume (the soldier’s helmet), which changes her identity and makes her resemble Artemis. And as she comes into sight, almost like an actress, on the high stage (ἄνωθεν) formed by the gate of the sanctuary, she becomes a dea ex machina. Interestingly, the daughter of Epigethes does not have a name in this narrative; she simply is ‘the daughter of Epigethes’. In a narrative which is so carefully composed, the mention of a name and the omission of another are anything but coincidental. It is part of the author’s selection of what he presents and how he presents it. Epigethes is not at all a common name. There are only four attesta5 6

7

Phylarchos, FGrHist 81 F 52 (Plutarch, Aratos 38); F 51, 59, and 60 (Plutarch, Kleomenes 5, 28, and 30). Polybios 2.56.7–8: ‘And being eager to stir the hearts of his readers to pity (εἰς ἔλεον) and to enlist their sympathies by his story, he talks of women embracing, tearing their hair, and exposing their breasts; and again of the tears and lamentations of men and women, led off into captivity along with their children and aged parents. And this he does again and again throughout his whole history, by way of bringing the terrible scene vividly before his readers (πρὸ ὀφθαλµῶν τιθέναι τὰ δεινά).’ On the meaning and significance of φρίκη see the study of Douglas Cairns in this volume (pp. 85–105).

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

57

tions in the six volumes of the Lexicon of Greek Personal Names. Besides this Epigethes (the only bearer of this name in the Peloponnese), there are two attestations in Athens and one in Macedonia; we may add the related name Epigethides in Tenos. Epigethes derives from ἐπιγηθέω, ‘to rejoice, to triumph over’. The ancient reader realised that Pellene was saved by the anonymous daughter of the ‘joyfully triumphant man’. I do not claim that the author made this name up; but he certainly chose to mention it precisely because it was yet another wonderful coincidence, a game of Tyche. There is a different version of the story, narrated by Polyainos in his Stratagems,8 having the battle take place during a festival of Athena. According to the custom of that day, her priestess, the tallest and most beautiful girl, impersonated the goddess wearing panoply and helmet.9 As she watched the citizens being armed from the citadel, she appeared to the enemies like an epiphany of the goddess. This is, again, a technique of narrative that can be observed in ancient historiography. By mentioning the fact that a historical event took place on the day of a festival (or by inventing this detail), or by making a ritual the narrative setting of an incident, a historian associates with this incident ideas and perceptions connected with the festival or the ritual.10 In this case, a battle that takes place during a celebration for Athena Polias, the protector of the citadel, arouses the hope of divine protection, without freeing the reader from the anguish that a tragic irony might occur and the goddess would not do justice to her epithet. These passages illustrate the pleasure Hellenistic narrators and audiences had in the portrayal of the unexpected, their fascination with sudden reversals, and with the assimilation of life with a drama. That the perceived director of this drama of life is a superhuman force (Fortune, the gods) makes us forget that the true director of the dramas is no other than the historian himself through the conscious selection of words and images and their artful composition. The aim of the Hellenistic historians was not only (or primarily) to describe the emotions of the ‘protagonists’ of their dramas, but to arouse the emotions of their readers and their audiences (during their public lectures). Their technique was the creation of illusions. 3 HELLENISTIC ILLUSIONS Illusion is generally defined as the belief of an individual or a group that something has taken place, although it has not – or has not taken place in the way they 8

9

10

Polyainos, Strategemata 8.59: τῆς Ἀθηνᾶς ἱέρεια κατά τι νόµιµον ἐκείνης τῆς ἡµέρας πανοπλίαν ἔχουσα καὶ τρίλοφον κράνος, ἡ καλλίστη καὶ µεγίστη τῶν παρθένων, ἀπὸ τῆς ἀκροπόλεως ἀπέβλεπεν ἐς τὸ πλῆθος τῶν ὁπλιζοµένων πολιτῶν. The impersonation of gods by priests during festivals was quite common: see Deshours 2006, 135f.; Chaniotis 2009a, 204f. notes 39 and 40. Examples: Lucian, Alexander 38–40; Pausanias 1.43.2; Sokolowski 1962, no. 51 lines 121–125. Ma 1994 (Plutarch); Schmitzer 1998 (Xenophon); Chaniotis 2005b, 17–19 (general remarks); Icks 2011 (Roman historiography); cf. Bathrelou 2012 (New Comedy).

58

Angelos Chaniotis

think it has; or, the belief in the existence of an entity or a being, which does not really exist. The combatants in Pellene thought that they saw a divine apparition; in other words, they had an experience, which their contemporaries would have interpreted and described as an epiphany. Henk Versnel once wrote an article asking the question ‘What did ancient man see when he saw a god?’11 The narrative concerning Pellene gives an answer. Epiphanies, manifestations of divine presence and intervention in human affairs, were a common theme in Hellenistic historiography – in fact, we know of Hellenistic historians who only treated epiphanies in their works (for instance, Syriskos in Chersonesos Taurica).12 This, however, is not an epiphany stricto sensu; it is not the appearance of Artemis, but of an ordinary woman, who is thought to be more than human, although she is not. This illusion should be distinguished from an epiphany, but it should also be distinguished from intentional deception, as a comparison with a very similar episode narrated by a much earlier historian shows. Herodotos narrates that Peisistratos had a tall and beautiful woman dress like Athena, put her on a chariot and entered Athens with her, having the heralds declare that Athena brings Peisistratos back to Athens.13 Peisistratos’ trick is presented as the result of human agency, as a case of conscious deception. In Pellene, the human agent, the soldier who put his helmet on the woman’s head, did not aim at deception. Unlike Peisistratos, he did not want the woman to appear like a goddess, but his deed ultimately created an illusion. Whether there was also a divine agent behind all that – an intervention of Artemis, making the girl run to the sanctuary’s gate – is not stated and is a different matter. The Greek word for illusion and deception is apate. Already Gorgias had discussed the connection between the beguilement produced by drama and emotion. Under certain conditions, apate had for Gorgias an educational value.14 In the Hellenistic period the word apate is often closely associated with the aesthetic pleasure that a spectacle or a fascinating narrative can offer to an audience.15 Illu11 12 13

14

15

Versnel 1987. IOSPE I2 344; FgrHist 807; Chaniotis 1988, 300f.; Dana and Dana 2001–03, 99–102. Herodotos 1.60.4–5; cf. Aristotle, Athenaion Politeia 14.4. Among the various interpretations of this incident that of Sinos 1993 (staged epiphany exploited for political purposes) seems to me more plausible. For different views see Gernet 1968 (evocation of rituals of sacred marriage); Connor 1987 (reversal of the established order); Blok 2000 (celebration of a victory); Anderson 2003 (procession celebrating the dedication of a new building on the Acropolis). Gorgias F 82 Β 23 edd. Diels/Κrantz.: ἤνθησε δ’ ἡ τραγῳδία καὶ διεβοήθη, θαυµαστὸν ἀκρόαµα καὶ θέαµα τῶν τότ’ ἀνθρώπων γενοµένη καὶ παρασχοῦσα τοῖς µύθοις καὶ τοῖς πάθεσιν ἀπάτην, ὡς Γοργίας φησίν, ἣν ὅ τ’ ἀπατήσας δικαιότερος τοῦ µὴ ἀπατήσαντος καὶ ὁ ἀπατηθεὶς σοφώτερος τοῦ µὴ ἀπατηθέντος. ὁ µὲν γὰρ ἀπατήσας δικαιότερος ὅτι τοῦθ’ ὑποσχόµενος πεποίηκεν, ὁ δ’ ἀπατηθεὶς σοφώτερος· εὐάλωτον γὰρ ὑφ’ ἡδονῆς λόγων τὸ µὴ ἀναίσθητον. I owe this reference to Professor Stavros Tsitsiridis (Patras). Heracleides periegeta 1: φιλοσόφων παντοδαπῶν ψυχῆς ἀπάται; Polybios 4.20.5: οὐ γὰρ ἡγητέον µουσικήν ... ἐπ᾿ ἀπάτῃ καὶ γοητείᾳ παρεισῆχθαι τοῖς ἀνθρώποις; SEG XXXV 744: οὐ µόνον πρὸς τὴν τῆς εὐωχίας [φρ]ονῶν [χρείαν, ἀλλὰ καὶ] τὴν θέαν καὶ τὴν ἀπά[τη]ν [καὶ τὴν διά]χυσιν τῆς ψ[υχῆς]; Dionysios of Halicarnassus, On Thucydides 7.3: οὐχ

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

59

sion is not only a subject of Hellenistic historiography; it is also the result of the vivid manner of presentation. With the enargeia, the painting of vivid images through words, and through detailed and colourful descriptions that appeal to the reader’s senses, the historian makes the reader think that he is witnessing a scene.16 In this context, criticising the historian Phylarchos, Polybios uses the phrase ‘to place something in front of (the reader’s) eyes’:17 Eager to arouse the pity (ἔλεον) of his readers and to create empathy toward what was being said, he introduces scenes of women clinging to one another, tearing their hair and baring their breasts, and in addition he describes the tears and lamentations of men and women as they are led away in captivity together with their children and aged parents. And he does this again and again throughout his whole history trying each time to bring the terrible scenes in front of his readers’ eyes (πρὸ ὀφθαλµῶν τιθέναι τὰ δεινά).

The same expression is used by Aristotle in his description of the task of a tragic poet,18 and by Polybios and the anonymous author of a decree in Olbia in connection with the way orators arouse the emotions of their listeners through vivid descriptions.19 Spectators, readers, and listeners are transformed into eye-witnesses through the power of the vivid description. Exactly as a skillful Hellenistic artist (see below p. 80) could create the illusion that the viewers of a work of art see something that is not really there, the contemporary historians and orators manipulated the thoughts and emotions of their audiences by making them think that

16

17 18 19

ἥρµοττεν ἐγκαταµίσγειν τῇ διηγήσει τὰς θεατρικὰς γοητείας οὐδὲ πρὸς τὴν ἀπάτην ἁρµόττεσθαι τῶν ἀναγνωσοµένων, ἣν ἐκεῖναι πεφύκασι φέρειν αἱ συντάξεις, ἀλλὰ πρὸς τὴν ὠφέλειαν. On enargeia in historiography see Zangara 2007, esp. 55–89; in rhetorical theory: Webb 2009, 87–105; in Hellenistic oratory: Chaniotis 2013a; in Hellenistic poetry: Otto 2009. Plutarch rightly observes that already Thucydides was a master of enargeia (Moralia 347a): τῶν ἱστορικῶν κράτιστος ὁ τὴν διήγησιν ὥσπερ γραφὴν πάθεσι καὶ προσώποις εἰδωλοποιήσας. ὁ γοῦν Θουκυδίδης ἀεὶ τῷ λόγῳ πρὸς ταύτην ἁµιλλᾶται τὴν ἐνάργειαν, οἷον θεατὴν ποιῆσαι τὸν ἀκροατὴν καὶ τὰ γινόµενα περὶ τοὺς ὁρῶντας ἐκπληκτικὰ καὶ ταρακτικὰ πάθη τοῖς ἀναγινώσκουσιν ἐνεργάσασθαι λιχνευόµενος (‘the best historian is the one who makes his narrative like a painting through vivid representations of emotions and individuals. In his writing, Thucydides always strives for this vividness, wishing to make the reader a spectator and to put in his mind the emotions of fright and consternation experienced by the eyewitnesses’). I owe this reference to Professor Stavros Tsitsiridis (Patras). Polybios 2.56.6–8. For this passage see below note 65. Aristotle, Poet. 1455a23: πρὸ ὀµµάτων τιθέναι. Polybios 22.8.11: ταῦτ᾽ οὖν τιθεὶς τοῖς Ἀχαιοῖς πρὸ ὀφθαλµῶν ἠξίου τὸν Εὐµένη µὴ διάφορα προτείνοντα θηρεύειν τὴν τῶν Ἀχαιῶν εὔνοιαν (‘bringing these things in front of the eyes of the Achaians, he asked Eumenes not to try to catch the benevolence of the Achaians with distracting propositions’); IOSPE I2 32 B lines 22–27: ὧν ἕνεκεν συνελθὼν ὁ δῆµος διηγωνιακὼς καὶ τὸγ κίνδυνον τὸµ µέλλοντα καὶ τὰ δεινὰ πρὸ ὀφθαλµῶν ποιούµενος παρεκάλει πάντας τοὺς ἰσχύοντας βοηθῆσαι καὶ µὴ περιιδεῖν τὴν ἐκ πολλῶν τετηρηµένηµ πατρίδα ὑποχείριον γενοµένην τοῖς πολεµίοις (‘because of this, the people convened in an assembly in deep despair; the demos (i.e. the orators in the assembly) brought in front of their eyes the danger that lay ahead and the terrors in store, and called on all who were able-bodied to help and not to watch with indifference their native city being subjected by the enemy, after it had been preserved for many years’).

60

Angelos Chaniotis

they were seeing things that they had not experienced or that would happen in he future. What the historian selects to present and how he presents it is his stagecraft that leads the audience to make associations and feel strong emotions. Two passages in Polybios illustrate this. One of the most dramatic narratives in what survives from Hellenistic historiography is Polybios’ description of Philip V’s siege and sacking of Abydos in 201 BCE (16.30–34). At the outset of his narrative he describes the self-confidence of he Abydenoi (πιστεύοντες ἑαυτοῖς ... ἐρρωµένως). But then their fortune turned, and when Philip demanded their surrender, they swore to die fighting, together with their women and children. Similar oaths of heroic and desperate sacrifice had been taken in the past by the Phokians and the Akarnanians, as Polybios notes – and years later the Aphrodisians would assert that all men, women, children, and slaves were ready to risk all for the Romans, and that they preferred death to losing the rule of the Romans (88 BCE). The Phokians and the Akarnanians were ultimately saved, and also in Aprodisias no mass suicide became necessary. The Abydenes had another thing coming. Their desperate last battle, when the crosswall fell, and then the mass suicide of those who abided by the oath are described by Polybios in a vivid manner:20 Not only did the Abydenes who fought in the first ranks mount the bodies of their dying enemies and fight with the outmost courage, not only did they fight desperately with sword and spear, but whenever any of these weapons was damaged and became useless or when they were forced to drop it, they took hold of the Macedonians with their hands and threw them down in their armour, or breaking their pikes they hit them and stabbed them repeatedly with the fragments on the face and the exposed parts of the body, thus throwing them into total confusion. ... When he [Philip] saw the number and the fury of those who were killing themselves, the children, and the women, by cutting throats, burning, hanging, throwing into wells and off the roofs, he was amazed and distressed at what was happening; he announced that he granted a respite of three days to those who wished to hang themselves and cut their throats. The Abydenes, remaining faithful to their original decision and thinking that they had almost betrayed those who had fought and died for their country, by no means accepted to live, except those whose hands had been stayed by fetters or similar forcible mean. All the rest rushed without hesitation in whole families to their death.

20

Polybios 16.33.2–4 and 16.34.9–12: οὐ γὰρ µόνον ἐπὶ τοὺς θνήσκοντας τῶν πολεµίων ἐπιβαίνοντες ἠγωνίζοντο µετὰ παραστάσεως οἱ προκινδυνεύοντες τῶν Ἀβυδηνῶν, οὐδὲ τοῖς ξίφεσι καὶ τοῖς δόρασιν αὐτοῖς ἐµάχοντο παραβόλως, ἀλλ᾽ ὅτε τι τούτων ἀχρειωθὲν ἀδυνατήσειεν ἢ µετὰ βίας προοῖντ᾽ ἐκ τῶν χειρῶν, συµπλεκόµενοι τοῖς Μακεδόσιν οὓς µὲν ἀνέτρεπον ὁµοῦ τοῖς ὅπλοις, ὧν δὲ συντρίβοντες τὰς σαρίσας αὐτοῖς τοῖς ἐκείνων κλάσµασιν ἐκ διαλήψεως τύπτοντες αὐτῶν ταῖς ἐπιδορατίσι τὰ πρόσωπα καὶ τοὺς γυµνοὺς τόπους εἰς ὁλοσχερῆ διατροπὴν ἦγον. ... θεωρῶν δὲ τὸ πλῆθος καὶ τὴν ὁρµὴν τῶν σφᾶς αὐτοὺς καὶ τὰ τέκνα καὶ τὰς γυναῖκας ἀποσφαττόντων, κατακαόντων, ἀπαγχόντων, εἰς τὰ φρέατα ῥιπτούντων, κατακρηµνιζόντων ἀπὸ τῶν τεγῶν, ἐκπλαγὴς ἦν, καὶ διαλγῶν ἐπὶ τοῖς γινοµένοις παρήγγειλε διότι τρεῖς ἡµέρας ἀναστροφὴν δίδωσι τοῖς βουλοµένοις ἀπάγχεσθαι καὶ σφάττειν αὑτούς. οἱ δ᾽ Ἀβυδηνοί, προδιειληφότες ὑπὲρ αὑτῶν κατὰ τὴν ἐξ ἀρχῆς στάσιν, καὶ νοµίζοντες οἷον εἰ προδόται γίνεσθαι τῶν ὑπὲρ τῆς πατρίδος ἠγωνισµένων καὶ τεθνεώτων, οὐδαµῶς ὑπέµενον τὸ ζῆν, ὅσοι µὴ δεσµοῖς ἢ τοιαύταις ἀνάγκαις προκατελήφθησαν: οἱ δὲ λοιποὶ πάντες ὥρµων ἀµελλήτως κατὰ συγγενείας ἐπὶ τὸν θάνατον.

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

61

The historian draws our attention to the tragic quality of this account by using the word peripeteia, which in the context of drama describes sudden changes of fortune.21 In the case of the sudden change of fortune (περιπέτεια) of the Abydenes one feels inclined to blame Fortune (τύχη) at the most, who, as if in pity, set right at once the misfortunes of the aforementioned peoples [the Phokians and Akarnanians], by granting both victory and safety to those who had lost hope; in the case of the Abydenes, however, she chose to do the opposite. For the men were killed, the city was taken, and the children together with their mothers fell into the enemy’s hands.

A similar focus on sudden changes of fortune (peripeteia), this time combined with an interest in another dramatic feature – hybris and its punishment –, can be observed in Polybios’ description of the defeat and death of Machanidas, Sparta’s king (or regent), in 207 BCE.22 Machanidas is introduced at the hight of power. Filled with confidence, he regards the Achaian attack a fulfillment of his prayers to the gods (11.11.1); and at the beginning fortune seems to smile to him. His mercenaries prevail and the Achaian troops flee in disorder towards the city of Mantineia, and everything seems lost for them. In his moment of triumph Machanidas makes a fatal mistake: instead of remaining in the field in order to strike the decisive blow, he follows his mercenaries and runs after the fugitive enemies. The opportunity is recognized by the Achaian general Philpoimen, who lets the pursuers pass by and then occupies the field they left, thus cutting them off from the rest of their army. The army of Machanidas is forced into a ditch around the walls of Mantineia, where it perishes (11.16.2f.). Machanidas’ attempt to escape gives Polybios the opportunity for a masterly close-up:23 21

22 23

Polybios 16.32.5f.: διὸ καὶ µάλιστ᾽ ἄν τις ἐπὶ τῆς Ἀβυδηνῶν περιπετείας µέµψαιτο τῇ τύχῃ, διότι τὰς µὲν τῶν προειρηµένων συµφορὰς οἷον ἐλεήσασα παραυτίκα διωρθώσατο, περιθεῖσα τὴν νίκην ἅµα καὶ τὴν σωτηρίαν τοῖς ἀπηλπισµένοις, περὶ δ᾽ Ἀβυδηνῶν τὴν ἐναντίαν εἶχε διάληψιν. οἱ µὲν γὰρ ἄνδρες ἀπέθανον, ἡ δὲ πόλις ἑάλω, τὰ δὲ τέκνα σὺν αὐταῖς µητράσιν ἐγένετο τοῖς ἐχθροῖς ὑποχείρια. Polybios 11.11–18; more detailed discussion in Chaniotis 2005a, 193–196, with further examples. Polybios 11.17.7–18.7: καθ᾽ ὃν δὴ καιρὸν ὁ τύραννος ἀπογνοὺς τὴν διὰ τῆς γεφύρας ὁδὸν παρήλαυνε παρὰ τὴν τάφρον, ἐνεργῶς διάβασιν ζητῶν. ὁ δὲ Φιλοποίµην, ἐπιγνοὺς τὸν Μαχανίδαν ἀπό τε τῆς πορφυρίδος καὶ τοῦ περὶ τὸν ἵππον κόσµου, τοὺς µὲν περὶ τὸν Ἀναξίδαµον ἀπολείπει, παρακαλέσας τηρεῖν ἐπιµελῶς τὴν δίοδον καὶ µηδενὸς φείδεσθαι τῶν µισθοφόρων διὰ τὸ τούτους εἶναι τοὺς συναύξοντας αἰεὶ τὰς ἐν τῇ Σπάρτῃ τυραννίδας: αὐτὸς δὲ παραλαβὼν Πολύαινον τὸν Κυπαρισσέα καὶ Σιµίαν, οἷς ἐχρῆτο τότε παρασπισταῖς, ἐκ τοῦ πέραν τῆς τάφρου τὴν ἀντιπαραγωγὴν ἐποιεῖτο τῷ τυράννῳ καὶ τοῖς µετ᾽ αὐτοῦ: δύο γὰρ ἦσαν οἱ τότε τῷ Μαχανίδᾳ συµµίξαντες, Ἀρηξίδαµος καὶ τῶν µισθοφόρων εἷς. ἅµα δὲ τῷ τὸν Μαχανίδαν κατά τινα τόπον εὔβατον τῆς τάφρου, προσθέντα τοὺς µύωπας, βίᾳ τὸν ἵππον ἐπάγειν καὶ διαπερᾶν, συναγαγὼν ἐκ µεταβολῆς ὁ Φιλοποίµην αὐτῷ καὶ πατάξας τῷ δόρατι καιρίως, καὶ προσενεγκὼν τῷ σαυρωτῆρι πληγὴν ἄλλην ἐκ διαλήψεως, ἐν χειρῶν νόµῳ διέφθειρε τὸν τύραννον. τὸ δὲ παραπλήσιον ἐγίνετο καὶ περὶ τὸν Ἀρηξίδαµον ὑπὸ τῶν παρίππων. ὁ δὲ τρίτος ἀπογνοὺς τὴν διάβασιν διέφυγε τὸν κίνδυνον κατὰ τὸν τῶν προειρηµένων φόνον. πεσόντων δ᾽ ἀµφοτέρων, εὐθέως οἱ περὶ τὸν Σιµίαν, σκυλεύσαντες τοὺς νεκροὺς καὶ συναφελόντες ἅµα τοῖς ὅπλοις τὴν τοῦ τυράννου κεφαλήν, ἠπείγοντο πρὸς τοὺς διώκοντας, σπεύδοντες

62

Angelos Chaniotis Meanwhile the tyrant, losing hope of making his way across the bridge, rode along the ditch vigorously trying to find a crossing. Philopoimen, recognising Machanidas by his purple cloak and the ornaments of his horse, left Alexidamos and his men, ordering them to guard the passage carefully and spare none of the mercenaries, for they had always increased the power of the Spartan tyrannies. Taking with him Polyainos of Kyparissia and Simias, who were at that time his aides-de-camp, he followed the tyrant and those with him – two men had joined him, Arexidamos and one of the mercenaries – along the opposite side of the ditch. When Machanidas, on reaching a place where the ditch was easily passable, set spurs to his horse and forced it across, Philopoimen turned to meet him. He first wounded him seriously with his spear and added another wound with the lower end of it, thus killing the tyrant hand to hand. Arexidamos suffered the same fate at the hands of those who were riding with Philopoimen, but the third man, despairing of crossing, escaped the danger while the other two men were being slain. When both had fallen, Simias’ companions stripped the bodies of the dead, taking together with the armour also the head of the tyrant and rushing back to the pursuers, eager to show to the troops those proofs of the death of the enemies’ commander.

The tragic irony in this detailed account, which resembles a hunting scene – hunting was an elite occupation, familiar to Machanidas, Philopoimen, Polybios, and his readers –, could not have escaped the notice of the ancient reader. The purple garment and the luxurious horse trappings, arrogant symbols of superiority, betray Machanidas, as the pursuer becomes the pursued. The historian, again, draws our attention to the unexpected:24 Indeed we often see that those, who already seem to have gained the day, totally fail shortly afterwards, whereas those who at first seemed to have failed, turn the tables and unexpectedly succeed in everything by their intelligence.

In these cases, the Hellenistic historian provides so many details (names, numbers, colours, objects, places) that he creates for us the illusion that we are eyewitnesses to this incident. And as if dealing with the audience of a dramatic performance in the theatre, the historian invites us to notice the reversals of fortune (peripeteia) and the unexpected developments (paradoxon). Polybios attributed to the sudden changes of fortune (αἱ τῶν ἄλλων περιπέτειαι) great instructive value.25 What makes the First Punic War so interesting for Polybios is the fact that it demonstrated more and greater changes of fortune (peripeteiai) than any other war.26 The historian never tires of pointing us to unexpected developoments and sudden changes of fortune. When he explains how a pro-Macedonian group under Askondas and Neon prevailed in Boiotia in the late third century BCE during the reign of Antigonos Doson, he repeatedly highlights the unexpected elements (20.5.6–11): Askondas and Neon got the upper hand, owing to a sudden change of fortune (γενοµένης τινὸς περιπετείας); an

24

25 26

ἐπιδεῖξαι τοῖς ὄχλοις τὴν ἀπώλειαν τοῦ τῶν ὑπεναντίων ἡγεµόνος χάριν τοῦ πιστεύσαντας ἔτι µᾶλλον ἀνυπόπτως καὶ τεθαρρηκότως ποιήσασθαι τὸν ἐπιδιωγµὸν τῶν ὑπεναντίων ἕως τῆς Τεγεατῶν πόλεως. Polybios 11.14.4: ἰδεῖν γοῦν ἔστι πολλάκις τοὺς µὲν ἤδη δοκοῦντας πεπροτερηκέναι µετ᾽ ὀλίγον τοῖς ὅλοις ἐσφαλµένους, τοὺς δ᾽ ἐν ἀρχαῖς δόξαντας ἐπταικέναι πάλιν ἐκ µεταβολῆς παρὰ τὴν αὑτῶν ἀγχίνοιαν τὰ ὅλα παραδόξως κατωρθωκότας. E.g., Polybios 1.1.2; 1.35.7; 5.75.5; cf. Walbank 1938, 64; Sacks 1981, 132–144. Polybios 1.13.11; cf. 3.97.8.

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

63

unexpectedly (παράδοξος) low tide kept Macedonian ships on the land at Larymna; Neon did not attack the Macedonians ‘contrary to what they expected’ (παρὰ τὴν προσδοκίαν), thus winning the Macedonian king’s gratitude for having been spared ‘during this change of fortune’ (κατὰ τὴν περιπέτειαν).27 In this way, the reader only sees what the historian has selected for him to ‘watch’. By diverting our attention to fortune’s whims, the historian makes us forget that the director of the drama of life is not fortune but the historian himself: it is he, who presents the events in such a sequence and such a manner that peripeteiai and paradoxa become apparent. Reminded of the fact that reversals and paradoxes are a universal human experience, we feel compassion for the victims of fortune’s whims and fear for what might lie ahead in our lives. Exactly as Scipio shed tears seeing his defeated opponent Syphax brought before him in chains, ‘thinking of the man’s former prosperity and kingly state’,28 the historian’s readers are expected to respond to historical narratives that highlight changes of fortune in the same manner that audiences in the theatre respond to the fate of a drama’s protagonists. The historical narrative resembles the theatrical stage. On this stage, the ‘protagonists’ of the narratives (orators, statesmen, kings) try to control the emotions of their contemporary audiences through theatrical behaviour, to which I now turn. 4 THEATRICAL BEHAVIOR OF STATESMEN IN HELLENISTIC NARRATIVES In one of the few lengthy fragments of his history, Poseidonios, a polymath of the late Hellenistic period, describes in detail a speech delivered by Athenion, an Athenian statesman and supporter of Mithridates VI of Pontos, in 88 BCE.29 The 27

28 29

Cf. Polybios’ comments on the adventures of the Peloponnesian city of Aigeira: ‘the Aigeiratans lost their city by their negligence, and recovered it again by their courage and valour, beyond expectation (παραδόξως)’ (4.58.12). For a similar interest of Hellenistic decrees on paradoxa see Chaniotis 2005a, 207–212 (IOSPE I2 32 and 352) and 2013a, 203, 207–209. See also below pp. 75f. Diodoros 27.6. Poseidonios, Histories fr. 247 ed. Theiler (Athenaios V 212 b-e): ἔπεµψαν ἐπὶ τὴν ἀνακοµιδὴν αὐτοῦ ναῦς µακρὰς καὶ φορεῖον ἀργυρόπουν. ἀλλ᾿ εἰσῄειν ἤδη, καὶ σχεδὸν τὸ πλεῖστον µέρος τῆς πόλεως ἐπὶ τὴν ἐκδοχὴν αὐτοῦ ἐξεκέχυτο· συνέτρεχον δὲ πολλοὶ καὶ ἄλλοι θεαταὶ τὸ παράδοξον τῆς τύχης θαυµάζοντες, εἰ ὁ παρέγγραφος Ἀθηνίων εἰς Ἀθήνας ἐπ᾿ ἀργυρόποδος κατακοµίζεται φορείου καὶ πορφυρῶν στρωµάτων... συνέτρεχον οὖν πρὸς τὴν θέαν ταύτην ἄνδρες, γυναῖκες, παῖδες τὰ κάλλιστα προσδοκῶντες παρὰ Μιθριδάτου... ἀπήντησαν δ᾿ αὐτῷ καὶ οἱ περὶ τὸν ∆ιόνυσον τεχνῖται, τὸν ἄγγελον τοῦ νέου ∆ιονύσου, καλοῦντες ἐπὶ τὴν κοινὴν ἑστίαν καὶ τὰς περὶ ταύτην εὐχάς τε καὶ σπονδάς. ... ἀφ᾿ ἧς (sc. οἰκίας) ἐξῄει χλαµύδα λαµπρὰν ἐπισύρων καὶ περικείµενος δακτύλιον χρυσίου ἐγγεγλυµµένην ἔχοντα τὴν Μιθριδάτου εἰκόνα ... ἐν δὲ τῷ τεµένει τῶν τεχνιτῶν θυσίαι τε ἐπετελοῦντο ἐπὶ τῇ Ἀθηνίωνος παρουσίᾳ καὶ µετὰ κήρυκος προαναφωνήσεως σπονδαί. ... ἀναβὰς οὖν ἐπὶ τὸ βῆµα... στὰς ἐπὶ τούτου καὶ περιβλέψας

64

Angelos Chaniotis

passage begins with a description of Athenion’s arrival in Athens. With the selection of the appropriate vocabulary and images, Poseidonios constructs for his account a framework full of theatrical connotations. This medium allows Poseidonios to underline the theatricality of this scene, to characterise Athenion’s speech as a spectacle and a performance in acting, through which Athenion controlled the emotions of the Athenian assembly. On a deeper level, this description gives Poseidonios the possibility to point to the paradoxical features of this event and thus arouse the appropriate feelings among his readers. For his return they dispatched warships and a chair with silver legs. He had barely arrived, and the largest part of the population was already out in the streets to receive him. Many other spectators (θεαταί) also came together, wondering at the strange reversal of fortune, that Athenion, who had acquired citizenship with fraud, was now being brought to Athens on a chair with silver legs and on a purple mattress. ... So men, women, and children were running to this spectacle (θέα), expecting the best from Mithridates.

Athenion’s success is presented as a game of Fortune (τὸ παράδοξον τῆς τύχης), and more paradoxa are to come. With the words theatai and thea, Poseidonios assimilates Athenion’s arrival with a spectacle; and in a subtle way he incorporates Athenion into the world of the theatre. Those who first receive him are the theatre artists; it is from the seat of their association that Athenion enters the agora. The Dionysiac artists received him, the messenger of the new Dionysos, inviting him to the common hearth and the prayers and libations on it. ... In the precinct of the artists there were sacrifices for the presence of Athenion and libations announced by the herald.

Athenion leaves the seat of the theatre artists in a costume, described in great detail, in order to make him appear as the caricature of a king. He came out from this house dragging a glamorous cloak and wearing a golden ring with the portrait of Mithridates engraved on it. Many servants walked before and behind him.

The ring on his finger was so huge that one could recognize Mithridates’ portrait on it – probably because Athenion was stretching his hand in such a way that everyone could see it. He was not wearing, but dragging his cloak (ἐπισύρων), which was obviously too large for him and not properly worn. Among the many details Poseidonios could have chosen to include in his narrative, he selected those that equated Athenion with an actor and his behaviour with a performance. Athenion is not only the caricature of a king; he is the caricature of an actor. This is obvious also in the next scene. Arriving in the agora, he ascends the tribune and delivers his speech. Here, the historian focuses on the description of body language, ges-

κυκληδὸν τὸ πλῆθος, ἔπειτ᾿ ἀναβλέψας «ἄνδρες Ἀθηναῖοι» ἔφη «τὰ πράγµατα µέν µε βιάζεται καὶ τὸ τῆς πατρίδος συµφέρον ἀπαγγέλλειν ἃ οἶδα, τὸ δὲ µέγεθος τῶν µελλόντων λέγεσθαι διὰ τὸ παράδοξον τῆς περιστάσεως ἐµποδίζει µε». ἁθρόως δ᾿ ἐπιβοησάντων αὐτῷ τῶν περιεστώτων θαρρεῖν καὶ λέγειν, «λέγω τοίνυν» ἔφη «τὰ µηδέποτε ἐλπισθέντα...»· µικρὸν δ᾿ ἐπισχὼν ἐπὶ τούτοις καὶ ἐάσας τοὺς πολλοὺς συλλαλῆσαι περὶ τῶν παραδόξως προηγγελµένων τρίψας τε τὸ µέτωπον «τί οὖν» εἶπε «συµβουλεύω;».

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

65

tures, pauses, the artful exploitation by Athenion of the paradoxon, and his interaction with the audience. He ascended the podium; ... he stood on it looking around at the crowd, and then raising his head he said, ‘Athenians, the affairs and the interests of the city urge me to say what I know, but the magnitude of the things that are to be said prevent me from saying them, because of the unexpected nature of the events.’ And when all who stood around urged him by shouting to have courage and talk, he said, ‘I will tell you, what you had never hoped for.’ Then he paused for a while after this, giving the crowd the opportunity to talk about the unexpected announcements. Then he scratched his forehead and said, ‘What is then the advice that I give you’?

There is hardly any detail in Poseidonios’ description that does not find a parallel either in contemporary acting or in the instructions given to orators by the authors of rhetorical treatises. For the pathetic tone of amplification, the author of Rhetorica ad Herrenium recommends that one ought to slap one’s thigh or beat one’s head, and sometimes use a calm and uniform gesticulation and a sad and disturbed expression. 30 Athenion’s body language – looking around, raising the head, scratching the forehead – corresponds to the body language of the actors, the comedians in particular, as we know it from terracotta statuettes,31 but also as it can be inferred from a passage in Plautus’ Miles Gloriosus, in which Periplectomenus describes the gestures of Palaestrio.32 Just look at him, how he stands there with bent brow, considering and cogitating. He is tapping his chest with his fingers. He intends to summon forth his intelligence, I suppose. Aha! He turns away! He rests his left hand on his left thigh, and reckons on the fingers of the right hand. He gives his right thigh a smack! ... He snaps his fingers! He is in distress, constantly changing his position! Look there, though; he is shaking his head – that idea won’t do! ... He is supporting his chin with a pillar. ... Glorious! A graceful pose, indeed! Just like the slaves in comedies!

Two features of Athenion’s performance – body language and pause – are mentioned by Poseidonios, in another passage, among the skills of the buffoon Saunio: ἐπιστροφὴ σώµατος and σιωπή.33 All this is also recommended by the authors of rhetorical treatises. According to the author of Rhetorica ad Herennium, pauses keep the voice strong and give the audience the chance to reflect on the speech. They also render the thoughts more clear-cut by separating them and leave the hearer time to think.34 At the beginning of his speech, Athenion looks around. As Quintilian explains, the teacher of the orator will insist that the speaker faces his audience, that the lips are not distorted nor the jaws parted to a grin, that the face is not thrown back, nor the eyes fixed on the ground, nor the neck slanted to left or right. ... I have seen many, who raised their brows whenever the voice was called upon for an effort, others who wore a perpetual frown, and yet 30 31 32 33 34

Rhetorica ad Herennium 3.15.26f. E.g., Bieber 1961, 41 figs. 160f., 47 figs. 197f., 80 fig. 294, 81 fig. 297, 100 fig. 372. Plautus, Miles Gloriosus 200–215. Poseidonios, Histories fr. 221 ed. Theiler = Diodoros 37.12.2: κατὰ τὴν σιωπὴν καὶ καθ᾿ ὁποίαν σώµατος ἐπιστροφὴν ἅπαντας ἐποίει τοὺς θεωµένους µειδιᾶν. Rhetorica ad Herennium 3.12.22.

66

Angelos Chaniotis others who could not keep their eyebrows level, but raised one towards the top of the head 35 and depressed the other until it almost closed the eye.

The similarity of Athenion’s theatrical behaviour with descriptions of orations in Hellenistic historiography, recommendations in handbooks of oratory, and above all with summaries of assembly speeches in Hellenistic decrees supports the conclusion that Poseidonios’ narrative was inspired by contemporary practices in the assembly, if not by a particular oration.36 Poseidonios’ narrative presents several similarities with the narrative of the battle at Pellene: in both episodes a historical event is rendered as a drama staged by Fortune. In both accounts there is an emphasis on dress in connection with illusion: in Pellene, a helmet creates the illusion of divine epiphany; in Athens, Athenion’s whole attire aims to create the illusion of dignity and cover Athenion’s origins and incompetence. In Poseidonios we have an additional element: the hypokrisis, a rhetorical delivery with clear theatrical overtones aiming to capture the audience’s attention and to arouse emotions. For the historian, such theatrical behavior is an object of observation that also serves as a medium of emotional arousal. Through theatrical behaviour – use of costume, voice, body-language, facial expressions, etc. – individuals construct an image of themselves which is at least in part deceiving, because it either is in contrast to reality or because it exaggerates or partly distorts reality. The Hellenistic statesmen who used theatrical behaviour aimed to gain control over the emotions and the thoughts of their audiences, to arouse respect, hope, fear, gratitude, trust and so on. The Hellenistic historians, who attract our attention, directly or indirectly, to this behavior, also aim at their readers’ (hostile) emotions: indignation, contempt, resentment, Schadenfreude. Neither theatrical behaviour nor illusion is a Hellenistic invention. Hardly any society can function without elements of staged behaviour or illusion. Artificial, pretentious, and staged behaviour or the display of exaggerated or false emotions is an ubiquitous phenomenon; and we are surrounded by illusions, whether we recognise them or not; some of them are harmless – for instance, creating the illusion of a successful performance by applauding after a bad lecture –, others have more serious consequences – for instance, the illusion of everlasting love or the illusion that parliamentary democracy guarantees the full and unlimited expression of the will of the people. The Hellenistic historians did not invent illusion and theatricality but they carefully observed it and they draw our attention to it in subtle ways. Polybios’ narrative of how Agathokles, Ptolemy IV’s chief minister, announced the king’s death to the army (203 BCE) is a case in point. Agathokles summoned a meeting of the Macedonians and appeared there together with his sister Agathokleia, the king’s mistress, and the young Ptolemy V, the king’s son.37 35 36 37

Quintilian, Institutio oratoria 1.11.10–14. Chaniotis 2013a, with discussion of some epigraphic evidence: cf. Chaniotis 2009a, 78–96. Polybios 15.26.1–3: καὶ τὰς µὲν ἀρχὰς ὑπεκρίνετο τὸν οὐ δυνάµενον εἰπεῖν ἃ βούλεται διὰ τὸ πλῆθος τῶν ἐπιφεροµένων δακρύων· ἐπεὶ δὲ πλεονάκις ἀποµάττων τῇ χλαµύδι

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

67

At first he pretended (ὑπεκρίνετο) that he could not say what he wished owing to the abundance of the tears that choked him, but after wiping his eyes many times with his cloak and subduing the outburst, he took the child in his arms and said, ‘Take this child whom his father on his death-bed placed in the arms of this woman’, pointing to his sister, ‘and placed in your trust, men of Macedon’.

Agathokles’ appearance resembles a dramatic performance, engaging careful use of the voice and the dress, facial expressions, gestures, and movements. The description of how Agathokles wipes his eyes with his cloak deserves our attention. In contemporary perception, the use of the garment has an intrinsic connection with acting and theatrical behaviour.38 The way Agathokles uses his cloak recalls a statuette of a comic actor in the role of a slave, wiping his eyes with his cloak.39 The theatrical use of the garment is also noted by Theophrastos, when he describes how the flatterer stuffs the corner of his cloak in his mouth as if he could not hold his merriment.40 The treatment of the garment by Hellenistic royalty is very often mentioned by Polybios, e.g. in connection with Philip V.41 After he celebrated the contest of the Nemea, he returned to Argos, taking off the diadem and the purple garment, wishing to create the impression of equality with the many, of a mild person, of a friend of the people. But the more popular the dress he wore, the more monarchical the power he possessed.

The dress and the removal of royal insignia were for Philip what the mask is for an actor: media for the creation of an illusion, in this case the illusion of the friendly ruler. Polybios, again, observed the same behaviour in the case of Antiochos IV.42 Many times he used to take off the royal garment and to wear of toga, going around in the agora, participating in the elections and asking the people for their vote, embracing some and begging others, in order to be elected as aedilis or tribunus.

This behaviour was intended to be taken as an expression of affability and respect of popular rule. Long before Polybios, another Hellenistic historian, Douris, had called attention to the theatrical behaviour of statesmen and kings and its intended

38 39 40 41

42

κατεκράτησε τῆς ἐπιφορᾶς, βαστάσας τὸ παιδίον «λάβετε» ἔφη «τοῦτον, ὃν ὁ πατὴρ ἀποθνήσκων εἰς µὲν τὰς ἀγκάλας ἔδωκε ταύτῃ» δείξας τὴν ἀδελφὴν «παρακατέθετο δ᾿ εἰς τὴν ὑµετέραν, ὦ ἄνδρες Μακεδόνες, πίστιν». Examples in Chaniotis 2009a, 74–77, 80f., and 110. Bieber 1961, 46 fig. 188 (Metropolitan Museum, New York, inv.no. 13.225.13). Theophrastos, Characters 2.4: καὶ σκώψαντι ψυχρῶς ἐπιγελάσαι τό τε ἱµάτιον ὦσαι εἰς τὸ στόµα ὡς δὴ οὐ δυνάµενος κατασχεῖν τὸν γέλωτα. Polybios 10.26.1f.: ὅτι Φίλιππος ὁ βασιλεὺς Μακεδόνων µετὰ τὸ ἐκτελέσαι τὸν τῶν Νεµέων ἀγῶνα αὖθις εἰς Ἄργος ἐπανῆλθε καὶ τὸ µὲν διάδηµα καὶ τὴν πορφύραν ἀπέθετο, βουλόµενος αὑτὸν ἴσον τοῖς πολλοῖς καὶ πρᾷόν τινα καὶ δηµοτικὸν ὑπογράφειν. ὅσῳ δὲ τὴν ἐσθῆτα δηµοτικωτέραν περιετίθετο, τοσούτῳ τὴν ἐξουσίαν ἐλάµβανε µείζω καὶ µοναρχικωτέραν. Polybios 26.1.4 (= Athenaios X 439 a): πολλάκις δὲ καὶ τὴν βασιλικὴν ἀποθέµενος ἐσθῆτα τήβενναν ἀναλαβὼν περιῄει κατὰ τὴν ἀγορὰν ἀρχαιρεσιάζων καὶ τοὺς µὲν δεξιούµενος, τοὺς δὲ καὶ περιπτύσσων παρεκάλει φέρειν αὑτῷ τὴν ψῆφον, ποτὲ µὲν ὡς ἀγορανόµος γένηται, ποτὲ δὲ καὶ ὡς δήµαρχος.

68

Angelos Chaniotis

emotional impact, focusing on the use of make-up and dress. In one of his fragments he comments on the efforts of Demetrios of Phaleron to appear merry and pleasant:43 He paid attention to his appearance, dying his hair blond and covering his face with rouge and with other creams; for he wished to appear pleasant and friendly to those who encountered him.

For Demetrios of Phaleron a merry face was not an expression of character but a mask, with which he demonstrated affability. In several of his fragments, Douris shows a vivid interest in garments (e.g. in the robe prepared for Demetrios Poliorketes), and it quite probable that Plutarch’s famous description of Demetrios Poliorketes after a lost battle, which inspired Cavafy’s King Demetrius, is taken from Douris.44 When Demetrios realized that his case is lost he went to his tent, and, as if he had been an actor and not a real king, put on a dark cloak in place of his stage-robes of royalty, and stole away unnoticed.

Phylarchos described a similar scene in connection with Mysta, the mistress of Seleukos II:45 When Seleucus was defeated by the Galatians, and was with difficulty able to save himself by flight, she put off the robes of a queen which she had been accustomed to wear, and assumed the garment of an ordinary servant, and being taken prisoner, was carried away with the rest of the captives.

Douris was also the main source of Plutarch for his life of Demetrios Poliorketes, the most theatrical of kings. I cannot discuss here how Douris contrasted the two Demetrioi (Demetrios of Phaleron and Demetrios Poliorketes) or how Plutarch presented the life of Demetrios Poliorketes as a drama,46 but one passage in Plutarch is relevant, since it shows the impact of theatricality on emotion. It concerns the staged appearance of Demetrios Poliorketes in the theatre of Dionysos in Athens.47 In the spring of 295 BCE Demetrios captured Athens and ordered the Athe43

44

45

46

47

Douris, FgrHist 76 F 10: ἐπιµελεῖτο δὲ καὶ τῆς ὄψεως, τήν τε τρίχα τὴν ἐπὶ τῆς κεφαλῆς ξανθιζόµενος καὶ παιδέρωτι τὸ πρόσωπον ὑπαλειφόµενος καὶ τοῖς ἄλλοις ἀλείµµασιν ἐγχρίων ἑαυτόν· ἠβούλετο γὰρ τὴν ὄψιν ἱλαρὸς καὶ τοῖς ἀπαντῶσιν ἡδὺς φαίνεσθαι. Plutarch, Demetrios 44: καὶ παρελθὼν ἐπὶ σκηνήν, ὥσπερ οὐ βασιλεύς, ἀλλ᾿ ὑποκριτής, µεταµφιέννυται χλαµύδα φαιὰν ἀντὶ τῆς τραγικῆς ἐκείνης, καὶ διαλαθὼν ὑπεχώρησεν. Cf. Douris’ interest in garments: FgrHist 76 F 12, 14, 50, 60. Phylarchos, FgrHist 81 F 30: ὑπὸ Γαλατῶν Σελεύκου νικηθέντος καὶ µόλις ἐκ τῆς φυγῆς διασωθέντος αὐτὴ [Μύστα] µεταµφιεσαµένη τὴν βασιλικὴν ἐσθῆτα καὶ ῥάκια λαβοῦσα θεραπαινίδος τῆς τυχούσης συλληφθεῖσα ἀπήχθη µετὰ τῶν ἄλλων αἰχµαλώτων. For theatricality in the life of Demetrios see Plutarch, Demetrios 18, 28, 41, 44, and 53. Discussions: Mastrocinque 1979; Chaniotis 1997, 244f. and 2009a, 111–128; Thonemmann 2005. On the contrast between Demetrios of Phaleron and Demetrios Poliorketes in Douris, see Chaniotis 2011, 177f. and 187. Plutarch, Demetrios 34 (after Douris?): εἰσελθὼν ὁ ∆ηµήτριος καὶ κελεύσας εἰς τὸ θέατρον ἀθροισθῆναι πάντας, ὅπλοις µὲν συνέφραξε τὴν σκηνὴν καὶ δορυφόροις τὸ λογεῖον περιέλαβεν, αὐτὸς δὲ καταβάς, ὥσπερ οἱ τραγῳδοί, διὰ τῶν ἄνω παρόδων, ἔτι µᾶλλον ἐκπεπληγµένων τῶν Ἀθηναίων, τὴν ἀρχὴν τοῦ λόγου πέρας ἐποιήσατο τοῦ δέους αὐτῶν.

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

69

nians to assemble in the theatre. He fenced the stage-building round with armed men, and encompassed the stage itself with his bodyguards. After these arrangements, and while the Athenians observed all this full of fright (ἐκπεπληγµένων), Demetrios finally appeared through one of the upper side-entrances like a tragic actor. The Athenians were now frightened more than ever (δέους), and Demetrios was in full control of their emotions. Soon their fears proved unjustified. With the choice of the right tone of voice (τόνου φωνῆς) and with the selection of the proper words (‘script’) Demetrios won from the Athenians the result aimed by his performance. We notice here, again, the importance of peripeteia, the sudden change. In this passage, we may distinguish three different layers of a historian’s treatment of emotions: the historian describes the emotions of the subjects of his narrative; he describes the theatrical means by which his protagonist controls the emotions of his audience; and through his vivid narrative he projects the change of emotions – the passage from agony and fear to relief – onto his readers (see also below pp. 71 and 73). This episode is described as taking place in a theatre. The theatre was indeed one of the stages of public life in the Hellenistic cities. But what is significant is the fact that Hellenistic historians often explicitly mention the theatre as the stage of a particular event – although they leave many other details unmentioned. Interestingly, they do so especially when the event that they describe has elements of theatricality. In such narratives the word theatre functions as an acoustic signal that prepares the reader to expect theatrical behaviour and a small drama. A few examples illustrate this point. In 217 BCE Philip V faced a riot of his troops because of dissatisfaction with the distribution of booty.48 Philip came in great haste from Lechaion to Corinth, running; there he assembled the Macedonians in the theatre and addressed them using both exhortation and rebuke for their conduct. Upon this there was a noise and much confusion of counsel, some advising that the offenders should be arrested and called to account, while others were in favour of reconciliation and grant of amnesty to all. Then the king, pretending (ὑποκριθείς) that he was convinced, addressed some words of exhortation to all and took his departure, well knowing who the originators of the sedition had been, but pretending ignorance owing to the pressure of circumstances.

Where the meeting took place does not add anything of substance to the story. The meeting could have taken place in the camp, in a sanctuary, or in a stadium. But through the combination of the words θέατρον and ὑποκρίνοµαι Polybios associates Philip’s actions with the stage of deception and characterises the king as a cunning actor. Polybios several times shows a keen interest in Philip’s theat-

48

καὶ γὰρ τόνου φωνῆς καὶ ῥηµάτων πικρίας φεισάµενος, ἐλαφρῶς δὲ καὶ φιλικῶς µεµψάµενος αὐτὸς διηλλάσσετο. Polybios 5.25.4f.: ὁ Φίλιππος ἧκε µετὰ σπουδῆς ἐκ τοῦ Λεχαίου θέων εἰς τὴν πόλιν. καὶ συναγαγὼν εἰς τὸ θέατρον τοὺς Μακεδόνας, τὰ µὲν παρεκάλει, τὰ δ᾿ ἐπέπληττε πᾶσιν ἐπὶ τοῖς πεπραγµένοις. θορύβου δ᾿ ὄντος καὶ πολλῆς ἀκρισίας, καὶ τῶν µὲν οἰοµένων δεῖν ἄγειν καὶ καταλύειν τοὺς αἰτίους, τῶν δὲ διαλύεσθαι καὶ µηδενὶ µνησικακεῖν, τότε µὲν ὑποκριθεὶς ὡς πεπεισµένος καὶ παρακαλέσας πάντας ἐπανῆλθε, σαφῶς µὲν εἰδὼς τοὺς ἀρχηγοὺς τῆς κινήσεως γεγονότας, οὐ προσποιηθεὶς δὲ διὰ τὸν καιρόν.

70

Angelos Chaniotis

rical behaviour, and in a passage which inspired Frank Walbank’s seminal article on tragic historiography,49 he explicitly uses a theatrical vocabulary to describe a change of fortune in Philip’s life, when he was forced to execute his son in 180 BCE:50 And the third tragedy (δρᾶµα) which Fortune produced at the same time was that concerning his sons. ... The quarrel of his sons burst into flame at the same time, as if Fortune was bringing their misfortunes on the stage (ἀναβιβαζούσης ἐπὶ σκηνήν) at one and the same time on purpose.

Mentioning the theatre as the backdrop of an event was a historian’s implicit comment on the use of theatricality, illusion, and deception to achieve arousal of emotions. Another example, again in Polybios, is the description of an assembly of the Achaean League in 168 BCE, to which Polybios was an eyewitness. 51 The Achaean politicians Andronidas and Kallikrates were supporting reconciliation between Ptolemy VIII and Antiochos IV. When they noticed that their arguments had no impact upon the assembly they applied a trick (µηχανή). They staged the sudden appearance of a messenger bearing a letter of the Roman consul urging the Achaeans to follow the Roman policy and make peace between the kings. Polybios’ choice of words is significant: the messenger, who arrived almost like a deus ex machina, is not described as arriving at the assembly (εἰς τὴν ἐκκλησίαν) but at the theatre (εἰς τὸ θέατρον). Thus the historian associated the scene with the sudden arrival of a messenger in Classical drama and its effect on the audience. The next example is from Poseidonios. The story goes that Nikias, one of the leading men at Engyon in Sicily, was urging his fellow citizens to go over to the Romans during the Second Punic War (c. 212 BCE). Knowing that his enemies were planning to arrest him and deliver him to the Carthaginians, he gave a performance during the assembly. Poseidonios twice mentions that the assembly took place in the theatre.52 Nikias suddenly (ἐξαίφνης) threw himself upon the ground, and waiting for a while, when silence and fright (ἡσυχίας σὺν ἐκπλήξει) naturally prevailed, he lifted his head, turned it about, and spoke in a low and trembling voice, little by little raising and sharpening its tone.

49 50

51

52

Walbank 1938. Polybios 23.10.12 and 16: τρίτον δ᾿ ἡ τύχη δρᾶµα κατὰ τὸν αὐτὸν καιρὸν ἐπεισήγαγεν τὸ κατὰ τοὺς υἱούς. ... καὶ τὸ κατὰ τοὺς υἱοὺς νεῖκος ἅµα τοῖς προειρηµένοις ἐξεκαύθη, τῆς τύχης ὥσπερ ἐπίτηδες ἀναβιβαζούσης ἐπὶ σκηνὴν ἐν ἑνὶ καιρῷ τὰς τούτων συµφοράς. Polybios 29.25: οὐδενὸς δὲ προσέχοντος αὐτοῖς ἐπεισήγαγον µηχανήν. Παρῆν γὰρ ἐκ πορείας εἰς τὸ θέατρον γραµµατηφόρος φέρων ἐπιστολὴν παρὰ Κοΐντου Μαρκίου, δι᾿ ἧς παρεκάλει τοὺς Ἀχαιοὺς ἀκολουθοῦντας τῇ Ῥωµαίων προαιρέσει πειρᾶσθαι διαλύειν τοὺς βασιλεῖς. Poseidonios, hist. fr. 93 a ed. Theiler (Plutarch, Marcellus 20): ... ἐξαίφνης ἀφῆκεν εἰς τὴν γῆν τὸ σῶµα, καὶ µικρὸν διαλιπών, οἷον εἰκός ἡσυχίας σὺν ἐκπλήξει γενοµένης, τὴν κεφαλὴν ἐπάρας καὶ περιενεγκών ὑποτρόµῳ φωνῇ καὶ βαρείᾳ, κατὰ µικρὸν συντείνων καὶ παροξύνων τὸν ἦχον, ὡς ἑώρα φρίκῃ καὶ σιωπῇ κατεχόµενον τὸ θέατρον, ἀπορρίψας τὸ ἱµάτιον καὶ περιρρηξάµενος τὸν χιτωνίσκον, ἡµίγυµνος ἀναπηδήσας ἔθεε πρὸς τὴν ἔξοδον τοῦ θεάτρου, βοῶν ὑπὸ τῶν Ματέρων ἐλαύνεσθαι. On the significance of φρίκη see the discussion by Dougals Cairns in this volume (pp. 85–107, esp. p. 97).

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

71

And when he saw the whole audience (θέατρον) struck dumb with horror (φρίκῃ καὶ σιωπῇ κατεχόµενον), he threw his mantle, tore off his tunic, and leaping up half naked, ran towards the exit of the theatre, crying out that he was being pursued by the Mother Goddesses.

Here we have a genuine acting performance in the theatre, with careful use of the voice, body language, and dress, and with a clear aim: to create the illusion of a divine epiphany und thus arouse the emotion of fear in the audience. Finally, a passage in Diodoros, most probably deriving from an earlier source, skilfully recreates for his readers the feeling of dread and amazement felt by the Rhodians, when the fleet of Demetrios Poliorketes approached their city for the most famous siege of the Hellenistic period:53 The soldiers of the Rhodians occupied the walls awaiting the approach of the enemy fleet, while the old men and the women watched from their houses, as the city is built like a theatre (θεατροειδοῦς); and all of them, terrified (καταπληττόµενοι) at the size of the fleet and at the bright light reflected by the shining weapons, were in great agony (οὐ µετρίως ἠγωνίων).

In this passage, the besieged Rhodians take their place in their city like the audience in a theatre, terrified by what they are watching. They were at the same time the spectators of their own war and a spectacle for the historian’s readers. The way the historian describes this scene is the result of careful consideration and skilful composition. He did not invent the similarity between Rhodes and a theatre54 but he chose to highlight it. Thus he transformed the battlefield into the stage of a drama and prepared his readers in a subtle manner for the dramatic scenes and continuous changes of fortune that were to follow.55 Already at the outset of the narrative the Hellenistic historian impresses the reader with a detailed description of Demetrius’ forces (20.82.4f.). The expectation of the assailants is contrasted with the fear of the defenders:56 As the land of the Rhodians had not been sacked for many years, a large number of those who were accustomed to make their profit from the misfortune of those defeated in war came together. ... The whole space between the island and the opposite shore was seen to be filled with his ships, which brought great fear and panic to those who were watching from the city.

In his account, Diodoros highlights unexpected turns of fortune caused by storms, Rhodian stratagems, and the heroism of the besieged, and describes how the emo-

53

54 55 56

Diodoros 20.83.2: οἱ µὲν γὰρ στρατιῶται τῶν Ῥοδίων διειληφότες τὰ τείχη τὸν ἐπίπλουν ἐκαραδόκουν τῶν πολεµίων, πρεσβῦται δὲ καὶ γυναῖκες ἀπὸ τῶν οἰκιῶν ἀφεώρων, οὔσης τῆς πόλεως θεατροειδοῦς, πάντες δὲ τό τε µέγεθος τοῦ στόλου καὶ τὴν αὐγὴν τῶν ἀποστιλβόντων ὅπλων καταπληττόµενοι περὶ τῶν ὅλων οὐ µετρίως ἠγωνίων. On the ‘theatre-like’ appearance of late Classical and Hellenistic cities see Caliò 2005; on Rhodes, ibid. 91–101, 109–111. Discussion of the dramatic aspects and the role of emotions in Diodoros’ description of the siege of Rhodes in Chaniotis 2013d, 451–454. Diodoros 20.82.5 and 20.83.1: πολλὰ γὰρ ἔτη τῆς χώρας τῆς Ῥοδίων ἀπορθήτου γεγενηµένης συνέρρει πανταχόθεν πλῆθος τῶν εἰωθότων ὠφελείας ἰδίας ἡγεῖσθαι τὰ τῶν πολεµουµένων ἀτυχήµατα. ... ὥστε πάντα τὸν ἀνὰ µέσον τόπον τῆς τε νήσου καὶ τῆς ἀντικειµένης παραλίας συµπεπληρωµένον φαίνεσθαι τοῖς πλοίοις καὶ πολὺν φόβον καὶ κατάπληξιν παρέχεσθαι τοῖς ἀπὸ τῆς πόλεως θεωροῦσιν.

72

Angelos Chaniotis

tions of the besieged Rhodians continually alternated between terror and hope.57 The most dramatic of these accounts is a description of a battle in the theatre of Rhodes:58 When day came and Demetrios raised the ensign, the men who were attacking the harbour and those who had been placed all around the wall shouted the battle cry, giving courage to the men who had occupied part of the area near the theatre. In the city, the crowd of children and women were terrified and in tears, under the impression that the city was being taken by storm. ... At first neither side withdrew from its position; but afterwards the number of the Rhodians constantly increased and they were eagerly facing the danger, like men fighting for their fatherland and the most valuable things. As the king’s soldiers were in distress, their commanders, Alkimos and Mantias, received many wounds and fell; most of the others were killed, some were captured, and only a few escaped to the king and were saved. ... Demetrios thought that Fortune had stolen from his hands the capture of the city ...

Demetrios’ campaign is likened to a spectacle also by Plutarch’s Hellenistic source:59 His enemies stood on shore and admired his ships of sixteen and fifteen rows of oars, as they sailed past their land; and his helepoleis [siege engines] were a spectacle to the besieged. ... (The helepolis) astounded the mind and delighted the eyes of those who watched.

My last example of this device of Hellenistic historians – to use theatres as the backdrop of dramatic narratives and descriptions of emotional scenes – comes from Plutarch’s life of Phokion. This passage describes how Phokion was brought to trial as an opponent of king Philip Arridaios in 318 BCE. Plutarch’s Hellenistic source (certainly an Athenian historian) describes the scene:60 57

58

59

60

Storms: 20.86.1; 96.2. Rhodian stratagems: 20.85.4; 86.3; 87.2; 88.3–6; 93.1; 94.2; 97.4. Battle scenes: 20.87.3; 88.8; 96.4–7; 98.4–9. Emotions: fear: 20.88.3; 92.1; 98.7f.; hope: 20.96.3. Diodoros 20.98.8–99.1: περικαταλαβούσης δ᾽ ἡµέρας καὶ τοῦ ∆ηµητρίου τὸ σύσσηµον ἄραντος οἱ µὲν τῷ λιµένι προσβαλόντες καὶ τὸ τεῖχος πάντοθεν περιεστρατοπεδευκότες συνηλάλαξαν, εὐθαρσεῖς ποιοῦντες τοὺς κατειληφότας µέρος τοῦ περὶ τὸ θέατρον τόπου, ὁ δὲ κατὰ τὴν πόλιν ὄχλος παίδων καὶ γυναικῶν ἐν φόβοις ἦν καὶ δάκρυσιν, ὡς τῆς πατρίδος κατὰ κράτος ἁλισκοµένης. ... τὸ µὲν πρῶτον οὐδέτεροι τῆς ἰδίας τάξεως ἐξεχώρουν, µετὰ δὲ ταῦτα τῶν µὲν Ῥοδίων ἀεὶ πλειόνων γινοµένων καὶ τὸν κίνδυνον ἑτοίµως ὑποµενόντων, ὡς ἂν ὑπὲρ πατρίδος καὶ τῶν µεγίστων ἀγωνιζοµένων, τῶν δὲ τοῦ βασιλέως θλιβοµένων Ἄλκιµος µὲν καὶ Μαντίας οἱ τὴν ἡγεµονίαν ἔχοντες πολλοῖς περιπεσόντες τραύµασιν ἐτελεύτησαν, τῶν δ᾽ ἄλλων οἱ πλεῖστοι οἱ µὲν ἐν χειρῶν νόµῳ διεφθάρησαν, οἱ δ᾽ ἥλωσαν, ὀλίγοι δὲ πρὸς τὸν βασιλέα φυγόντες διεσώθησαν. ... ∆ηµήτριος δὲ τὴν τῆς πόλεως ἅλωσιν ὑπολαβὼν ἐκ τῶν χειρῶν αὐτοῦ τὴν τύχην ἀφῃρῆσθαι. Plutarch, Demetrios 20.4 and 21.2: καὶ τὰς µὲν ἑκκαιδεκήρεις αὐτοῦ καὶ τὰς πεντεκαιδεκήρεις ἐθαύµαζον ἑστῶτες οἱ πολέµιοι παρὰ τὴν γῆν αὐτῶν πλεούσας, αἱ δ᾽ ἑλεπόλεις ὡς θέαµα τοῖς πολιορκουµένοις ἦσαν. ... θάµβος ἅµα τῇ ψυχῇ καὶ χάριν τινα τῇ ὄψει τῶν θεωµένων παρεῖχε [sc. the ἑλέπολις]. Plutarch, Phokion 34: καὶ προσῆν τὸ σχῆµα τῇ κοµιδῆ λυπηρόν, ἐφ᾿ ἁµάξαις κοµιζοµένων αὐτῶν διὰ τοῦ Κεραµεικοῦ πρὸς τὸ θέατρον· ἐκεῖ γὰρ αὐτοὺς προσαγαγὼν ὁ Κλεῖτος συνεῖχεν, ἄχρι οὗ τὴν ἐκκλησίαν ἐπλήρωσαν οἱ ἄρχοντες, οὐ δοῦλον, οὐ ξένον, οὐκ ἄτιµον ἀποκρίναντες, ἀλλὰ πᾶσι καὶ πάσαις ἀναπεπταµένον τὸ βῆµα καὶ τὸ θέατρον παρασχόντες. ἐπεὶ δὲ ἥ τε ἐπιστολὴ τοῦ βασιλέως ἀνεγνώσθη, λέγοντος αὐτῷ µὲν ἐγνῶσθαι προδότας γεγονέναι τοὺς ἄνδρας, ἐκείνοις δὲ διδόναι τὴν κρίσιν ἐλευθέροις τε

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

73

Their transportation [of Phokion and his associates] presented a sad spectacle, as they were brought from Kerameikos to the theatre (πρὸς τὸ θέατρον). After they had been brought there Kleitos put them under arrest, until the magistrates called the assembly, allowing everyone, man and woman alike, access to the podium and the theatre (τὸ θέατρον παρασχόντες), not preventing anyone from attendance, neither slaves nor foreigners nor citizens who had lost their citizenship. Then the letter of the king was read out, in which he said that as for his part, he regarded these men as traitors, but since the Athenians are free and autonomous, they have the right to pass their own judgement. After that, Kleitos presented the men. Then some covered their heads and others looked down shedding tears. Someone found the courage to stand up and say that since the king entrusted such an important decision to the people, it was proper that the slaves and the foreigners leave.

As in the description of Demetrios’ performance in the theatre of Athens, we distinguish here the same three layers of a historian’s treatment of emotions: he describes the emotions of the Athenians; he describes the theatrical setting which arouses emotions among the contemporary viewers; and through his vivid narrative he arouses emotions in his readers. The historian evokes in a subtle way the image of a spectacle. He mentions a detail that at first sight seems insignificant: ‘they were brought from Kerameikos to the theatre’. For his readers, who knew the topography of Athens, the significance was clear. If the prisoners were brought from Dipylon, the northwest gate of Athens, at Kerameikos to the theatre of Dionysos, on the south slope of the Acropolis, they can only have followed the processional road of the Great Panathenaia. The historian mentions that this assembly took place in the theatre (not in the Pnyx). But what an assembly! It did not consist of the citizens, but instead of the usual audience of the theatre: men and women, citizens and foreigners, free and slaves. This is not a people’s assembly but a parody of an assembly, a spectacle. And then the letter of the king is read, in which he recognises the right of the Athenians to pass their free judgement, but only after he informed them about his own judgment. In the venue of spectacle, illusion, deception, and emotion, emotions are displayed, the illusion of freedom is maintained, and the assembly becomes a theatrical mask that conceals the bitter reality of asymmetrical relations. 5 WITH TEARS IN THEIR EYES: CREATING EMPATHY IN HELLENISTIC AUDIENCES The historian Phylarchos is better known through Polybios’ criticism than through the surviving passages of his work. In a famous passage, Polybios describes how Phylarchos achieved his aim to arouse empathy among his readers:61

61

δὴ καὶ αὐτονόµοις οὖσι, καὶ τοὺς ἄνδρας ὁ Κλεῖτος εἰσήγαγεν, οἱ µὲν ἐνεκαλύψαντο καὶ κάτω κύψαντες ἐδάκρυον, εἷς δὲ ἀναστὰς ἐτόλµησεν εἰπεῖν ὅτι, τηλικαύτην κρίσιν ἐγκεχειρικότος τῷ δήµῳ τοῦ βασιλέως, καλῶς ἔχει τοὺς δούλους καὶ ξένους ἀπελθεῖν ἐκ τῆς ἐκκλησίας. For veiling (ἐνεκαλύψαντο) in connection with weeping see Cairns 2009. Polybios 2.56.6–8: βουλόµενος δὴ διασαφεῖν τὴν ὠµότητα τὴν Ἀντιγόνου καὶ Μακεδόνων, ἅµα δὲ τούτοις τὴν Ἀράτου καὶ τῶν Ἀχαιῶν, φησὶ τοὺς Μαντινέας γενοµένους ὑποχειρίους µεγάλοις περιπεσεῖν ἀτυχήµασι, καὶ τὴν ἀρχαιοτάτην καὶ

74

Angelos Chaniotis In order to illiustrate the cruelty of Antigonos and the Macedonians, as well as that of Aratos and the Achaeans, he asserts that, when the Mantineans fell into their hands, they suffered great misfortunes; and that the most ancient and greatest of all the cities in Arkadia had to struggle against calamities so terrible as to move all Greece to dread and tears. And eager to arouse the pity (ἔλεον) of his readers and to create empathy toward what was being said, he introduces scenes of women clinging to one another, tearing their hair and baring their breasts, and in addition he describes the tears and lamentations of men and women as they are led away in captivity together with their children and aged parents. And he does this again and again throughout his whole history trying each time to bring the terrible scenes in front of his readers’ eyes.

What Polybios is condemning in book 2 he himself is doing in book 4, when he describes the destruction of Lyttos on Crete. We read of an unexpected turn of fortune (παραλόγως) – of a city abandoned by its defenders and utterly destroyed. We see in front of our eyes women and children being lead to captivity and men sheding tears. And as in the case of Mantineia, the narrative ends with a lament for the oldest city of Crete, the fatherland of the bravest men:62 At about that time, while the Lyttians were campaigning with their entire force against the enemy’s territory, the Knossians noticed this and seized Lyttos, which was left without defenders. They sent the children and the women off to Knossos; as for the city, they burned it and razed it to the ground and damaged it in every possible way. And then they returned home. When the Lyttians came back to the city from their campaign and saw, all together, what had happened, their spirit was so severely affected, that not a single one of those present had the heart to enter the native city. They all marched round it, bewailing and lamenting the fate of their fatherland and their own many times, and then they turned their backs and retreated to Lappa. ... Thus having become in one day cityless and strangers instead of citizens they went on fighting against Knossos with the other allies. Thus was Lyttos utterly and unexpectedly made away with, a colony and relative of the Lacedaemonians, the most ancient city in Crete ever, as all acknowledged, the breeding-place of her bravest men.

62

µεγίστην πόλιν τῶν κατὰ τὴν Ἀρκαδίαν τηλικαύταις παλαῖσαι συµφοραῖς ὥστε πάντας εἰς ἐπίστασιν καὶ δάκρυα τοὺς Ἕλληνας ἀγαγεῖν. σπουδάζων δ᾽ εἰς ἔλεον ἐκκαλεῖσθαι τοὺς ἀναγινώσκοντας καὶ συµπαθεῖς ποιεῖν τοῖς λεγοµένοις, εἰσάγει περιπλοκὰς γυναικῶν καὶ κόµας διερριµµένας καὶ µαστῶν ἐκβολάς, πρὸς δὲ τούτοις δάκρυα καὶ θρήνους ἀνδρῶν καὶ γυναικῶν ἀναµὶξ τέκνοις καὶ γονεῦσι γηραιοῖς ἀπαγοµένων. ποιεῖ δὲ τοῦτο παρ᾽ ὅλην τὴν ἱστορίαν, πειρώµενος ἐν ἑκάστοις ἀεὶ πρὸ ὀφθαλµῶν τιθέναι τὰ δεινά. Polybios 4.54.1–6: κατὰ δὲ τοὺς αὐτοὺς καιροὺς Λυττίων ἐξωδευκότων εἰς τὴν πολεµίαν πανδηµεί, συννοήσαντες οἱ Κνώσιοι τὸ γεγονὸς καταλαµβάνονται τὴν Λύττον, ἔρηµον οὖσαν τῶν βοηθησόντων. καὶ τὰ µὲν τέκνα καὶ τὰς γυναῖκας εἰς Κνωσὸν ἀπέπεµψαν, τὴν δὲ πόλιν ἐµπρήσαντες καὶ κατασκάψαντες καὶ λωβησάµενοι κατὰ πάντα τρόπον ἐπανῆλθον. οἱ δὲ Λύττιοι παραγενόµενοι πρὸς τὴν πόλιν ἀπὸ τῆς ἐξοδείας, καὶ συνθεασάµενοι τὸ συµβεβηκός, οὕτως περιπαθεῖς ἐγένοντο ταῖς ψυχαῖς ὥστε µηδ᾽ εἰσελθεῖν µηδένα τολµῆσαι τῶν παρόντων εἰς τὴν πατρίδα. πάντες δὲ περιπορευθέντες αὐτὴν κύκλῳ, καὶ πολλάκις ἀνοιµώξαντες καὶ κατολοφυράµενοι τήν τε τῆς πατρίδος καὶ τὴν αὑτῶν τύχην, αὖθις ἐξ ἀναστροφῆς ἐπανῆλθον εἰς τὴν τῶν Λαππαίων πόλιν. ... οὗτοι µὲν ἀντὶ πολιτῶν ἀπόλιδες ἐν ἡµέρᾳ µιᾷ καὶ ξένοι γεγονότες ἐπολέµουν πρὸς τοὺς Κνωσίους ἅµα τοῖς συµµάχοις. Λύττος δ᾽ ἡ Λακεδαιµονίων µὲν ἄποικος οὖσα καὶ συγγενής, ἀρχαιοτάτη δὲ τῶν κατὰ Κρήτην πόλεων, ἄνδρας δ᾽ ὁµολογουµένως ἀρίστους ἀεὶ τρέφουσα Κρηταιέων, οὕτως ἄρδην καὶ παραλόγως ἀνηρπάσθη.

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

75

What Polybios criticises in Phylarchos and practices himself is what contemporary orators offered their audiences in the assembly; it is also what their audiences expected to experience. Numerous decrees show that emotional display and dramatic narratives dominated public discourse in the assembly of Hellenistic cities.63 A decree of Xanthos provides an interesting insight in the oral performances in the assembly, displaying a striking similarity to Polybios’ narrative on Lyttos.64 Three envoys from Kytenion in Doris had arrived in Xanthos in Lykia in order to request financial aid for the reconstruction of the fortification wall of their city. The Xanthians should help, they argued, because they were connected with them with bonds of kinship; when the ancestors of the Xanthians were in need, an ancestor of the Kytenians, the Heraclid Aletes, saved them and married their king’s daughter. The feeling of gratitude and the affection towards kinsfolk should motivate the Xanthians to act. At the same time the envoys appealed to compassion by describing the recent disasters that had fallen upon them. Their speech does not survive but the response of the Xanthians gives us an impression of its content:65 They demonstrated that the colonists, sent out from our land by Chrysaor, the son of Glaukos, the son of Hippolochos, received protection from Aletes, one of the descendants of Herakles; for Aletes, starting from the land of the Dorians, came to their aid when they were being warred upon. Putting an end to the danger by which they were beset, he married the daughter of Aor [the Sword], the son of Chrysaor [the Golden Sword].

Then the envoys narrated the events which had brought their fatherland to a desperate situation. A summary of their account is preserved in a letter:66 It occurred that in the time when king Antigonos had invaded Phokis parts of the city walls of all the cities had collapsed because of the earthquakes and the younger men had marched to the sanctuary of Apollo in Delphi in order to protect it. When the king arrived in Doris he destroyed the walls of all our cities and burned down our houses.

The narrative of the envoys ended with a dramatic appeal to the distant relatives in Xanthos not to show indifference:67

63 64 65

66

67

For a discussion of some evidence see Chaniotis 2013a, 2013b, and 2103c. SEG XXXVIII 1476. Main commentaries: Bousquet 1988; Curty 1995, 183–191; Jones 1999, 61f., 139–143; Chaniotis 2009a, 249–269; 2013a, 205–207. SEG XXXVIII 1476 A LL. 22–30: ... ἔτι τε παρεδείκνυον τῶν ἀποικισθέντων ἐκ τῆς ἡµετέρας ὑπὸ Χρυσάορος τοῦ Γλαύκου τοῦ Ἱππολόχου πρόνοιαν πεποιηµένον Ἀλήτην, ὄντα τῶν Ἡρακλειδῶν· ὁρµηθέντα γὰρ αὐτὸν ἐκ τῆς ∆ωρίδος βοηθῆσαι πολεµουµένοις καὶ τὸν περιεστηκότα κίνδυνον λύσαντα συνοικῆσαι τὴν Ἄορος τοῦ Χρυσάορος θυγατέρα. SEG XXXVIII 1476 D 93–103: συµβαίνει γὰρ ἁµῶν, καθ᾿ ὃν καιρὸν ὁ βασιλεὺς Ἀντίγονος ἐνέβαλε ἐν τὰν Φωκίδα [228 BCE], τῶν τε τειχέων µέρη τινὰ καταπεπτώκειν ὑπὸ τῶν σεισµῶν πασᾶν τᾶµ πολίων καὶ τοὺς νεωτέρους εἰσβοαθοήκε⟨ι⟩ν ἐν τὸ ἱερὸ[ν] τοῦ Ἀπόλλωνος τοῦ ἐν ∆ελφοῖς· παραγενόµενος δὲ ὁ βασιλεὺς ἐν τὰν ∆ωρίδα τά τε τείχη ἁµῶν κατέσκαψε πασᾶν τᾶµ πολίων καὶ τὰς οἰκίας κατέκαυσε. SEG XXXVIII 1476 A 14–17: παρακαλοῦσιν ἡµᾶς ἀναµνησθέντας τῆς πρὸς αὐτοὺς ὑπαρχούσης συγγενείας ἀπό τε τῶν θεῶν καὶ τῶν ἡρώων µὴ περιιδεῖν κατεσκαµµένα τῆς

76

Angelos Chaniotis They ask us to bring to our memory our kinship to them, which originates in the gods and the heroes, and not to remain indifferent to the fact that the walls of their fatherland have been razed to the ground. ... They requested not to look on the elimination of the largest city among the cities of the Metropolis (the Mother-City) with indifference (περιιδεῖν).

The Xanthian magistrate who formulated the responses had no choice but to display compassion:68 We should respond that all the Xanthians felt the same grief with you (συνηχθέσθησαν) for the misfortunes (ἀκληρήµατα) which have befallen your city.

The response of the Xanthians is the response that a tragedy provokes: ἔλεος and φόβος, fear and compassion for human fate. It was the appropriate response to a narrative that must have highlighted the agony of the population, tragic ironies, and unexpected turns of fortune. The similarity with Polybios is striking: Both Lyttos and Kytenion were attacked while they were defenseless. In the case of Kytenion, the enemy exploited a moment of weakness, with the fortifications destroyed by earthquakes and the young warriors absent in Delphi, to invade Phokis. In the case of Lyttos, all the warriors were absent on a campaign. Both cities were burned. In both cases the narrators highlighted the seniority of Lyttos and Kytenion over the other cities of the region, as did Phylarchos in the case of Mantineia: the more ancient a city, the greater the shock for its utter destruction. Important media for the arousal of empathy are the selection of suitable images and the use of emotional language,69 as well as the delivery (hypokrisis) of the orations. For the latter – the appropriate use of the voice and body-language as we observed it in Athenion’s oration in Athens (pp. 64–66) – the inscription from Xanthos does not provide any information, but it is reported of other contemporary ambassadors that they delivered their speeches with tears in their eyes.70 Nonetheless, even the summary of the envoys’ speech reflects the use of emotional language and allows a comparison with other emotionally loaded accounts both in historiography and in decrees that reflect orations in the assembly. First, we notice the use of the rare word ἀκληρήµατα (misfortunes; literally: unjust deprivation of one’s lot). The word is used in the Xanthian response (τοῖς περὶ τὴν πόλιν γεγενηµένοις ἀκληρήµασιν; ‘for the misfortunes which have befallen your city’), which, however, adopts the vocabulary and the formulations that were used in the debate in the assembly. Ἀκληρήµατα and related words (ἀκληροῦντες, ἀκληρία) appear in orations with emotional overtones in Polybios. The Ai-

68 69 70

πατρίδος αὐτῶν τὰ τείχη ...; A32f.: ἠξίουν µὴ περιιδεῖν τὴν µεγίστην πόλιν τῶν ἐν τῆι Μητροπόλει ἐξαλειφθεῖσαν. A 42-44: δεδόχθαι ἀποκρίνασθαι αὐτοῖς ὅτι µὲν τοῖς περὶ τὴν πόλιν γεγενηµένοις ἀκληρήµασιν πάντες Ξάνθιοι συνηχθέσθησαν. Chaniotis 2013c, 346-350 (‘words as emotive acoustic signals’). For dramatic elements in narratives of battles in Hellenistic historiography see Chaniotis 2005a, 189–213. Diodoros 31.3 and 31.5 (on envoys of the Rhodians and the Thracians in Rome, 167 BCE): ... (the Rhodian envoys) started mourning, ... they besought them with tears. ... ... (The Thracian envoys) gave vent to tears as they made their petitions. On reference to tears in Hellenistic historiography see Lateiner 2009, 122–125.

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

77

tolian Chlaineas, for instance, claims that the Aitolians alone faced the Macedonian King Antipatros for the sake of the security of the unfortunate victims (τῆς τῶν ἀδίκως ἀκληρούντων) of his injustice (9.30.3f.); in an assembly of the Achaian Koinon the Aiginetan orator Kassandros pointed to the misfortune (ἀκληρία) of his countrymen (12.8.9). Second, in their appeal for help, the Xanthian envoys used the phrase µὴ περιορᾶν (‘not to look upon one‘s need with indiference’) a standard phrase in the context of arousal of pity or in the praise of people who did not neglect the obligation to help and exhibited courageous, responsible, or honourable behaviour. We find it so often in orations or summaries of orations that it can be labelled as an acoustic signal of empathy.71 Thirdly, the burning down of houses that the envoys highlighted was generally regarded as an act of cruelty and represented the ultimate fear of any Greek community. For this reason, orators and historians used the image of burned houses, temples, and altars in order to arouse pity for the victims, anger against the assailants, and fear for the future.72 Two passages in Polybios illustrate this. The first passage – from a speech by the Aitolian Chlaineas – aims to arouse anger against Philip V, while the second aims at empathy. Chlaineas’ purpose was to stir up opposition to Macedonian rule (211 BCE); one of the shocking images with which he operates is precisely the image of burned houses: ‘he damaged [Lakonia] destroying the land; he damaged setting the houses on fire.’73 In the second passage, Polybios describes how an anonymous ambassador to an assembly of the Aitolians in 207 BCE motivated the Aitolians to abandon their alliance with barbarians (the Romans). The envoy mentions the burning of houses as barbaric behaviour:74 71

72

73 74

I.Ephesos 2001 line 13 (Ephesos, c. 300–297 BCE): µὴ περιιδεῖν ἀλλοτριωθὲν τὸ φρούριον (‘not to look with indifference at the loss of the fort’); IOSPE I2 32 B 25f. (Olbia, c. 200 BCE): µὴ περιιδεῖν τὴν ἐκ πολλῶν τετηρηµένηµ πατρίδα ὑποχείριον γενοµένην τοῖς πολεµίοις (‘not to watch with indifference how their native city, after it had been preserved for many years, is subjected by the enemy’); I.Oropos 307 ll. 19f. (Oropos, c. 150 BCE): µὴ περιιδε[ῖν] πόλιν Ἑλληνίδα ἐξανδραποδισθεῖσαν (‘not to remain indifferent toward the enslavement of a Greek city’); I.Olympia 53 lines 10 (Elis, late first century BCE): µὴ πε̣ριιδ̣ῖν κ̣[ ει]µέ̣ν ην ἐπ’ ἐδάφους (‘not to look with indifference at her [the city?] lying on the ground’). Cf. IG XII Suppl. 364 (Thasos, first century CE); SEG XIX 1613 line 16 (Skythopolis, early second century BCE); IG II2 1092 B line 23; 1224 e line 7 (Athens, second century CE). Polybios often mentions the burning of cities and sanctuaries in order to arouse hostility against his greatest enemies: the Aitolians and Philip V: e.g. 4.62 (the Aitolians burn the city and sanctuary at Dion); 4.67.3 (the Aitolians burn parts of the sanctuary at Dodona); 5.9.2f. (Philip V burns buildings in the Aitolian federal sanctuary at Thermos); 16.1.5 (Philip V burns temples and altars); 18.3.3 (oration of Alexander against Philip V). See also p. 74 and the following notes. Polybios 9.28.3: καὶ κατέφθειρε µὲν τέµνων τὴν γῆν, κατέφθειρε δ᾽ αἴθων τὰς οἰκίας. Polybios 11.5.6–7: καὶ κυριεύσαντες µὲν αὐτοὶ πόλεως οὔτ᾽ ἂν ὑβρίζειν ὑποµείναιτε τοὺς ἐλευθέρους οὔτ᾽ ἐµπιπράναι τὰς πόλεις, νοµίζοντες ὠµὸν εἶναι τὸ τοιοῦτο καὶ βαρβαρικόν. συνθήκας δὲ πεποίησθε τοιαύτας, δι᾽ ὧν ἅπαντας τοὺς ἄλλους Ἕλληνας ἐκδότους δεδώκατε τοῖς βαρβάροις εἰς τὰς αἰσχίστας ὕβρεις καὶ παρανοµίας.

78

Angelos Chaniotis If you were to capture a city, you would not allow yourselves to insult freemen or to burn their cities, because you regard this as cruel and barbarous. Yet you have made a treaty by which you have handed over all other Greeks to the barbarians, to be exposed to the most shameful insults and lawlessness.

These orations in Polybios find their correspondence in oration in contemporary assemblies, summarised in decrees. A decree of Eresos, rejecting the request of the descendants of former tyrants to be granted amnesty, is a good example (late fourth century BCE).75 Those who opposed their return used affective memory against the exiles. By narrating the tyrants’ crimes against the people in vivid detail, they aroused empathy and hostile emotions in the assembly:76 ... he exacted 20 thousand staters from the citizens, plundered the Greeks, demolished the altars of Zeus Philippios, started a war against Alexander and the Greeks, depriving the citizens of their weapons and expelling many of them from their city. He seized their wives and daughters and kept them captive on the acropolis, exacting through blackmail 3,200 staters. With his bandits he plundered the city and the sanctuaries, he set them on fire, even burning citizens. ...

The imagery employed by the anonymous orator – plundering, stealing, capturing married women and virgins, committing sacrileges – culminates with the image of fire and the burning of sanctuaries and people. We see precisely the same image in Octavian’s letter to Mylasa (31 BCE), in which the Roman ruler summarises what the envoys of Mylasa had told him about their fatherland’s fate during the war:77 When the city was captured, you lost many prisoners, many citizens were killed – some were even burned together with the city, as the cruelty of the enemies did not even stop in front of the most sacred sanctuaries. They also informed me about the plundered land and the burned farmsteads.

The aim of such narratives embedded in orations was empathy. In Eresos, those who were born after the fall of the tyrants should feel, after the narration of the tyrants’ crimes, the same anger as those who witnessed them. Octavian should acknowledge the suffering of the Mylaseans; the Xanthians should join their dis75 76

77

IG XII.2.526 + Suppl.; OGIS 8. The most recent discussion of this dossier and its historical context is by Ellis-Evans 2012. Lines 2–13: τοὶ[ς πολίτα]ις δισµυρίοις στάτηρας εἰσέπραξε [καὶ τοὶ]ς Ἔλλανας ἐλαΐζετ[ο] καὶ τοὶς βώµοις ἀ[νέσ]καψε τῶ ∆ίος τῶ [Φ]ιλιππί[ω], καὶ πόλεµον ἐξε[νικ]άµενος πρὸς Ἀλέξανδρον καὶ τοὶς Ἔλλανας τοὶς µὲν πολίταις παρελόµενος τὰ ὄπλα ἐξεκλάϊσε ἐκ τᾶς πόλιος [πα]νδάµι, ταὶς δὲ γύνα[ι]κας καὶ ταὶς θυγάτερας συλλάβων καὶ ἔρξα[ις] ἐν τᾶ ἀκροπόλι τρισχιλίοις καὶ διακοσίο[ις] στάτηρας εἰσέπραξε, ταν δὲ πόλιν καὶ τὰ ἶρ[α] διαρπάσαις µετὰ τῶν [λα]ΐσταν ἐνέπρησε κα[ὶ] σ[υ]γκατέκαυσε σώµατα [τῶν] πολίταν. I.Mylasa 602 lines 11–19: κρατ̣η̣[θεί]σης τῆς πόλεως, πολλοὺς µὲν αἰχµαλώτο[υς] ἀποβαλῖν, πολίτας οὐκ ὀλίγους µὲν φονευθέντας, τινὰς δὲ καὶ συνκαταφλεγέτας τῇ πόλε[ι,] τῆς τῶν πολεµίων ὠµότητος οὐδὲ τῶν ναῶν οὐδὲ τῶν ἱερῶν τῶν ἁγιωτάτων ἀποσχοµέ[ν]ης· ὑπέδιξαν δέ µοι καὶ περὶ τῆς χώρας̣ τῆς λελεηλατηµένης καὶ τῶν ἐπαύλεων τῶν ἐµπεπρησµένων. Cf. similar formulations in the fragmentary decree for Sotas in Priene (c. 270 BCE), in which the cruelty of the Galatians is castigated: I.Priene 17.

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

79

tant relative from Kytenion in their grief – and they claim that they did (συνηχθέσθησαν). Hellenistic orators pursued the same aim as the historian Phylarchos, and many of his Hellenistic colleagues. What Polybios criticises in Phylarchos is what contemporary orators were seeking to achieve (see pp. 63–66). CONCLUSIONS This small selection of sources illustrates the interest of Hellenistic historians in descriptions of emotions and theatrical behaviour, in the perception of life as a drama, in illusions, and in emotional arousal. This interest should be seen in the context of Hellenistic history and culture. It reflects both real-life experiences in the Hellenistic world and trends of contemporary literature and art. Theatricality, illusion, and emotional display were not historiographical inventions, but real aspects of political life in the Hellenistic world, inter alia necessary for a balanced relationship between the ruling elite and people in the cities, and between cities and kings.78 A crucial issue in the public image of statesmen and kings likewise was to maintain a balance between affability, necessary for their popularity, and inequality, necessary for their leadership. It is in this context that we should see theatricality and emotional display in public life in the Hellenistic period: as a strategy of communication in asymmetrical relations. Given the established constitutional status of the assembly in the cities, the Hellenistic statesmen had to rely on delicate skills of performance in order to manipulate the masses in the assembly and to preserve the illusion of the rule of the people. The role of theatricality in the fragile balance of power between the king and the polis was quite similar. An important element of their relation was the illusion of the city’s autonomy. In theory, a city should be free (eleuthera), administered by its own laws (autonomos), free of garrisons (aphrouretos), and free of tribute (aphrologetos). The kings had to construct an image of supremacy, which would legitimate their rule, and at the same time respect the illusion of civic autonomy. When the balance was disturbed, with too much affability or too great a distance, theatrical behaviour was regarded as ridiculous or as an expression of madness.79 It is the reality of theatrical behaviour and illusion that explains the interest of Hellenistic historians in these subjects. But theatricality and illusion were not only subjects treated by historians; they were also means that historians themselves applied, when they shaped their narratives as dramas. This practice calls for a different explanation, one which is related to more general characteristics of the Hellenistic world: the ubiquitous presence of spectacles, the refined art of acting, the illusions created by Hellenistic art.

78 79

I discuss this in detail in Chaniotis 2009a, 103–139 and 176–197; cf. Chaniotis 2010. Diodoros 29.32.

80

Angelos Chaniotis

In his Natural History, Pliny mentions the most famous mosaicist of the Hellenistic period, a certain Sosos:80 The most famous artist in this genre was Sosos, who laid the floor in Pergamon which they call the asarotos oikos (the unswept room) because by means of small tesserae tinted in various colors he depicted on a paved floor the debris from a meal and other such things as are customarily swept away, making it seem as if they had been left there. A marvellous feature in that place is a dove drinking water and casting the shadow of its head upon it, while other doves sun themselves and preen themselves on the rim of a large drinking vessel.

Exactly as a Hellenistic mosaicist made the viewers of his work see something that was not there, the historian’s narrative made his audience see an incident, listen to its protagonists, and feel with them. Phylarchos was the great master of this art but, as we have seen, Polybios, Poseidonios, and Diodors followed similar trends. Phylarchos was a main source of Plutarch in his Lives of Agis and Kleomenes. Although it is unlikely that the following passage is a verbatim quotation from his work,81 it still gives us an impression of how he might have described a dramatic episode of Spartan history. In 241 BCE Agis, the reformer king, expelled his fellow king Leonidas and made Leonidas’ son-in-law Kleombrotos king. Chilonis, Leonidas’ daughter, and Kleombrotos’ wife, faced a dilemma: to follow her father into exile, or to stay in Sparta with her husband. She decided to follow her father. A change of fortune made Leonidas king again and sent Kleombrotos as suppliant in the sanctuary of Poseidon. Plutarch describes the scene in which Leonidas, Kleombrotos, and Chilonis finally meet as a small drama, with dialogues, emotions, and vivid descriptions of body language.82 80

81 82

Pliny, Natural History 36.62: celeberrimus fuit in hoc genere Sosus, qui pergami stravit quem vocant asaroton oecon, quoniam purgamenta cenae in pavimentis quaeque everri solent velut relicta fecerat parvis e tessellis tinctisque in varios colores. mirabilis ibi columba bibens et aquam umbra capitis infuscans; apricantur aliae scabentes sese in canthari labro. Pelling 2002, 91–115, has argued that Plutarch extensively reworked his source material. Plutarch, Agis 16.3–18.1 (after Phylarchos?): ὁ δὲ Κλεόµβροτος εἰς τὸ τοῦ Ποσειδῶνος ἱερὸν ἐλθὼν ἱκέτευε· καὶ γὰρ ἐδόκει τούτῳ µᾶλλον ὁ Λεωνίδας χαλεπὸς εἶναι, καὶ παρεὶς τὸν Ἆγιν ἐπὶ τοῦτον ἀνέβη στρατιώτας ἔχων, καὶ κατηγόρει µετ’ ὀργῆς, ὅτι γαµβρὸς ὢν ἐπεβούλευσεν αὐτῷ καὶ τὴν βασιλείαν ἀφείλετο καὶ συνεξέβαλε τῆς πατρίδος. Ὁ µὲν οὖν Κλεόµβροτος οὐδὲν εἶχεν εἰπεῖν, ἀλλ’ ἠπορηµένος ἐκάθητο καὶ σιωπῶν· ἡ δὲ Χιλωνίς, ἡ τοῦ Λεωνίδου θυγάτηρ, πρότερον µὲν ἀδικουµένῳ τῷ πατρὶ συνηδικεῖτο, καὶ τοῦ Κλεοµβρότου τὴν βασιλείαν παραλαβόντος ἀποστᾶσα τὴν τοῦ πατρὸς συµφορὰν ἐθεράπευε, καὶ παρόντι µὲν συνικέτευε, φεύγοντος δὲ πενθοῦσα καὶ χαλεπῶς ἔχουσα πρὸς τὸν Κλεόµβροτον διετέλει· τότε δ’ αὖ πάλιν ταῖς τύχαις συµµεταβαλοῦσα µετὰ τοῦ ἀνδρὸς ἱκέτις ὤφθη καθεζοµένη, περιβεβληκυῖα τὰς χεῖρας ἐκείνῳ καὶ τῶν παιδίων τὸ µὲν ἔνθεν, τὸ δ’ ἔνθεν ὑφ’ αὑτὴν ἔχουσα. θαυµαζόντων δὲ πάντων καὶ δακρυόντων ἐπὶ τῇ χρηστότητι καὶ φιλοστοργίᾳ τῆς γυναικός, ἁψαµένης τῶν πέπλων καὶ τῆς κόµης ἀτηµελῶς ἐχόντων, «τοῦτο» εἶπεν «ὦ πάτερ ἐµοὶ τὸ σχῆµα καὶ τὴν ὄψιν οὐχ ὁ Κλεοµβρότου περιτέθεικεν ἔλεος, ἀλλ’ ἀπὸ τῶν σῶν κακῶν καὶ τῆς σῆς φυγῆς µεµένηκέ µοι σύντροφον καὶ σύνοικον τὸ πένθος. πότερον οὖν δεῖ µε σοῦ βασιλεύοντος ἐν Σπάρτῃ καὶ νικῶντος ἐγκαταβιῶναι ταύταις ταῖς συµφοραῖς, ἢ λαβεῖν ἐσθῆτα λαµπρὰν καὶ βασιλικήν, ἐπιδοῦσαν ὑπὸ σοῦ τὸν παρθένιον ἄνδρα φονευόµενον; ὃς εἰ µὴ παραιτεῖταί σε µηδὲ πείθει τέκνων καὶ γυναικὸς δάκρυσι, χαλεπωτέραν ἢ σὺ βούλει δίκην ὑφέξει τῆς κακο-

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

81

And Kleombrotos came as a suppliant to the sanctuary of Poseidon, for he thought that Leonidas was bitter against him. Indeed, Leonidas left Agis in peace and went up against Kleombrotos with soldiers. He angrily accused Kleombrotos of plotting against him, although he was his son-in-law, robbing him of the royal power, and helping in sending him into exile. Having nothing to say, Kleombrotos sat perplexed and silent. Chilonis, however, Leonidas’ daughter, when her father was wronged, had joined him in his suffering. When Kleombrotos took the royal power, she left him and attended to her father in his misfortunes, joining him in his suppliant life while he was in the city, and in his exile continually grieving for him and having bitter feelings against Kleombrotos. This time, as Chilonis’ fortune changed back again with the changed fortunes of the men, she was seen sitting as a suppliant with her husband, embracing him, with a little child clinging to her on either side. When all were moved to wonder and tears at her virtue and affection, she touched her robes and her hair, alike unkempt, and said: ‘This garb, my father, and this appearance, are not due to pity for Kleombrotos. Ever since your sorrows and your exile grief has been my steadfast mate and companion. Must I, then, now that you are king in Sparta and victorious, continue to live in this misfortune, or put on a splendid and royal attire, after seeing the man I married as a virgin slain at your hands? Unless he persuades and wins you over by the tears of his wife and children, he will suffer for his evil plans a much harsher punishment than you wish; for he shall see me, his most beloved one, dead, before he is. For with what confidence could I live and face the other women, when I have failed to arouse pity in either husband or father? Both as wife and as daughter I was born to share only the misfortune and dishonour of those nearest and dearest to me. As for my husband, even if he had some plausible excuse for his actions, I robbed him of it by supporting you and giving testimony against his deeds; but you make it easy for him to defend himself for his wrong-doings, by showing that royal power is a thing so great and so worth fighting for that for its sake it is right to kill a son-in-law and ignore a child.’ Uttering such supplications Chilonis rested her face upon the head of Kleombrotos and turned her eyes, all melted and marred with grief, upon the bystanders. Then Leonidas, after consulting with his friends, ordered Kleombrotos to leave his asylum and go into exile, but asked his daughter to stay, and not to abandon him, since he loved her so much and since he had saved her husband’s life for her sake.

Although Plutarch must have reworked his source, I suspect that this passage reflects the empathy that Hellenistic tragic historiography sought to arouse. Hellenistic historiography had a performative element. It was not only read by readers; it was also read out aloud in public lectures (akroaseis) in theatres, gymnasia, festivals, and symposia.83 The dramatic features in a narrative, such as the one in the last passage, must have been stressed in its oral performance by the historian. The

83

βουλίας, ἐπιδὼν ἐµὲ τὴν φιλτάτην αὐτῷ προαποθανοῦσαν. τίνι γὰρ ἐµὲ δεῖ ζῆν παρρησίᾳ πρὸς ἄλλας γυναῖκας, ᾗ µήτε παρ’ ἀνδρὸς δεοµένῃ µήτε παρὰ πατρὸς ἔλεός ἐστιν; ἀλλὰ καὶ γυνὴ καὶ θυγάτηρ συνατυχεῖν καὶ συνατιµάζεσθαι τοῖς ἐµαυτῆς ἐγενόµην. τούτῳ µὲν οὖν εἰ καί τις ἦν λόγος εὐπρεπής, ἐγὼ τοῦτον ἀφειλόµην, τότε σοὶ συνεξετασθεῖσα καὶ καταµαρτυρήσασα τῶν ὑπὸ τούτου γενοµένων· σὺ δ’ αὐτῷ τὸ ἀδίκηµα ποιεῖς εὐαπολόγητον, οὕτω µέγα καὶ περιµάχητον ἀποφαίνων τὸ βασιλεύειν, ὥστε δι’ αὐτὸ καὶ γαµβροὺς φονεύειν καὶ τέκνων ἀµελεῖν εἶναι δίκαιον». Ἡ µὲν Χιλωνὶς τοιαῦτα ποτνιωµένη τό τε πρόσωπον ἐπὶ τὴν κεφαλὴν ἐπέθηκε τοῦ Κλεοµβρότου καὶ τὸ βλέµµα διεφθαρµένον καὶ συντετηκὸς ὑπὸ λύπης περιήνεγκεν εἰς τοὺς παρόντας. ὁ δὲ Λεωνίδας διαλεχθεὶς τοῖς φίλοις, τὸν µὲν Κλεόµβροτον ἐκέλευσεν ἀναστάντα φεύγειν, τῆς δὲ παιδὸς µένειν ἐδεῖτο καὶ µὴ καταλιπεῖν ἑαυτόν, οὕτω φιλοῦντα καὶ δεδωκότα χάριν τὴν τοῦ ἀνδρὸς αὐτῇ σωτηρίαν. On pity in the Chilonis episode see Pelling 2005, 285f. Chaniotis 1988, 367f. and 2009b.

82

Angelos Chaniotis

performative context of Hellenistic historiography often was that of the celebration, the festive gathering in a sanctuary or a theatre, in public spaces associated with competitions and spectacles, in front of audiences, who loved to watch. The historians could not present images in a material form (bloody battles, sophisticated siege engines, impressive garments and scenery, the thrill of a competition), but what they could do was to use language and narrative in order to paint mental pictures of a scene, such as the episode in Pellene, with which I opened this study, and the drama of Chilonis, with which I close it. Hellenistic historians treated the theatrical behaviour of their protagonists (statesmen and kings), the illusions that the subjects of their narratives had (the masses in the cities), and emotional display and arousal in public life. But while they were doing so, they were themselves agents of theatrical behaviour, illusion, and emotion. This needs to be taken into consideration not only when we study the works of Hellenistic historians as literary compositions, but also when we use them as sources of historical information. In his criticism of Phylarchos, Polybios criticises drama, more generally:84 The tragic poet seeks to thrill (ἐκπλῆξαι) and entertain (ψυχαγωγῆσαι) his audience for the moment by the most plausible words possible, but the historian’s task is to present true actions and words, in order to instruct and persuade for all time those who seek learning. In the first case [drama] priority is given to plausibility, even if what is said is untrue, in order to beguile the spectators (διὰ τὴν ἀπάτην τῶν θεωµένων); in the second case [history] priority is given to truth, in order to benefit those who love learning.

Hellenistic audiences preferred to be charmed than to be educated; they liked to be thrilled, they wanted to show pity. To be successful, Hellenistic historians had to become masters of empathy. Their instruments were illusion, enargeia, emotional language, and theatricality. BIBLIOGRAPHY Anderson, G. (2003) The Athenian Experiment: Building an Imagined Political Community in Ancient Attica, 508-490 B.C., Ann Arbor. Bathrellou, E. (2012) Menander’s Epitrepontes and the Festival of the Tauropolia, Classical Antiquity 31, 151–192. Beaton, R. (1983) The History Man, Journal of the Hellenic Diaspora 10, 1983, 23–44. Blok, J. H. (2000) Phye’s Procession: Culture, Politics and Peisistratid Rule, in H. SancisiWeerdenburg (ed.), Peisistratos and the Tyranny: A Reappraisal of the Evidence, Amsterdam, 17–48. Bousquet, J. (1985) La stèle des Kyténiens au Létôon de Xanthos’, Revue des Études Grecques 101, 12–53. Bowra, C. M. (1967) Constantine Cavafy and the Greek Past, in id., The Creative Experiment, London, 29–60. 84

Polybios 2.56.11f.: ἐκεῖ µὲν γὰρ δεῖ διὰ τῶν πιθανωτάτων λόγων ἐκπλῆξαι καὶ ψυχαγωγῆσαι κατὰ τὸ παρὸν τοὺς ἀκούοντας, ἐνθάδε δὲ διὰ τῶν ἀληθινῶν ἔργων καὶ λόγων εἰς τὸν πάντα χρόνον διδάξαι καὶ πεῖσαι τοὺς φιλοµαθοῦντας, ἐπειδήπερ ἐν ἐκείνοις µὲν ἡγεῖται τὸ πιθανόν, κἂν ᾖ ψεῦδος, διὰ τὴν ἀπάτην τῶν θεωµένων, ἐν δὲ τούτοις τἀληθὲς διὰ τὴν ὠφέλειαν τῶν φιλοµαθούντων.

Empathy, Emotional Display, Theatricality, and Illusion in Hellenistic Historiography

83

Cairns, D. (2009) Weeping and Veiling: Grief, Display, and Concealment in Ancient Greek Culture, in T. Fögen (ed.), Tears in the Graeco-Roman World, Berlin, 37–57. Caliò, L. M. (2005) Theatri curvaturae similis: note sull’urbanistica delle città a forma di teatro, Archeologia Classica 56, 49–130. Chaniotis, A. (1988) Historie und Historiker in den griechischen Inschriften. Epigraphische Beiträge zur griechischen Historiographie, Stuttgart. ––– (1997) Theatricality Beyond the Theater: Staging Public Life in the Hellenistic World, in B. Le Guen (ed.), De la scène aux gradins. Thêatre et représentations dramatiques après Alexandre le Grand dans les cités hellénstiques (Pallas 41), Toulouse, 219–259. ––– (2005a) War in the Hellenistic World: A Social and Cultural History, Malden. ––– (2005b) Griechische Rituale der Statusänderung und ihre Dynamik, in M. Steinicke and S. Weinfurter (eds.), Investitur- und Krönungsrituale, Cologne/Weimar, 1–19. ––– (2009a) Θεατρικότητα καὶ δηµόσιος βίος στὸν ἑλληνιστικὸ κόσµο, Herakleion. ––– (2009b) Travelling Memories in the Hellenistic World, in R. Hunter and I. Rutherford (eds.), Wandering Poets in Ancient Greek Culture: Travel, Locality, and Panhellenism, Cambridge, 249–269. ––– (2010) Illusions of Democracy in the Hellenistic World, November 2010: http:// athensdialogues.chs.havard.edu/cgibin/WebObjects/athensdialogues.woa/wa/dist?dis=43”) ––– (2011) The Ithyphallic Hymn for Demetrios Poliorcetes and Hellenistic Religious Mentality, in P. P. Iossif, A. S. Chankowski, and C. C. Lorber (eds.), More than Men, Less than Gods. Studies in Royal Cult and Imperial Worship. Proceedings of the International Colloquium Organized by the Belgian School at Athens (1-2 November 2007), Leuven, 157–195. ––– (2013a) Paradoxon, Enargeia, Empathy: Hellenistic Decrees and Hellenistic Oratory, in C. Kremmydas and K. Tempest (eds.), Hellenistic Oratory: Continuity and Change, Oxford, 201–216. ––– (2013b) Affective Epigraphy: Emotional Display in Public Inscriptions of the Hellenistic Period, Mediterraneo Antico (forthcoming). ––– (2013c) Emotional Language in Hellenistic Decrees and Hellenistic Histories, in M. Mari and J. Thornton (eds.), Parole in movimento. Linguaggio politico e lessico storiografico nel mondo ellenistico, Pisa, 339–352. ––– (2013d) Under Siege: Challenges, Experiences, and Emotions, in B. Campbell and L. Tritle (eds.), The Oxford Handbook of Warfare in the Classical World, Oxford, 438–456. Connor, W. R. (1987) Tribes, Festivals and Processions: Civic Ceremonial and Political Manipulation in Archaic Greece, Journal of Hellenic Studies 107, 40–50. Curty, O. (1995) Les parentés légendaires entre cités grecque, Geneva, 183–191. Dallas, J. (1974) Καβάφης καὶ ἱστορία. Αἰσθητικὲς λειτουργίες, Athens. Dana, M. and D. Dana (2001–03) Histoires locales dans le Pont-Euxin Ouest et Nord. Identité grecque et construction du passé, Il Mar Nero 5, 91–111. Deshours, N. (2006) Les mystères d’Andania. Étude d’épigraphie et d’histoire religieuses, Geneva. Ellis-Evans, A. (2012) The Tyrants Dossier from Eresos, Chiron 42, 183–212. Gernet, L. (1968) Anthropologie de la Grèce antique, Paris. Hirst, A. (1995) Philosophical, Historical, and Sensual: An Examination of Cavafy’s Thematic Collections, Byzantine and Modern Greek Studies 19, 33–93. Icks, M. (2011) Elevating the Unworthy Emperor: Ritual Failure in Roman Historiography, in A. Chaniotis (ed.), Ritual Dynamics in the Ancient Mediterranean: Agency, Emotion, Gender, Representation, Stuttgart, 347–376. Jones, C. P. (1999) Kinship Diplomacy in the Ancient World. Cambridge Mass./London. Jusdanis, G. (1987) The Poetics of Cavafy: Textuality, Eroticism, History, Princeton. Keeley, E. (1995) Cavafy’s Alexandria. Study of a Myth in Progress, Cambrige, Mass. Lateiner, D. A. (2009) Tears and Crying in Hellenic Historiography, in T. Fögen (ed.), Tears in the Greco-Roman World, Berlin, 105–134.

84

Angelos Chaniotis

Ma, J. (1994) Black Hunter Variations, Proceedings of the Cambridge Philological Society 40, 4980. Marincola, J. (2010) Aristotle’s Poetics and ‘Tragic History’’, in S. Tsitsiridis (ed.), Παραχορήγηnµα - Μελετήµατα γιὰ τὸ ἀρχαῖο θέατρο πρὸς τιµήν τοῦ καθηγητῆ Γρηγόρη Μ. Σηφάκη, Herakleion, 445–460. Mastrocinque, A. (1979) Demetrios tragodoumenos. Propaganda e letteratura al tempo di Demetrio Poliorcete, Athenaeum 57, 260–276. Otto, N. (2009) Enargeia: Untersuchung zur Charakteristik alexandrinischer Dichtung, Stuttgart. Pelling, C. B. R. (1995) Plutarch’s Adaptation of His Source Material, in B. Scardigli (ed.), Essays on Plutarch’s Lives, Oxford, 125–154. ––– (2002) Plutarch and History, London/Swansea. ––– (2005) Pity in Plutarch, in R. H. Sternberg (ed.), Pity and Power in Ancient Athens, Cambridge, 277–312. Sacks, K. (1981) Polybius on the Writing of History, Berkeley. Schepens, G. (2004) Polybius on Phylarchus’ ‘Tragic” Historiography’, in Schepens and Bollansée (eds.) 2004, 141–164. ––– and J. Bollansée (eds.) (2004) The Shadow of Polybius. Intertextuality as a Research Tool in Greek Historiography. Proceedings of the International Colloquium, Leuven, 21-22 September 2001, Louvain. Schmitzer, U. (1998) Sieben Thebaner gegen Theben. Bemerkungen zur Darstellungsform in Xenophon, hell. 5,4,1–12, Würzburger Jahrbücher für die Altertumswissenschaft NF 22, 123–139. Scholten, J. B. (2000) The Politics of Plunder. Aetolians and their Koinon in the Early Hellenistic Era, 279-219 B.C., Berkeley. Sinos, R. H. (1993) Divine Selection: Epiphany and Politics in Archaic Greece, in C. Dougherty and L. Kurke (eds.), Cultural Poetics in Archaic Greece: Cult, Performance, Politics, Cambridge, 73–91. Sokolowski, F. (1962) Lois sacrées des cités grecques. Supplément, Paris. Thonemmann, P. (2005) The Tragic King: Demetrios Poliorketes and the City of Athens, in O. Hekster and R. Fowler (eds.), Imaginary Kings. Royal Images in the Ancient Near East, Greece and Rome, Stuttgart, 63–86. van der Stockt, L. (2004) ‘Πολυβιάσασθαι’? Plutarch on Timaeus and ‘Tragic History’, in Schepens and Bollansée (eds.) 2004, 271–305. Versnel, H. S. (1987) What did Ancient Man See When He Saw a God? Some Reflections on Greco-Roman Epiphany, in D. van der Plass (ed.), Efigies Dei. Essays on the History of Religions, Leiden, 42–55. Walbank, F. W. (1938) Philippos tragoidoumenos. A Polybian Εxperiment, Journal of Hellenic Studies, 58, 55–68. Walbank, F. W. (1955) Tragic History: A Reconsideration, Bulletin of the Institute of Classical Studies 2, 4–14. Walbank, F. W. (1960) History and Tragedy, Historia 9, 216–234. Webb, R. (2009) Ekphrasis, Imagination, and Persuasion in Ancient Rhetorical Theory and Practice, Farnham. Zangara, A. (2007) Voir l’histoire: Théories anciennes du récit historique IIe siècle avant J.-C. IIe siècle après J.-C., Paris.

A SHORT HISTORY OF SHUDDERS Douglas Cairns I take my starting point in this paper from the first passage to be discussed in the stimulating and persuasive contribution to this volume by Angelos Chaniotis (pp. 55–57).1 On Plutarch’s description of the battle in Sikyon,2 Chaniotis remarks (p. 56): ‘Far more interesting than the description of emotions in this narrative is the arousal of emotions through the narrative.’ The emotions referred to in the passage in question are φρίκη and θάµβος. Φρίκη, as we shall see, can be the name of an emotion, but its primary significance lies in its reference to a physical symptom that is common to a range of emotional and non-emotional events. The nature of this symptom is suggested by the behaviour of the cognate noun, φρίξ, in Homeric poetry. Φρίξ is used (three times in the Iliad, once in the Odyssey) of the rippling effect of the sea; the word recurs thereafter mainly in Homeric scholarship, though it does appear occasionally in medical writers as a synonym of φρίκη (just as φρίκη itself is occasionally used in the sense that belongs to φρίξ in Homer). While φρίξ is used mainly of bodies of water, and φρίκη of humans and other animals, the former is informative about the latter in that its reference is to a form of movement that is visible to an observer. Φρίξ is something that happens to the sea, but the term is applied on the basis of how that experience appears to those who observe it. In the same way, φρίκη is an experience of an animal, but what the application of the term pinpoints is the visible aspect of that experience in the eyes of others. When this term is applied to an emotional experience, what we are dealing with is (in the strict sense) the phenomenology of emotion, that is, the shared, third-person perspective that we all have (notwithstanding the standard philosophical puzzles about the communicability of qualia) of what it is like to experience the emotion in the first person. To be sure, φρίκη is a subjective experience, but it is a subjective experience with an external, visible aspect, and it is this external, visible aspect that allows us to relate that person’s visible shudder, via the implicit theory of mind that

1

2

I am grateful to audiences in Amsterdam, Leeds, Berlin, Gießen, Rethymno, and Kyoto for valuable discussion, and to Jan Bremmer, Elizabeth Craik, Christoph Harbsmeier, Oliver Overwien, Keith Rutter, Richard Smith, and Philip van der Eijk for advice and assistance. I should also like to thank the Leverhulme Trust, whose award of a Major Research Fellowship enabled me to pursue the project of which this paper is part, and the Alexander von Humboldt-Stiftung, whose funding enabled me to make use of Berlin’s excellent academic libraries in bringing the paper to completion. Plutarch, Aratus 32.1–2.

86

Douglas Cairns

we develop from infancy, to our own subjective experience of shuddering and of the emotions of which shuddering is a symptom. Φρίκη belongs, in its primary sense, to the basic somatic level of emotion. Sources such as the Hippocratic corpus, other medical writings, and the collections of Problemata attributed to Aristotle and Alexander of Aphrodisias all give ample evidence of its basic somatic aspect.3 These sources are well-nigh unanimous in relating φρίκη, φρίσσω, etc. to bodily temperature: we shudder when we are cold, and when we shudder or shiver in other circumstances (for instance when we are afraid, when we are suffering from various physical ailments, when we sneeze, when we urinate, after eating, etc.) variations in bodily temperature are normally also implicated.4 The link between shuddering/shivering and piloerection (vestigial in humans) is frequently noted,5 and this can in turn provide a cue for comment on the occurrence of φρίκη also in non-human animals, both in circumstances which we should describe as emotional and in other, non-emotional scenarios.6 Φρίκη, therefore, is an involuntary bodily movement, one that is part 3

4

5

6

Cf. Burkert 2010, 48f. The Hippocratic corpus has 60 occurrences of the noun, φρίκη, 36 of the verb, φρίσσω (cf. one instance of φρικάζω), and 53 of the derivative adjective, φρικώδης (plus one of the synonym, φρικάλεος). In Galen, the figures are 110, 49, and 86 respectively. In medical writers, φρίκη is especially associated with fever and cold sweats (e.g., Hippocrates, Aphorismi 7.4; de morbo sacro 1.23–5; cf. Zink 1962, 19 n. 49; Berrettoni 1970, 262; Op de Hipt 1972, 210f.). In the Aristotelian Problemata, see especially Book 8 (887b10– 889b9) on chill (ῥῖγος) and shivering (φρίκη – nine occurrences of the root; there are a further 25 occurrences elsewhere in the work). Cf. e.g., Hesiod, Opera et dies 539f. (φρίσσω of human piloerection in cold weather); Plutarch, De primo frigido 947C (φρίκη and τρόµος are names for the ‘battle’ between hot and cold). Hippocrates and Galen each draw distinctions between φρίκη and ῥῖγος (the former is milder than the latter, according to Hippocrates, De morbo sacro 1.24; the former affects only the skin, the latter the whole body, according to Galen, De tremore vii, 612.9–12 ed. Kühn), though Galen notes that ‘all other medical writers’ use the terms interchangeably (De tremore vii, 611.18–612.4). Two late sources confirm Galen’s view of his fellow professionals (Palladius, Synopsis de febribus 24 in Ideler 1841, 117f.; Theophilus and Stephanus of Athens, De febrium differentia in Sicurus 1862, 30–32). Cf. the grammarians in n. 35 below. Cf. Berrettoni 1970, 263. Galen, however, insists on the existence of other causes, e.g. the application of bitter drugs (De tremore vii, 627.11–629.5 ed. Kühn). He also distinguishes between φρίκη and ῥῖγος as symptoms of fear and as signs of physical cold (ibid., 628.2–4); contrast [Aristotle], Problemata 889a15–25, on the role of bodily temperature in the emotions of fear and anger. In terms of etymology it is apparently ῥῖγος and not φρίκη that is related to Latin frigus, ‘cold’: see Chantraine 1968–80, 1249. E.g. [Aristotle], Physiognomonica 812b30; Problemata 888a38, 889a26; [Alexander of Aphrodisias], Problemata 2.26; cf. [Theocritus], Idyllia 25.244; Plutarch, fr. 73 ed. Sandbach. Cf. the frequency of the association between ‘goose bumps’ (UK English ‘goose pimples’) and physical cold (as also with fever and other biological functions such as sneezing) in the studies of Schurtz et al. 2012. Sophocles, fr. 875 ed. Radt; [Aristotle], Physiognomonica 812b30 (again); Nicander, Theriaca 721, 727; Plutarch, Aristeides 18.2 (developing the Homeric image by which weapons etc. bristle like the fur of an angry animal); Dio Chrysostom, Oratio 58.4; Achilles Tatius 1.12.3; fourteen times in Aelian, De natura animalium; [Alexander of Aphrodisias], Problemata 4.159.

A Short History of Shudders

87

of human beings’ pre-human inheritance and rooted in basic systems of bodily regulation that respond to changes in the temperature of the organism and of the environment. Φρίκη is thus an experiential concept. Though the mechanisms by which it is said to be produced (for instance in the Aristotelian and Alexandrian Problemata, in the Hippocratic corpus, in Galen, etc.) depend on physiological models (both popular and scientific) that we no longer share, the phenomenon itself is an undeniable fact of human and non-human physiology. As a symptom of emotion, and especially of fear-like emotions, it is a member of a set of related symptoms that are also recognised in our own folk models (‘I shudder to think’, ‘it gives me the shivers’, ‘he was in a cold sweat’, ‘she’s got cold feet’, ‘it was a chilling/hairraising experience’), and confirmed by empirical investigation.7 The fact that φρίκη is a symptom of fear and similar emotions in Greek is an important one, for the same symptom remains an important sign of such emotions and an important aspect of the concept of those emotions in English and other modern languages.8 This simple point is often minimised by those who would write the history of emotion: there are substantial aspects of emotional experience that depend on the biological heritage of our species and are deeply rooted in basic mechanisms of bodily regulation that human beings share with other animals.9 Where such aspects are prominent in cultures’ concepts of emotion we cannot expect the history of those concepts or the history of emotion itself to be one of unconstrained conceptual and cultural variation; change, development, or transformation will be, at least to some extent, constrained by physical embodiment. If this point is recognised at all by constructionist theorists of emotion, it is normally dismissed as trivial or uninteresting. Attention to emotions’ physical symptoms is typically stigmatised as reductionist and condemned for underestimating the crucial role in the conceptualisation of emotion of the evaluative, lin7

8

9

Specifically on symptoms of fear, see Darwin 1889 [1998], 70f., 346f. (trembling), 100f., 104f., 291f., 295–298 (piloerection), 291, 346f. (temperature changes), with P. Ekman’s comments and further reading where relevant; cf. Eibl-Eibesfeldt 1989, 371 (on piloerection), 479 (on changes in skin temperature); Balcombe 2010, 48 (on changes in body and skin temperature as symptoms of fear and other emotions in humans and animals); cf. also Burkert 2010, 46. Keltner and Haidt 2003 associate human piloerection, as a symptom of emotion, primarily with awe, as do Schurtz et al. 2012, who find an association between piloerection and fear in only a small number of their respondents (US college students, whose exposure to genuinely fear-inducing scenarios may be limited, as the authors observe, p. 210). We shall discuss the connexion between φρίκη and Greek forms of awe below; for the moment, I observe only that the association with fear and fear-like states is dominant in the ancient Greek context. For low body temperature as a metonym for fear in various cultures, see Kövecses 2000, 5, 23f.; for a survey of psychological applications of words meaning ‘warm’ and ‘cold’ in Greek, see Zink 1962, especially 15–30 on ‘“Kälte” als Ausdruck einer unangenehmen Gefühlslage wie Schreck, Angst, Furcht, Entsetzen, Grauen’; cf. also Bouvier 2011. On the relation between physical temperature and the metaphorical concepts of emotional warmth and coldness, see Williams and Bargh 2008; Zhong and Leonardelli 2008. See in particular Damasio 1995.

88

Douglas Cairns

guistic, social, and cultural aspects of emotional experience and concepts. 10 That these charges are unfounded can be demonstrated by further study of φρίκη. In Aratus 32 (see pp. 54–57), for example, the point of Plutarch’s use of that term is not primarily to draw attention to the bodily changes experienced by the Aetolians in reaction to the sudden appearance of Epigethes’ daughter. If physical shuddering is relevant at all, it is not as such, that is, as a simple, visible, bodily reaction, but as a sign of a more inclusive emotional experience, one that involves perception, an evaluation of that perception, cultural norms (in this case, specific concepts of divinity and of the possible modes of interaction between human and divine), and a characteristic pattern of behaviour. The term φρίκη refers not to one very limited aspect of that experience, the physical shudder that is a symptom of fear or something like it, but to the total experience. If this were not in any case clear, it would be strongly supported by the fact that the other term used, θάµβος, refers not to a physical symptom of emotion, but to an emotion as such.11 The information conveyed by the use of the term φρίκη is not that the Aetolians, for some reason, shuddered, but that they were (let us say) afraid; shuddering is a sign of that emotion, and regardless of whether we are to picture the Aetolians in this case as actually shuddering or not, it is as a sign and not merely as a symptom that the term is being used in this passage. The word φρίκη means not, or not only, that they shuddered, but that they were afraid. In short, φρίκη in this passage is being used in metonymy (or, if you prefer, synecdoche) for the total emotional experience of which it is a sign. Metonymy is a fundamental mechanism for extending the meaning of terms and in the formation of conceptual categories.12 Something of this importance can be gauged through the study of the metonymous extension of φρίκη. First, let us recall that, in emotional contexts, the primary reference of φρίκη and its cognates is to a certain perceptible physical movement, felt by the subject but also perceived by others. There are numerous passages in which φρίκη etc. refer to this movement as such. At Aeschylus, Agamemnon 1242–1244,13 for example, πέφρικα seems to highlight the spontaneous physical reaction, while the phrase καὶ φόβος µ’ ἔχει indicates its nature. At Euripides, Hecuba 85f., Hecuba’s personified φρήν shudders out of τάρβος,14 and this pattern, in which φρίκη the symptom is mentioned alongside the emotion which is its cause, is common.15 Similarly, in 10 11 12 13

14 15

For an egregious example, see Gross 2006. For θάµβος, θαῦµα, etc. in such contexts, see below n. 55. See above all Lakoff 1987. See also Theodoropoulou 2012, especially 454-457 on the interaction of metonymy and metaphor. [Chorus] τὴν µὲν Θυέστου δαῖτα παιδείων κρεῶν | ξυνῆκα καὶ πέφρικα, καὶ φόβος µ’ ἔχει | κλύοντ’ ἀληθῶς οὐδὲν ἐξῃκασµένα (‘Thyestes’ feast of children’s flesh I understand, and shudder. Fear grips me: what I hear is no fiction, but the truth’). Οὔποτ’ ἐµὰ φρὴν ὧδ’ ἀλίαστον | φρίσσει ταρβεῖ (‘never before has my mind shuddered so incessantly with dread’). On the personification, cf. below, note 21. E.g. Sophocles, Ajax 693; Plato, Phaedrus 251a; Menander, Epitrepontes 901; Plutarch, Agesilaus 24.5; De curiositate 516D; Aetia physica 914E; Philostratus, Heroicus 748.15–17; Appian, Bella civilia 1.11.97; Dio Chrysostom, Oratio 58.4; cf. Zink 1962, 20.

A Short History of Shudders

89

Plutarch’s Life of Alexander, an involuntary shudder is one of several physical symptoms (quivering, dizziness) experienced by Cassander at the sudden sight of a statue of Alexander at Delphi.16 And φρίκη regularly occurs in lists or explicit descriptions of the physical symptoms of emotion.17 In such cases, an actual symptom of emotion is presented in the context of a wider emotional response, and φρίκη is both an aspect of a total emotional scenario and a symptom that helps convey the ‘flavour’ of that scenario. In other cases, the relevant terms are used to describe an actual symptom, but that symptom is itself used as a sign, a kind of shorthand, that the emotional response with which it is typically associated is taking place. At Aeschylus, Supplices 346, for example, Pelasgus is no doubt referring to an actual shudder in describing his reaction to the Danaids’ supplication,18 but the point of that reference is to indicate his apprehension with regard to the power of the ritual and the sanctions that it entails. Similarly, in two passages of Achilles Tatius, the verb is used to describe actual shudders that function as signs of particular emotional reactions.19 The full extent of the metonymous extension of the term, however, is visible where there is no reference to any actual bodily change, and the relevant term is used to refer to the occurrence of the emotion as such. At Euripides, Troades 182f., for example, if it is to be the cause rather than a symptom of ἔκπληξις,20 φρίκη must be fear, not shuddering. Similarly, at Phoenissae 1284f., φρίκη is the cause of the chorus’ trembling φρήν,21 and it is thus much more likely that φρίκη is the emotion of fear 16

17

18 19

20 21

Plutarch, Alexander 74.6: εἰκόνος Ἀλεξάνδρου φανείσης ἄφνω πληγέντα φρῖξαι καὶ κραδανθῆναι τὸ σῶµα, καὶ µόλις ἀναλαβεῖν ἑαυτόν, ἰλιγγιάσαντα πρὸς τὴν ὄψιν (‘struck by the sudden appearance of a statue of Alexander he physically shuddered and trembled, dizzy at the sight’). Cassander’s vividly presented and powerful symptoms depend on his memory of a severe beating, years before, at Alexander’s hands; cf. below, note 50. [Aristotle], Physiognomonica 812b29f.; Problemata 887a1–3, 949a9–20, 964b22–29; cf. Aristotle, De motu animalium 701b20–22; Alexander of Aphrodisias, De anima 77.2–5 ed. Bruns. Πέφρικα λεύσσων τάσδ’ ἕδρας κατασκίους (‘I shudder as I look upon these shaded shrines’). Achilles Tatius 4.15.3 (the narrator shudders, ἔφριξα, at the mention of the name of Gorgias, which Leucippe had previously uttered in a dream); 5.25.4 (the narrator shudders when caught by his mistress, Melite, in possession of a letter from his true love, Leucippe); cf. [Aeschylus], Prometheus Vinctus 695, where the chorus’ spontaneous shudder at the tale of Io is construed as φόβος by Prometheus in line 696; or Theophrastus, Characteres 16.14, where the superstitious man’s shudder at the sight of a madman or epileptic is a sign of the superstitious fear that causes him to spit into the fold of his garments. Ὀρθρεύουσαν ψυχὰν | ἐκπληχθεῖσ’ ἦλθον φρίκᾳ (‘I have come, my soul struck with terror, φρίκη, in the sleepless dawn’). Αἰαῖ αἰαῖ, τροµερὰν φρίκᾳ | τροµερὰν φρέν’ ἔχω (‘alas, my mind trembles in terror, φρίκη’). NB also (as in Hecuba 85f., cited at note 14 above) the mechanism of personification by which the visible fear and trembling of the whole person is presented as the experience of an internal agent detectable only by the subject. For such imagery in the presentation of emotion as an internal experience invisible to others cf. Odyssey 10.374, 20.9–21, 20.301; and especially Odyssey 23.215–217 (αἰεὶ γάρ µοι θυµὸς ἐνὶ στήθεσσι φίλοισιν | ἐρρίγει, µή τίς µε βροτῶν ἀπάφοιτ’ ἐπέεσσιν | ἐλθών – ‘constantly my heart shivered in my breast, lest some mortal should come and deceive me with words’); cf. Plutarch, De curiositate 516D

90

Douglas Cairns

of which trembling is the sign than that it refers tautologously to a physical symptom.22 That a metonymy of this kind is in play is especially clear when the verb which primarily refers to the physical symptom takes the construction appropriate to the verb which names the emotion with which that symptom is particularly associated. This is the case when φρίσσω, ‘I shudder’, governs a direct object in the same way as verbs meaning ‘to fear’.23 Thus, in a famous passage, Helen contrasts the kindness of Hector with the horror that she occasions in the other Trojans:24 For I no longer have anyone else in broad Troy who is gentle or kind – all the others shudder 25 at me.

That the verb φρίσσω in such locutions stands for a verb of fearing is particularly clear in Euripides, Hippolytus 415–418, where Phaedra expresses her incredulity that an adulteress should be able to conceal her guilty conscience from her husband:26 How, oh Cyprian, mistress of the deep, can they look their husbands in the face and not shudder at the darkness, their partner in crime, or at the timbers of the house, lest they at some stage speak?

The fact that the verb φρίσσω is here followed not only by a direct object, but also by a noun clause of the sort that regularly specifies the propositional content of a verb of fearing indicates that ‘shudder’ here is a simple metonymy for ‘fear’; shudders as such have no propositional content.27

22 23 24 25

26 27

(the ψυχή shudders in fear). In this way, the visible physical symptom that is central to the inter-subjective phenomenology of the emotion is transferred even to the presentation of emotion as an undetectable subjective experience, emphasising how important such symptoms are to the ways in which we think and talk about emotion. Cf. Plutarch, De Alexandri magni fortuna aut virtute 343e, where φρίκῃ διέτρεσαν is more likely ‘trembled in terror’ than ‘trembled and shuddered’. For this phenomenon, cf. Apollonius Dyscolus, De constructione 413.5–415.2. Iliad 24.774f.: οὐ γάρ τίς µοι ἔτ’ ἄλλος ἐνὶ Τροίῃ εὐρείῃ | ἤπιος οὐδὲ φίλος, πάντες δέ µε πεφρίκασιν. Cf., e.g., Aeschylus, Septem contra Thebas 720f.; Euripides, Cyclops 320; Hippolytus 855; Sophocles, Antigone 997; Aristophanes, Nubes 1132f.; Diodorus Siculus 14.66.1, 28.2.1; Plutarch, Aemilius 29.5, 35.3; Cicero 22.2; Demosthenes 20.3; Tiberius et Gaius Gracchus 22.5; Otho 1.3; Quomodo adulescens poetas audire debeat 36E; De superstitione 166DE; De Alexandri magni fortuna aut virtute 331E–F; De tranquillitate animi 476D; De vitioso pudore 536CD; De sollertia animalium 980F; Ad principem ineruditum 781E; Appian, Hannibalica 183. The same phenomenon is observable when the noun, φρίκη, governs an objective genitive: Euripides, Ion 898; Plutarch, Timoleon 22.6. Αἳ πῶς ποτ’, ὦ δέσποινα ποντία Κύπρι, | βλέπουσιν ἐς πρόσωπα τῶν ξυνευνετῶν | οὐδὲ σκότον φρίσσουσι τὸν ξυνεργάτην | τέραµνά τ᾿ οἴκων µή ποτε φθογγὴν ἀφῇ; Cf. Odyssey 23.216 (note 21 above), where the verb in question is ῥιγέω. The response on which Phaedra comments in the hypothetical adulteress, of course, involves a failure to experience the guilty fear of exposure that Phaedra herself would feel in such a situation; thus, though still a form of fear, φρίκη is here implicated in a scenario that also encompasses prospective and retrospective shame. On this aspect of the wider context, see Cairns 1993, 321– 340.

A Short History of Shudders

91

Once a symptom of emotion such as φρίκη is established as a metonym for the emotion itself, the term then becomes implicated in a wider network of emotional imagery, in two ways: first, like any other emotional term, such a metonym attracts the standard sorts of metaphor in which emotional concepts are expressed; second, the metonym exists in relation to a nexus of other terms which turn upon the same aspect of the relevant emotion as it does. There is a straightforward illustration of the first phenomenon in the example from Plutarch’s Aratus from which we started: Epigethes’ daughter ‘put φρίκη and θάµβος in’ the enemy (32.2), that is, the emotion is a substance and those who experience it are its containers. The simple ‘container metaphor’ by which emotions and other psychological experiences are ‘in’ us or we are ‘full of’ them is a fundamental way in which human beings conceive of mental events in terms of embodied experience.28 Equally dependent on the ontological metaphor by which emotions are substances is the conduit metaphor, in which the emotion is conveyed from one person to another.29 Also related, and fundamental to the view of emotions as phenomenologically passive experiences, is the metaphor of emotion as an antagonist or external force, whether as an entity that ‘seizes’ or ‘holds’ one,30 or as something that ‘comes over’ or ‘comes upon’ one from outside.31 And we have also noted the mechanism by which the visible and physical symptoms of emotions are presented as experiences of personified internal agents (see note 21). In addition, the experiential network in which φρίκη belongs (in which φρίκη-type emotions are typified by the lowered body temperature, shivering, and piloerection that are their physical symptoms) encompasses a range of other terms which express complementary aspects of the same imagery. Thus Helen, before whom the Trojans πεφρίκασιν at Iliad 24.775 is described as ῥιγεδανή at 19.325: the metonymy naturally encompasses both patient and object – Helen is ‘cold’ and she makes the Trojans ‘shiver’. The verbs ῥιγέω and ῥιγόω, when they are not used literally of physical temperature, participate in this network mainly in the description of emotional symptoms, though ῥιγέω is regularly used as a metonym for the emotion itself.32 The comparative ῥίγιον appears to be almost exclusively 28

29 30

31 32

Plutarch offers several additional examples using φρίκη: see Crassus 26.5 (φρίκη ‘comes to be in’); Marius 43.6 (‘full of φρίκη’); Pelopidas 27.7 (φρίκη ‘caused in’); Comparatio Lysandri et Sullae 2.4 (‘causing φρίκη in’); De cohibenda ira 461E (‘putting φρίκη in’); cf. Aratus 54.6 (‘filled the kingdom with φρίκη and hatred towards him’). On the container metaphor, cf. Lakoff and Johnson 1980, 29–32, 58–60, 73–75, 92–96, 98–105, 161, 148; for its use of emotion, cf. Kövecses 2000, 141–163 and passim. E.g., Plutarch, Sulla 11.1: ὥστε φρίκην µὲν τῷ δήµῳ ... παρασχεῖν (‘so as to give φρίκη to the people’); cf. Lakoff and Johnson 1980, 11; Kövecses 2000, 89. Plutarch, Aemilius 17.8 (κατεῖχε); Caesar 66.9 (εἶχε); Publicola 7.1 (εἶχε); Chariton, De Callirhoe narrationes amatoriae 1.8.2 (ἐλάµβανεν); cf. its occurrence at Hippocrates, Epidemiae 7.1.11 of φρίκη as a symptom of physical disease (φρῖκαί τε αὐτῇ ἔστιν ὅτε ἐνέπιπτον; ‘she suffered from occasional attacks of φρίκη’). Cf. Kövecses 2000, 68–70. E.g., Herodotus 6.134 (ὑπελθεῖν); Achilles Tatius 5.21.2 (ὑποδραµεῖν); Philostratus, Heroicus 666.7f. (ὑπελθεῖν). In Homer, ῥιγέω seems to refer consistently to shuddering or shivering, whether as a symptom of emotion (twelve times in the Iliad, twice in the Odyssey) or as a metonym for it (Iliad

92

Douglas Cairns

metaphorical, used of unpleasant eventualities and states of affairs that occasion fear and similar emotions in those who encounter or contemplate them;33 the superlative ῥίγιστος and other derivative adjectives can also be used in a similar sense. 34 Also part of the same nexus of meaning are the adjectives κρυόεις and ὀκρυόεις, which likewise have a metaphorical application to the objects of fear and similar emotions – hence Helen is not only ῥιγεδανή at Iliad 19.325 but also ὀκρυόεσσα at 6.344.35 The implications of all this for the cross-cultural study of emotion are considerable. The importance of emotional symptoms in the construction of emotional concepts underlines the fundamental importance of physical embodiment in the concept of emotion itself. In the case of φρίκη, the symptom is one that has its roots in basic somatic mechanisms of temperature regulation, that is manifested in a range of non-emotional contexts, and that is shared with other animals. From these materials, universal in humans and extending beyond the human species, is constructed a concept of (a kind of) fear in which physical symptoms are intimately related to cognitive appraisals and evaluations. The mechanism by which this occurs is the universal one of metonymy, by which the name of the symptom comes to function as a name of the emotion. This metonym is then enmeshed in the wider network of emotional metaphor that is, again, a feature of the conceptualisation of emotion in all cultures and languages known to me. At all these levels, the concept of φρίκη is typical in locating the language and thought of emotion in embodied physical experience. There is nothing in any way surprising or unfamil-

33

34

35

3.353f., 5.350f., 7.113f., 17.174f.; Odyssey 23.215–217; cf. Pindar, Nemea 5.50; Apollonius Rhodius 3.437f.; Rhianus, fr. 1.7 ed. Powell). It seems never to be used of simply being physically cold, a sense which is reserved for ῥιγόω (only at Odyssey 14.481): see Zink 1962, 15f. In post-Homeric literature the question of a distinction between the two verbs is complicated by the existence of ambiguous forms, and though ῥιγόω has irregular contractions that seem designed to distinguish it from ῥιγέω, still the two can be used interchangeably (as at [Aristotle], Problemata 948b13–15). Ῥίγιον: thrice in the Iliad, twice in the Odyssey; cf. Hesiod, Opera et dies 703; Mimnermus, fr. 4.2 ed. West; Semonides, fr. 6.2 ed. West; Apollonius Rhodius 3.402f., 429f.; Odyssey 17.191 is the only use of the term with reference to physical temperature; cf. Zink 1962, 22f.. For ῥίγιστος, see e.g. Iliad 5.873f.; Lycophron, Alexandra 2.292; Apollonius Rhodius 2.215f.; Nicander, Theriaca 64; cf. ῥιγηλός: [Hesiod], Scutum 131; Nicander, Alexipharmaca 220. Cf. Iliad 8.64 (of war); Apollonius Rhodius 2.607 (of fear); Quintus Smynaeus, 1.133, 539 (battle), 13.88 (death). Cf. κρυόεις Iliad 5.740 (battle), 9.2 (rout); Hesiod, Theogony 936 (war); [Hesiod], Scutum 255 (Tartarus); Stesichorus, S11.5 PMGF (death); Pindar, Pythia 4.73 (prophecy); Bacchylides, fr. 60.12 ed. Maehler (war); Euphorion, Supplementum Hellenisticum 415 col. 2.3 edd. Lloyd-Jones/Parsons (war); Batrachomyomachia 73 (fear); Quintus Smynaeus 7.363 (fear); cf. Zink 1962, 24f. For these various adjectives as synonyms, see Apollonius Sophista, Lexicon Homericum 138.32 (ῥίγησεν ἔφριξεν, καὶ ῥίγιον τὸ φρικτὸν καὶ χαλεπόν, καὶ ῥιγέστατα τὰ φρικτὰ καὶ χαλεπά, καὶ «ῥιγεδανῆς Ἑλένης» τῆς φρικώδους); Hesychius. κ 4252 (κρυόεσσα = φρικτή), 4253 (κρυόεσσα ἰωκή = ἡ φρικτὴ καὶ φοβερὰ βοή), 4264 (κρυώδους = φρικώδους), ο 494 (ὀκρύοεν = φρικῶδες), 495 (ὀκρυοέσσης = φρικτῆς), ῥ 299 (ῥιγεδανῆς = φρικώδους), 300 (ῥιγεδανόν = φρικῶδες), 301 (ῥίγιστα = φρικωδέστατα).

A Short History of Shudders

93

iar about this – the point is precisely that ancient Greek emotional concepts are, to a large extent, built up out of the same materials as our own, materials that draw on our experience as physically embodied beings interacting with our physical and social environments. What needs to be emphasised, however, is that this experiential, embodied nature of emotion is not just an aspect of a shared biological substratum; it is a feature also of language and of thought. It is not that embodiment is relevant only in terms of emotions’ physical changes, symptoms, and expressions and is left behind when emotional concepts take root in language, thought, and culture. There is no disjunction, but rather a fundamental continuity between emotions as physical experiences and emotional concepts as linguistic and cultural categories. In terms of the development of emotional concepts, there is no wedge to be driven between the body on the one hand, and language and culture on the other. Linguistic and literary studies of emotion have tended to concentrate exclusively on the sense and reference of the terms that label emotional experiences and scenarios. Unsurprisingly, they have found that cultures label emotions in different ways; the emotional labels of one culture are often not coextensive in reference with those of others; emotional scenarios that are labelled in one culture may go unlabelled, ‘hypocognised’, or even unrecognised in another. But there is much more to the language of emotion than this. Attention to aspects of emotion language beyond the semantics of emotion terms not only reinstates the role of embodied experience but provides better evidence of a culture’s phenomenology of emotion, getting us as close as we can get to a culture’s attempt to encapsulate the subjective experience of emotion in language.36 There is, then, an irreducibly physical and instinctive core to the emotion of φρίκη, and the term can also stand as a metonym for the fear-like emotion of which it is a symptom. Though there can be disagreement over which emotions are ‘basic’, over the distinction between basic and non-basic emotions, or over the differences between basic and non-basic forms of the same emotion, only the most extreme social constructionist would deny that both shivering and feeling afraid are not only cross-cultural but also inter-specific experiences. I do not think, however, that we should simply leave things there. It is not enough to say (for example) that φρίσσω means ‘I shudder’ and that ancient Greeks, like modern Glaswegians, said ‘I shudder’ as a way of saying ‘I am afraid’, because, for them as for us, shuddering is a symptom of fear that can be used as a metonym for the emotion itself. While this is true, φρίκη requires more specific attention than that. Fear is not single. There are occurrent and dispositional forms, for instance, the sudden shock that comes over us when we see a snake and the dispositional fear of snakes that leads us, on the whole, to avoid situations in which we might see one. There is a similar distinction between instinctive (action-programme) and more reflective (higher cognitive) forms:37 my sudden reaction to an unexpected noise has something in common with, but is not exactly the same as, my appre36 37

Cf. Burkert 2010, 54. The terms are those of Griffiths 1997.

94

Douglas Cairns

hension about what the UK government may do to our public services. Equally, though φρίκη and its cognates can take the place of φόβος, δέος, etc. in purely general senses, it none the less remains possible to discern certain specific nuances in the usage of φρίκη etc. that make it unwise simply and wholly to assimilate them to either ancient Greek or modern English concepts of fear. First, φρίκη etc., even when they are used of emotional reactions rather than of purely physical conditions, do not always seem to refer to fear. There are many cases in which it is possible to be uncertain that what we call fear is precisely what the terms imply. No doubt each of us would choose a slightly different list of examples, but for my money, one such would be the passage we have already discussed from Iliad 24 in which φρίκη is the Trojans’ reaction to Helen (24.775). This response is contrasted with Hector’s kindness and friendliness, and while one could make a case for Helen as an object of fear – as a symbol or omen, perhaps, of the dire consequences that her presence in Troy portends – it is at least possible that we are meant to think of something more like anger or disgust.38 An episode in the sixth book of Josephus’ Bellum Iudaicum (6.201–219) confirms that φρίκη can presuppose a strong element of revulsion. During the siege of Jerusalem, a starving woman cooks and eats her own son in a desperate attempt to avenge herself upon the Jewish guards whose depredations have reduced her to this level. The guards are immediately struck dumb in φρίκη and παρέκστασις at this (210);39 the news spreads at once through the city, and all visualise and shudder at the event as if they had committed it themselves.40 The Romans, for their part, react with disbelief, pity, and µῖσος. The reaction adumbrated by φρίκη and φρίσσω in this passage is not fear of something threatening or dangerous, but shock and revulsion at something that has already happened. In another example, a passage of Demosthenes’ speech On the Crown vividly depicts the non-verbal behaviour of those who go around the agora beaming with joy at the successes of Athens’ enemies, holding out their right hands and spreading the good news, but shudder, groan, and hang their heads when events turn Athens’ way.41 Perhaps the shuddering of such people expresses their fear that the city’s success entails negative consequences for themselves, but on the face of it πεφρικώς, like the groaning and dejection, seems to focus on the ‘bad news’ as such, and since this has already occurred, the reaction is more like sorrow than fear.42 Fear is certainly not 38

39 40 41

42

Cf. Zaborowski 2002, 235f. Small but not entirely insignificant numbers of respondents associate piloerection with anger or disgust also in the contemporary study (of US college students) conducted by Schurtz et al. 2012: see especially their Table 1, p. 209. Παρέκστασις (found in one MS and printed by Niese) occurs only here; all other MSS (and testimonia) have φρενῶν ἔκστασις. The choice of reading does not affect my argument. Josephus, Bellum Judaicum 213: πρὸ ὀµµάτων ἕκαστος τὸ πάθος λαµβάνων ὥσπερ αὐτῷ τολµηθὲν ἔφριττε. Demosthenes 18.323: οὐκ ἐπὶ µὲν τοῖς ἑτέρων εὐτυχήµασι φαιδρὸς ἐγὼ καὶ γεγηθὼς κατὰ τὴν ἀγορὰν περιέρχοµαι, τὴν δεξιὰν προτείνων καὶ εὐαγγελιζόµενος τούτοις οὓς ἂν ἐκεῖσ᾽ ἀπαγγελεῖν οἴωµαι, τῶν δὲ τῆς πόλεως ἀγαθῶν πεφρικὼς ἀκούω καὶ στένων καὶ κύπτων εἰς τὴν γῆν, ὥσπερ οἱ δυσσεβεῖς οὗτοι. Demosthenes alludes to Aeschines’ alleged behaviour (cf. 217, 244, 291); cf. Aeschines 3.164. So Wankel 1976, 1351 ad loc.

A Short History of Shudders

95

the issue at Sophocles, Ajax 693, where the chorus’ shiver (a veritable frisson) expresses their joy at Ajax’s apparent salvation, a joy that they hyperbolically describe as the intense, passionate, and sexual emotion of ἔρως.43 On occasion, an analogy with the phenomenon of piloerection in animals makes φρίκη an aggressive and angry response rather than a fearful one.44 Plutarch’s Life of Aristides (18.2), for example, develops the implicit Homeric analogy between an army’s bristling weapons (or their helmets’ bristling crests) and the piloerection of an angry animal in presenting an image of the Greek forces at Plataea as an angry beast.45 Two further passages in Plutarch use φρίσσω as a metonym not for fear but for awe or admiration, that of Philip for the power of Demosthenes’ oratory to mobilise opposition and that of Alexander for Diogenes the Cynic.46 Moreover, even when φρίκη etc. refer to something like fear, it is often the case that the reference is to fear of a more or less specific kind. Φρίκη and its cognates seem, first of all, to be particularly associated with immediate, automatic, and instinctive responses to sudden or direct visual or aural stimuli. The Aristotelian Problemata discuss φρίκη as a spontaneous reaction to various unpleasant sounds (886b9–11, 964b34–37), and this reflex reaction is then explicitly related to fear, on the basis that such sounds are instinctively regarded as signs of impending trouble (887a1–3).47 This particular association between φρίκη and immediate visual or aural stimuli is widely confirmed,48 and is reflected in the way that the adjective φρικώδης very often qualifies sights and sounds: though many of these passages include a reference to the ominous connotations or negative import of the sights or sounds in question, it is clear that in many cases the adjective also highlights the capacity of the stimulus to elicit an instinctive emo43

44

45

46 47

48

They associate the intensity of their emotion with the god Pan, in whose honour they sing and dance (693–700): see Borgeaud 1979, 219f. (with 221f. note 140). For the connexion of φρίκη and ἔρως, cf. Plato, Phaedrus 251a (discussed below); for the medicalised symptoms of emotion in this last passage, cf. Respublica 387bc (of the fear of death). For φρίκη, piloerection, and anger in animals cf. e.g. [Hesiod], Scutum 171, 390f.; [Theocritus], Idyllia 25.244f. Comparable is Dio Chrysostom, Oratio 32.65, where the patient is the centaur, Chiron, a figure capable of both animal piloerection and human emotion. Piloerection links fear and aggression as action-programme responses: cf. Burkert 2010, 46, 48; also Darwin 1889 [1998], 100f., 104f.; Frijda 1986, 140f., 144f.; Schurtz et al. 2012, 206. Plutarch, Aristeides 18.2: ἥ τε φάλαγξ ὄψιν ἔσχεν αἰφνιδίως ἑνὸς ζῴου θυµοειδοῦς πρὸς ἀλκὴν τρεποµένου καὶ φρίξαντος (‘suddenly the phalanx took on the appearance of a single angry beast, bristling in its own defence’). For the Homeric image, see Iliad 4.282, 7.62, 13.339; Odyssey 19.446 (and cf. Burkert 2010, 50); for the same image of crops in a field, cf. Iliad 23.599. Plutarch, Demosthenes 20.3; de Alexandri magni fortuna aut virtute 331F. The relation of the startle reflex to the emotion of fear is similarly in question at 964b22–29, where the φρίκη caused by being touched by another person is explained in terms of the fear aroused by what is sudden and unexpected. The link with vision is especially frequent in Plutarch’s Lives: see Alexander 74.6; Aratus 32.3 (on which cf. below); Cicero 49.2; Marius 44.9; Numa 10.6; cf. φρίσσω etc. + participle of verb of seeing, e.g., Aeschylus, Supplices 346; Prometheus Vinctus 695. For φρίκη as a reaction to loud, sudden, uncanny, or unexpected noises, cf. Cassius Dio 48.37.2 = ii.275 ed. Boissevain (cf. 36.49.2 = i.387 ed. Boissevain); Philostratus, Heroicus 748.14–17.

96

Douglas Cairns

tional response. Plutarch, for example, refers to the deep and horrific roar (φθογγῆς βαρείας καὶ βρόµου φρικώδους), the low and terrible tone, a mixture of bestial roaring and the clap of thunder (τὰ δὲ φθέγγεται βύθιόν τι καὶ δεινόν, ὠρυγῇ θηριώδει καὶ τραχύτητι βροντῆς µεµειγµένον), produced by the Parthians’ percussion instruments as they face the Romans in battle, commenting that the Parthians have clearly understood the impact of such sounds on the emotions and morale of their opponents.49 That this vivid, perceptual aspect of φρίκη is relevant even in response to second-hand narratives of events is powerfully illustrated by the story of the Jewish woman’s cannibal feast in Josephus (above).50 The sense of φρίκη etc. is also very often specified by the occurrence of the relevant terms in connexion with the divine and the supernatural. Many of these passages exhibit the connotations of immediacy and spontaneity that we have just discussed, and thus φρίκη appears as an instinctive response to sights and sounds of a numinous or supernatural nature (cf. note 49). This makes φρίκη an especially appropriate response to epiphany, presumed epiphany, quasi-epiphany, or other presumed signs of divine presence:51 thus a heavenly light occasions φρίκη 49

50

51

Crassus 23.8–9: εὖ πως συνεωρακότες ὅτι τῶν αἰσθητηρίων ἡ ἀκοὴ ταρακτικώτατόν ἐστι τῆς ψυχῆς καὶ τὰ περὶ ταύτην πάθη τάχιστα κινεῖ καὶ µάλιστα πάντων ἐξίστησι τὴν διάνοιαν. Cf. Euripides, Hippolytus 1201f. (sounds of supernatural origin), 1215f. (ditto); Andromache 1147f. (ditto); Aristophanes, Ranae 1335f. (ditto); Andocides, De mysteriis 29 (both sound and content, i.e. horrific tales of religious transgressions); [Aristotle], Mirabilium auscultationes 843a15f. (the mere sight of waves in the Straits of Messina); Apollonius Rhodius 4.1339–1342 (sound as sign of danger); Plutarch, Marius 19.1–20.3 = Poseidonius fr. 201 ed. Theiler (the groans and lamentations of their defeated opponents echo through the hills at night and terrify the Romans); Josephus, Bellum Iudaicum 4.286f. (thunder), 6.2 (the sight of piles of corpses), 6.83f. (the sight of one centurion’s prodigious massacre of the enemy); Antiquitates Iudaicae 19.344f. (the appearance of Agrippa as a quasi-epiphany); Cornutus, De natura deorum 11.12 (the appearance of the Eumenides); Plutarch, Marcellus 15.4 (the θέαµα φρικῶδες of a ship raised into the air by one of Archimedes’ defensive machines); Sulla 14.3 (the sound of trumpets and horns); Comparatio Lysandri et Sullae 2.4 (the sight of slaughter); De facie in orbe lunae 944B (the appearance of the moon’s ‘angry face’); Lucian, Philopseudes 22 (the Gorgon-like aspect of a female monster); Aristeides, Hieroi logoi 2, 297.20f. ed. Jebb (mysteries; cf. 3, 320.5; Eleusinios 256.24); Achilles Tatius 3.17.7 (the sight of Leucippe emerging, mutilated but alive, from her coffin); Pollux 4.85 (φρικώδης is a good epithet for the sound of the trumpet); Philostratus, Heroicus 682.9f. (the voice of Ajax from his tomb), 748.14–17 (divine singing); Julian, Panegyricus Constantii 25 (the sight of weapons in the eyes of the coward). In these φρίκη-causing sights and sounds the element of fear, or at least of the unnerving or uncanny, is prominent; Greek writers seem not to present φρίκη as a response to stirring or awe-inspiring sights or sounds (such as works of art or pieces of music) as such; contrast the subjects investigated by Schurtz et al. 2012, and cf. Keltner and Haidt 2003, 300f., 303f., 306f. On the importance of ὄψις and ἐνάργεια in the Josephus narrative and its affinities with Greek tragedy, see Chapman 1998, 51–106; cf. Chapman 2007. For φρίκη and the visualisation of events ‘in the mind’s eye’, cf. Cassander’s response to the statue of Alexander at Delphi, Plutarch, Alexander 74.6 (note 16 above). Cf. the shudders that respond to epiphany at Hesiod, fr. 165. 4f. edd. Merkelbach/West and to the divine sign from Zeus that marks Oedipus’ heroisation at Oedipus Coloneus 1606f., though in both these places the verb employed is ῥιγεῖν. For the ‘holy shudder’, cf. especially

A Short History of Shudders

97

before the divine, but confidence in the face of the enemy, in Cyrus and his men,52 and φρίκη is the reaction of the audience to the illusion of divine presence or possession created by the Sicilian statesman, Nicias, in Plutarch’s Life of Marcellus.53 The terms are also used of reactions to the supernatural communications believed to occur in dreams,54 or to a variety of miracles, portents, and omens – as when the people of Adranum greet Timoleon with reports of their own φρίκη and θαῦµα when, at the beginning of Timoleon’s battle against Hicetas of Leontini, the gates of their temple spontaneously flew open to reveal the cult-statue’s speartip trembling, sweat running down the god’s face.55 In some of these passages, and indeed in a number of others, φρίκη might be regarded simply as fear of divine power, anger, or punishment.56 These are the kinds of response deprecated by Plutarch in De superstitione, when he castigates people who shudder and tremble before those who are in fact their saviours.57 Yet it is also clear that awe and deference are often also in play. Thus, while (according to Cornutus) the physical appearance of the Eumenides is φρικώδης and they

52

53

54 55

56

57

Burkert 2010, 50–54; also Keltner and Haidt 2003, 298f., 308–310 on awe and religion. Only a very small number of respondents in the survey of Schurtz et al. 2012 refer their goose pimples to religious experiences (p. 209); but this may simply reflect the limited scope for profound religious experiences in the lives of typical US college students over the four-week period of the survey (as the authors note, p. 210). Xenophon, Cyropaedia 4.2.15: λέγεται φῶς τῷ Κύρῳ καὶ τῷ στρατεύµατι ἐκ τοῦ οὐρανοῦ προφανὲς γενέσθαι, ὥστε πᾶσι µὲν φρίκην ἐγγίγνεσθαι πρὸς τὸ θεῖον, θάρρος δὲ πρὸς τοὺς πολεµίους. Marcellus 20.8 = Posidonius fr. 93a Theiler = FGrHist 87 F 43 ed. Jacoby. Cf. Chaniotis, in this volume, pp. 70f. For φρίκη in the context of quasi- or assumed epiphany (i.e., when the appearance or behaviour of a mortal suggests or is assimilated to epiphany), cf. Josephus, Antiquitates Iudaicae 19.344f. (the quasi-epiphany of Agrippa in the theatre); Plutarch, De Alexandri magni fortuna aut virtute 343E (the appearance of Alexander as quasi-epiphany). See Josephus, Bellum Iudaicum 3.353; Plutarch, De superstitione 165F; Philostratus, Heroicus 666.6–8; Achilles Tatius 5.25.4. Plutarch, Timoleon 12.9: µετὰ φρίκης καὶ θαύµατος ἀπαγγέλλοντες, ὡς ἐνισταµένης τῆς µάχης οἱ µὲν ἱεροὶ τοῦ νεὼ πυλῶνες αὐτόµατοι διανοιχθεῖεν, ὀφθείη δὲ τοῦ θεοῦ τὸ µὲν δόρυ σειόµενον ἐκ τῆς αἰχµῆς ἄκρας, τὸ δὲ πρόσωπον ἱδρῶτι πολλῷ ῥεόµενον. Again, the basic phenomenon goes back to Homer (Iliad 12.208f., a physical shudder at the sight of an omen, though the verb there is ῥιγεῖν), but Plutarch proves especially rich in instances: see Aemilius 17.8 (eclipse); Agesilaus 24.5 (daylight as quasi-divine sign, associated with Eleusis); Sulla 11.1 (an omen that takes place in the theatre). For ‘wonder’ (θάµβος, θαῦµα, etc.) as a stock feature of epiphanies, often coupled with ‘fear’, see e.g. Homer, Iliad 3.398; Odyssey 1.322f., 3.372f., 16.178f., 19.36–40; Hymnus Homericus ad Apollinem 134f.; Hymnus Homericus ad Venerem 81–90, etc.; cf. Richardson 1974, 208f. on Hymnus Homericus ad Cererem 188–190; Faulkner 2008, 164 on Hymnus Homericus ad Venerem 83–90. For fear of divine power, anger, or unpleasant consequences as the principal focus of supernatural φρίκη, e.g. Aeschylus, Septem contra Thebas 720f.; Euripides, Cyclops 320; Plato, Respublica 387c; Callimachus, Aetia fr. 75.6 ed. Pfeiffer; Nicaenetus fr. 1. 6 ed. Powell; Cornutus, De natura deorum 74.2–4; Plutarch. Aemilius, 35.3 (cf. Otho 1.3: quasi-religious fear of Otho as a quasi-avenging deity); De virtute morali 450A; Philostratus, Vita Apollonii 3.35. See De superstitione, 165D–167A in general, especially166D, 166E; cf. Quomodo adulescens poetas audire debeat 26B – the student should not be afraid to criticise the works he studies, shuddering (φρίττειν) and prostrating himself like the superstitious in a temple.

98

Douglas Cairns

inspire fear, none the less their role in punishing the ἀσεβεῖς is salutary, and thus they truly merit the name of Semnai.58 According to Menander Rhetor, hymns to personifications (or ‘fictions’) which are divine require a style which is more σεµνός, by contrast with those to more ‘human’ personifications, that is, ὅσα οὐ παντάπασιν φρικώδη καὶ θεῖα, such as Penia (Poverty) and Agrypnia (Sleeplessness).59 Another late source, the emperor Julian, corroborates the association of φρίκη with positive respect rather than simple fear when he compares his feelings towards the gods with those which one might experience towards good masters, teachers, fathers, and protectors.60 It is in this sense of awe and respect that sacred places, such as temples and shrines, are said to attract φρίκη.61 This basic, instinctive response to the awesomeness of the numinous as such is attested (albeit with the verb ῥιγεῖν) as early as the shudder with which Ajax looks upon the works of the gods at Iliad 16.119f. This association with the expression of awe, humility, and deference before the divine links this kind of φρίκη with the Plutarchan passages already mentioned in which φρίσσω is a metonym for admiration.62 Similarly, it is humility and deference, rather than merely fear, which Hecuba desiderates in Helen when she expresses the opinion that, on account of her transgressions, she should go trembling with φρίκη, dressed in rags and with her head shaved, in order to manifest a due sense of shame and respect for others.63 It seems, then, that φρίκη is implicated in hierarchies of status and strategies of demeanour and deference in both the human and the divine spheres. In Greek terms, it seems to have as much to do with σέβας, a type of awe or respect that responds to legitimate status and authority, the salutary fear recommended in Aeschylus’ Eumenides,64 as it does with simple fear of danger. 58 59 60

61

62 63

64

Cornutus, De natura deorum 11.3–18. Menander Rhetor 342.15–19. Contra Heracleium 8.14–17: οὕτω δή τι τοὺς θεοὺς πέφρικα καὶ φιλῶ καὶ σέβω καὶ ἅζοµαι καὶ πάνθ’ ἁπλῶς τὰ τοιαῦτα πρὸς αὐτοὺς πάσχω, ὅσαπερ ἄν τις καὶ οἷα πρὸς ἀγαθοὺς δεσπότας, πρὸς διδασκάλους, πρὸς πατέρας, πρὸς κηδεµόνας. Cf. id., Epistulae 89b.169–175: as one who loves his father (etc.) likes looking at images of his father, so one who loves the gods likes looking at statues and pictures of the gods, σεβόµενος ἅµα καὶ φρίττων ἐξ ἀφανοῦς ὁρῶντας εἰς αὐτὸν τοὺς θεούς (‘in awe and φρίκη before the gods who, unseen, look upon him’). See Demosthenes, 23.74 (of the Delphinion, the court with jurisdiction over justifiable homicide, qua holy place); Josephus, Bellum Iudaicum 4.181f., 6.123; Plutarch, Tiberius et Gaius Gracchus 21.5; NB especially Pollux 1.23, where φρικῶδες appears after σεµνόν, ἔνθεον, and θεῖον in a list of appropriate epithets for temples. Demosthenes 20.3 and de Alexandri magni fortuna aut virtute 331F. See note 46. Euripides, Troades 1025–1028: ἣν χρῆν ταπεινὴν ἐν πέπλων ἐρειπίοις, | φρίκῃ τρέµουσαν, κρᾶτ’ ἀπεσκυθισµένην | ἐλθεῖν, τὸ σῶφρον τῆς ἀναιδείας πλέον | ἔχουσαν ἐπὶ τοῖς πρόσθεν ἡµαρτηµένοις. See Cairns 1993, 213f., and 137f., 157, 206–214, in general on σέβας. On awe as a social emotion, and especially on its positive aspects, see Keltner and Haidt 2003; Schurtz et al. 2012, 210–216. For Keltner and Haidt 2003, 306f. awe towards social superiors is the emotion’s ‘primordial’ form, the application to elicitors in the natural world, art, or music a secondary development. Schurtz et al.’s investigation of the physical symptom of ‘goose bumps’ likewise concentrates on social factors (for which their respondents did indeed provide much

A Short History of Shudders

99

It is at least partly in this guise that φρίκη is implicated in the institution of the oath, a ritual that publicly puts at stake the honour both of the human actors and its divine guarantors in an often elaborate and solemn ceremony involving prescribed roles and formulas. To be sure, in this context φρίκη remains, at bottom, an instinctive and involuntary emotional response, but its association with the oath reminds us that such responses are regularly embedded in highly structured and specific cultural practices. A large number of passages from the Imperial Period and later link φρίκη and the oath, the majority being simple references to ὅρκοι φρικώδεις,65 but the association between the oath and the physical reaction that φρίκη represents is as old as Greek literature itself, as we see in the case of Priam’s shudder (ῥίγησεν) in response to the request that he perform oathsacrifice at Iliad 3.259.66 To be sure, the oath entails frightening consequences in the event of its breach,67 but it is also an institution in which τιµή is deeply implicated – that of the god is invested in the solemnity of the ritual itself and that of the human participants is committed to its maintenance. Hence, though ὅρκον σέβεσθαι occurs only in the second of the Golden Verses of Pythagoras,68 a connexion between the oath and σέβας is regular in the frequent presentation of honouring one’s oaths as a requirement of εὐσέβεια,69 while the idea that the oath itself is an entity worthy of honour and respect is apparent from the recurrence of the phrase ὅρκον αἰδεῖσθαι.70 The same sense of awe and respect for the τιµή invested in ritual as such seems to lie behind Pelasgus’ shudder on seeing his city’s altar bedecked with the symbols of the Danaids’ supplication at Aeschylus, Supplices 346: the Danaids themselves relate this reaction to the prospect of Zeus’ anger at 347, but they have equally urged Pelasgus to show αἰδώς for the altar itself at 345; Pelasgus’ response lies somewhere between the αἰδώς that the suppliants demand and the fear that reinforces their claims.71

65

66 67 68 69 70 71

evidence). But the rootedness of such symptoms in evolutionarily old capacities that humans share with other species might suggest a different evolutionary hypothesis, less specifically focused on human social hierarchies. See e.g. Philo Iudaeus, De decalogo 141.3; Josephus, Vita 275; Bellum Iudaicum 2.139; [Clemens Romanus], Homiliae 5.5.2; Plutarch, Alexander 30.11; Arrian, fr. 94.2–3 ed. Jacoby (with stress on sanctions for perjury); Pollux 1.39; Cassius Dio 8.36.29 = i.107 ed. Boissevain; Porphyrius, De abstinentia 4.13; and so on into the writings of the Church Fathers and beyond (e.g. seven times in John Chrysostom, four times in Palladius, etc.) Cf. ῥιγίστη of the oath sworn by the Styx at Apollonius Rhodius 2.291f.; more remotely, ῥίγιστος of Zeus Hikesios, ibid. 2. 215. Cf. curses at Plutarch, Crassus 16.7; Timoleon 5.3; Philostratus, Vita sophistarum 2, p. 599. 9–11; magic spells: Lucian, Philopseudes 31. Iamblichus, De vita Pythagorica 28.144. Sophocles, Trachiniae 1222f.; Euripides, Medea 754f.; Hippolytus 656f., 1308f.; Iphigenia Taurica 743; Antiphon 6.33, etc. See Aeschylus, Eumenides 483f., 680, 710; Sophocles, Oedipus Tyrannus 647; Appian, Prooemium 23.2; Achilles Tatius 5.26.4; cf. Cairns 1993, 209f. For αἰδώς and σέβας as related ways of responding to the τιµή of sacred objects, cf. the αἰδώς for the altar demanded at Supplices 345 (cf. Choephori 106–108) with the σέβας for the same object recommended by Danaus at 222f.

100

Douglas Cairns

Probably the most elaborate of ancient Greek ritual processes, in terms of the way the emotions of the participants are aroused, manipulated, and incorporated in a highly structured series of individual and collective experiences, is that of the mystery religions. Here we have a tantalising fifth-century glimpse of the relation between φρίκη and the mysteries in an exiguous fragment of an unknown play by Aeschylus: ἔφριξ’· ἔρως δὲ τοῦδε µυστικοῦ τέλους.72 Later, in 400 or 399 BCE, Andocides uses the adjective φρικώδης with reference to his accusers’ allegations, where the adjective may encompass both the sanctity of the rites allegedly profaned and the horror of the alleged transgression.73 But the locus classicus on the phenomenology of the initiand’s experience is the celebrated fragment of Plutarch’s On the Soul, fr. 178 ed. Sandbach, in which the immortal soul’s experience of death is compared extensively to that of initiation into the Eleusinian Mysteries, a process, which itself no doubt mirrors or represents the journey of the initiate’s soul after death. Here, φρίκη belongs to the stage prior to initiation, in which the initiand is subjected to the darkness and disorientation that are preliminary to the illumination to follow:74 At first there are wanderings, wearisome running around, and inconclusive, anxious journeys through the darkness; then, before the consummation itself, all the terrors – φρίκη, trembling, sweating, and amazement (θάµβος). But after that one encounters a kind of miraculous light, and is welcomed by pure, open places and meadows, with voices and dancing and the awe75 inspiring majesty of sacred sounds and holy visions ...

Given the ritual’s highly dramatised and stage-managed construction of the initiand’s journey from darkness to light, involving special effects, costumes, and props, there can be no doubt that the arousal of emotions such as φρίκη, spontaneous and authentic as they may be in each individual’s case, is a deliberate aim of the initiatory process as social drama. As a response to the sanctity, solemnity, and power of the ritual, the link between φρίκη and the mysteries is attested throughout antiquity and beyond.76 72

73 74

75 76

Aeschylus fr. 387 ed. Radt = Σ Sophocles, Oedipus Coloneus 1049. On the text (Radt prints Dindorf’s ἐρῶ δὲ) and possible context (including a possible relation between φρίκη and ?ἔρως here and ἔφριξ’ ἔρωτι at Sophocles, Ajax 693), see Radt 1985, 433f. ad loc. Andocides, De mysteriis 29. Cf. Josephus, Bellum Iudaicum 5.378 (φρίκη at talking about the divine in the hearing of non-believers). Πλάναι τὰ πρῶτα καὶ περιδροµαὶ κοπώδεις καὶ διὰ σκότους τινὲς ὕποπτοι πορεῖαι καὶ ἀτέλεστοι, εἶτα πρὸ τοῦ τέλους αὐτοῦ τὰ δεινὰ πάντα, φρίκη καὶ τρόµος καὶ ἱδρὼς καὶ θάµβος· ἐκ δὲ τούτου φῶς τι θαυµάσιον ἀπήντησεν καὶ τόποι καθαροὶ καὶ λειµῶνες ἐδέξαντο, φωνὰς καὶ χορείας καὶ σεµνότητας ἀκουσµάτων ἱερῶν καὶ φασµάτων ἁγίων ἔχοντες [κτλ.]. On the emotional aspects of the ritual here, cf. Clinton 1992, 85f.; cf. Seaford 1996, 39, 201. E.g., Demetrius, De elocutione 101 (καὶ τὰ µυστήρια ἐν ἀλληγορίαις λέγεται πρὸς ἔκπληξιν καὶ φρίκην, ὥσπερ ἐν σκότῳ καὶ νυκτί – ‘the mysteries too are spoken of allegorically, to arouse ἔκπληξις and φρίκη, e.g. “in darkness and night”’); cf. Josephus, Bellum Iudaicum 2.133; Lucian, Iuppiter tragoedus 30; Aristeides, Hieroi logoi 2, 297.20f. ed. Jebb (cf. 256.24, 320.5). Φρικώδης etc. are frequently used in Christian writers’ representations of Christian dogma and practice, especially the sacrament, as mysteries (e.g., φρίκη twice, φρίσσω three times, φρικώδης nineteen times, φρικτός 59 times in John Chrysostom).

A Short History of Shudders

101

Many of the associations of φρίκη discussed above are evoked in a splendid passage of Plato’s Phaedrus (251a). In a context in which the passion of ἔρως is assimilated to the philosopher’s pursuit of metaphysical knowledge, and in which the cycles of reincarnation that allow us, if we are lucky, to retain dim memories of the true world of Forms, are presented in terms of mystic initiation, φρίκη appears as at once a symptom of ἔρως (as at Sophocles, Ajax 693), as a response to divine epiphany, as an experience of the mystic initiate, and as an emotion akin to both fear and σέβας:77 But when the recent initiate, one who saw much of what was visible then, sees a god-like face or some bodily form that is a good imitation of beauty, at first he shudders, and something of the fear he experienced then comes over him; then he gazes upon it and reveres it as if it were a god, and if he were not afraid of a reputation for utter madness, he’d offer sacrifice to his beloved as one would to a statue of a god.

Φρίκη, then, can be a spontaneous, involuntary, and automatic physical reaction common to emotional and other πάθη. It very often appears as an immediate response to direct visual and auditory stimuli. It is particularly associated with extreme sights, sounds, and situations, and especially with the numinous. But it is also capable of being manipulated and manufactured in highly structured contexts, when it is enlisted as an element in elaborate social and ritual dramas. This makes φρίκη highly appropriate as a spontaneous response to the visual representation of (often divinely inspired) extremes of suffering on the tragic stage, a function with which it is credited by both Gorgias,78 in an early but clearly influential account of the emotional effects of poetry, and Aristotle in the Poetics.79 In the latter passage, Aristotle takes for granted that the visual representation of others’ suffering can arouse φρίκη, but opines that it is a mark of superior poetic skill to do so via the plot itself, so that one who merely heard rather than witnessed the events 77

78

79

Ὁ δὲ ἀρτιτελής, ὁ τῶν τότε πολυθεάµων, ὅταν θεοειδὲς πρόσωπον ἴδῃ κάλλος εὖ µεµιµηµένον ἤ τινα σώµατος ἰδέαν, πρῶτον µὲν ἔφριξε καί τι τῶν τότε ὑπῆλθεν αὐτὸν δειµάτων, εἶτα προσορῶν ὡς θεὸν σέβεται, καὶ εἰ µὴ ἐδεδίει τὴν τῆς σφόδρα µανίας δόξαν, θύοι ἂν ὡς ἀγάλµατι καὶ θεῷ τοῖς παιδικοῖς. See Helen (B 11 edd. Diels-Kranz) 8: τὴν ποίησιν ἅπασαν καὶ νοµίζω καὶ ὀνοµάζω λόγον ἔχοντα µέτρον· ἧς τοὺς ἀκούοντας εἰσῆλθε καὶ φρίκη περίφοβος καὶ ἔλεος πολύδακρυς καὶ πόθος φιλοπενθής, ἐπ’ ἀλλοτρίων τε πραγµάτων καὶ σωµάτων εὐτυχίαις καὶ δυσπραγίαις ἴδιόν τι πάθηµα διὰ τῶν λόγων ἔπαθεν ἡ ψυχή (‘all poetry I regard and describe as speech with metre. Into those who listen to it comes a fearful shuddering and a tearful pity and a longing that loves to lament, and at the success and failure of others’ affairs and persons the soul undergoes, through words, a certain experience of its own’). On this passage, see Heath 1987, 7f. Cf. e.g. the reaction of Plato’s Ion to his own representation of terrible events (ὅταν τε φοβερὸν ἢ δεινόν [sc. λέγω], ὀρθαὶ αἱ τρίχες ἵστανται ὑπὸ φόβου, ‘whenever I tell a frightening or terrible tale, my hair stands on end out of fear’, Ion 535c), a reaction that he claims to be able to arouse in others too (535e). Poetica 14, 1453b3–7: δεῖ γὰρ καὶ ἄνευ τοῦ ὁρᾶν οὕτω συνεστάναι τὸν µῦθον ὥστε τὸν ἀκούοντα τὰ πράγµατα γινόµενα καὶ φρίττειν καὶ ἐλεεῖν ἐκ τῶν συµβαινόντων· ἅπερ ἂν πάθοι τις ἀκούων τὸν τοῦ Οἰδίπου µῦθον (‘for the plot ought to be so composed that, even without seeing a performance, one who merely hears what happens will shudder and feel pity as a result of the events – as indeed one would on hearing the plot of the Oedipus’).

102

Douglas Cairns

dramatised in the play would still experience pity and fear; and he gives Sophocles’ Oedipus Tyrannus as an example of a tragedy that fulfils this requirement. In the Oedipus Tyrannus itself, however, though the play’s sole instance of a term from the φρίκη family does come immediately after the περιπέτεια in which Oedipus’ recognition of his true status instantiates the profound change from good fortune to bad of which the chorus sing in the fourth stasimon, it is none the less also powerfully expressive of the chorus’ instinctive horror at the re-appearance of the now blind and horribly mutilated Oedipus (his self-mutilation represented visually by the new mask he will have donned during the choral ode). Here, the chorus’ φρίκη represents a visceral revulsion towards both the spectacle that the blind Oedipus presents and the horror of his downfall:80 What suffering, terrible for humans to see, most terrible of all that I have ever encountered! What madness came upon you, wretched one? What divine being was it that leapt further than the longest leap on top of your unhappy fate? Alas, poor man: I cannot even look at you, though there is much I want to ask, much to hear, and much to look at; such is the φρίκη you cause in me.

This passage brings us full circle, back to the passage from Plutarch’s Life of Aratus with which we began. There are a number of similarities between the two contexts. First there is the visual element: in both passages it is above all an object of sight that excites the φρίκη of those who witness it.81 Then there is the suspicion (or rather, in the Oedipus Tyrannus at least, the certainty) that events that appear to happen by chance in fact betray some kind of divine purpose.82 Hence both the 80

81

82

Oedipus Tyrannus 1297–1306: ὦ δεινὸν ἰδεῖν πάθος ἀνθρώποις, | ὦ δεινότατον πάντων ὅσ’ ἐγὼ | προσέκυρσ’ ἤδη. τίς σ’, ὦ τλῆµον, | προσέβη µανία; τίς ὁ πηδήσας | µείζονα δαίµων τῶν µηκίστων | πρὸς σῇ δυσδαίµονι µοίρᾳ; | φεῦ φεῦ δύστην’, ἀλλ’ οὐδ’ ἐσιδεῖν | δύναµαί σ’, ἐθέλων πόλλ’ ἀνερέσθαι, | πολλὰ πυθέσθαι, πολλὰ δ’ ἀθρῆσαι· | τοίαν φρίκην παρέχεις µοι. Cf. Aratus 32.2: τοῖς πολεµίοις φάσµα θεῖον ὁρᾶν δοκοῦσι φρίκην ἐνέβαλε καὶ θάµβος (‘she caused φρίκη and θάµβος in the enemy, who imagined they were seeing a divine epiphany’). Cf. the emphasis on the visual not only in the Achaeans’ response to the daughter of Epigethes (32.2: αὐτοῖς τε τοῖς πολίταις θέαµα σεµνότερον ἢ κατ’ ἄνθρωπον ἐφάνη, ‘to the citizens themselves she seemed a sight of more than human majesty’), but also in the alternative version in which it is the sight of the cult-image of the goddess that repels the Aetolians (32.3: µηδένα προσβλέπειν ἐναντίον, ‘no one looks directly at it;’ ὅραµα φρικτὸν εἶναι καὶ χαλεπόν, ‘a φρίκη-inspiring and painful sight;’ 32.4: ἀντιπρόσωπον, ‘face to face’). The Oedipus Tyrannus’ references to τύχη are virtually all ironic, serving to confirm that explanations in these terms, rather than in terms of divine purpose, are mistaken: see 977f., 1080f.; cf. 78–81, 87f., 145, 262f., 442, 773, 776, 948f., 998, 1036, 1478f.; cf. Eidinow 2011, 57–59. In the same way, the play makes thematic capital by exploiting the many and manifest coincidences of its plot in such a way that they appear to betray the action of Apollo: see above all Cameron 1968, 67f., 73–78; Peradotto 1992, 3–10. In this, the play conforms to Aristotle’s observation that chance events are more striking when they appear to happen on purpose (Poetica 9, 1452a6–10: καὶ τῶν ἀπὸ τύχης ταῦτα θαυµασιώτατα δοκεῖ ὅσα ὥσπερ ἐπίτηδες φαίνεται γεγονέναι, οἷον ὡς ὁ ἀνδριὰς ὁ τοῦ Μίτυος ἐν Ἄργει ἀπέκτεινεν τὸν αἴτιον τοῦ θανάτου τῷ Μίτυι, θεωροῦντι ἐµπεσών· ἔοικε γὰρ τὰ τοιαῦτα οὐκ εἰκῇ γίνεσθαι – ‘even of things that happen by chance those that seem to come about on purpose

A Short History of Shudders

103

Aetolians and the Achaeans in Plutarch treat the appearance of the daughter of Epigethes as a kind of epiphany,83 just as Sophocles’ chorus want to know the identity of the δαίµων who leapt upon Oedipus’ µοῖρα, a µοῖρα that was already characterised by a δαίµων’s malign intentions (1300–1302). Both as an instinctive response to a sudden and unexpected visual stimulus and as a reaction typical of those who sense that the influence of the gods is at work, then, φρίκη is eminently at home in both these passages. The particular associations of φρίκη make this emotion especially suited to the creation of that ἐνάργεια that Angelos Chaniotis has correctly diagnosed as integral to the theatricality of this and similar passages. The essence of tragic emotion is sympathy, the ability to feel the pity that, Aristotle tells us in the Rhetoric, depends on a sense of the vulnerability that we share with those who are suffering,84 and the fear, which (in both Poetics and Rhetoric) relies on the sense that such things might also happen to us.85 In the Oedipus Tyrannus, the pity that the chorus and others feel, despite their revulsion, for Oedipus, and especially their presentation of him (in the fourth stasimon) as a paradigm of the shared human vulnerability on which pity rests help to guide and condition an audience’s response;86 the same seems to me to be true of their φρίκη in the passage quoted

83

84 85

86

appear more remarkable, such as when the statue of Mitys at Argos fell upon the man who was responsible for Mitys’ death and killed him while he was looking at it; for such things do not seem to happen randomly’); the same principle seems to underlie the presentation of τύχη also in the Aratus passage. On τύχη in Plutarch, especially the Lives, see Swain 1989. It is puzzling that the assumed epiphany is of Artemis, when the main ‘prop’ which contributes to its effect is the three-crested helmet (τριλοφία), an attribute exclusively associated with Athena. The prototype is Phidias’ Athena Parthenos at Athens, though the model is employed also by late fifth-century Syracusan coinage in a type that either appropriates the Athenian depiction of the goddess (Holloway 1991, 131) or adapts it to that of Arethusa (Seltman 1955, 126); see Rutter 2012, 79f. For the coins, see Tudeer 1913, no. 58, R36; Sylloge Numorum Graecorum. The Collection of the American Numismatic Society nos. 308– 313; cf. among literary sources Eratosthenes, Catasterismi 1.13.21 ed. Olivieri. Polyaenus’ version (Strategemata 8.59.1) makes the expected association between the τριλοφία and Athena; perhaps in collating different sources Plutarch (or a predecessor) conflated a version involving an epiphany of Athena (as in Polyaenus) with a version involving the cult of Artemis – e.g. that attributed to the Pellenians at Aratus 32.3, if that version did in fact refer to a cult-image of Artemis. But it is also possible that the passage’s sole occurrence of the goddess’s name at 32.1 is an error (authorial or scribal), Artemis for Athena (and thus τῆς θεοῦ in the Pellenians’ version at 32.3 would also be a reference to Athena). For Porter 1937, 70, it is Polyaenus who gets it wrong. Aristotle, Rhetorica 2.8, 1385b13–33, 1386a25–29. Aristotle, Poetica 13, 1453a4–6; cf. Rhetorica 1386b27–29, where the things we pity in others are said to be the kind of things that we fear may happen to ourselves. For the shared vulnerability to vicissitude as a condition for pity in traditional Greek ethics, cf. Iliad 24.485– 512; Bacchylides 5.155–162 (especially 160–162 and cf. 89–92); Sophocles, Ajax 121–126; Philoctetes 501–506; Oedipus Coloneus 566–568; Euripides, Hecuba 282–287; Herodotus 1.86.6, 7.46.2; cf. Pelling 2005, 289, 291f. on Plutarch. On the role of fear for oneself in Aristotle’s conception of pity see Konstan 2001, 130–136 with Cairns 2004, 66f. Pity: 1194, 1211, 1216–1221, 1286, 1296, 1299, 1303, 1347; revulsion: especially the chorus at 1217f., 1297–1299, 1303–1306, 1348, all, significantly, associated in context with their

104

Douglas Cairns

above – the horror at Oedipus’ suffering is a prerequisite for the recognition that his suffering differs in degree but not in kind from that which might befall any of us. It is interesting, therefore, that φρίκη can occasionally be used of the kind of sympathetic yet still fearful response that is central to tragic aesthetics in the Poetics and to the Rhetoric’s concept of pity. In particular I have in mind the koryphaios’ comment on Heracles’ sufferings at Sophocles, Trachiniae 1044f.,87 the chorus’ on those of Io at Prometheus Vinctus 695,88 and two passages in Plutarch’s Life of Aemilius Paullus in which φρίκη appears to be conditioned by reflection on the vulnerability of all human beings to vicissitude.89 In all of these cases the patients of φρίκη respond to others’ misfortunes both as objects of fear and as grounds for sympathy (or at least for reflection on the liability of all human beings to suffer in similar ways). CONCLUSION In conclusion, the phenomenon that the Greeks called φρίκη is clearly not a wholly culture-specific one; its roots, as an involuntary movement, antedate the

87 88 89

pity; cf. Creon at 1424–1431. On the ‘hermeneutic’ function of the choral voice in the fourth stasimon, see e.g. Calame 1999, 139; for the same general phenomenon (internal audience response guiding external) in Plutarch, see Pelling 2005, 282f. Κλύουσ’ ἔφριξα τάσδε συµφοράς, φίλαι, | ἄνακτος, οἵαις οἷος ὢν ἐλαύνεται (‘I shudder as I hear of our king’s sufferings, friends; what terrible afflictions for a man like him’). Πέφρικ’ εἰσιδοῦσα πρᾶξιν Ἰοῦς (‘I shudder as I contemplate the state of Io’s fortunes’). Plutarch, Aemilius 29.5: booty from the sack of the cities of Epirus produces no more than eleven drachmas per solider, so that ‘everyone shuddered at the outcome of the war, that the division of an entire nation’s wealth should yield so little profit and gain for each individual’ (φρῖξαι δὲ πάντας ἀνθρώπους τὸ τοῦ πολέµου τέλος, εἰς µικρὸν οὕτω τὸ καθ’ ἕκαστον λῆµµα καὶ κέρδος ἔθνους ὅλου κατακερµατισθέντος). Cf. 35.3: the untimely death of two of Aemilius’ sons, one immediately before and the other immediately after his triumph, leads the Romans to ‘shudder at the cruelty of Fortune, that she did not scruple to introduce so much sorrow into a household so admired, so full of joy and sacrifices, or to mix laments and tears together with victory paeans and triumphs’ (φρῖξαι τὴν ὠµότητα τῆς τύχης ἅπαντας, ὡς οὐκ ᾐδέσατο πένθος τοσοῦτον εἰς οἰκίαν ζήλου καὶ χαρᾶς καὶ θυσιῶν γέµουσαν εἰσάγουσα, καὶ καταµειγνύουσα θρήνους καὶ δάκρυα παιᾶσιν ἐπινικίοις καὶ θριάµβοις). Cf. Pelling 2005, 209. On some of the issues raised by the appearance of quasi-tragic narrative patterns in Plutarch and his predecessors among the historians, cf. Pelling 2002, passim, especially 97f., 101f., 106, 111 n. 27, 117–141 (especially 120f., 130f.), 197–206, 239, 248, 355, 358, 387; Pelling 2005, 280–283; Chaniotis, pp. 80–82 in this volume). Cf. also Lateiner 2009 on the presentation of tears in historiography and in Plutarch’s Lives. In this connexion, it is perhaps worth noting that among pre-Christian (non-medical) authors Plutarch is by far the most prolific user of φρίκη-words (φρίκη 143 times, φρίσσω 27 times, φρικώδης sixteen times). To an extent, this reflects chronological semantic change, as words which were used sparingly and for particular effect in Classical poetry and as technical terms in the Hippocratic and Aristotelian corpora become common in post-Classical prose. At the same time, however, the qualities of φρίκη, as I have outlined them above, do chime very well with Plutarch’s predilection for vivid narrative, dramatic changes of fortune, and moralising on the ways in which his subjects’ lives exemplify recurrent human types and patterns.

A Short History of Shudders

105

origin of the human species; as a symptom of emotion, it belongs with basic responses that are rooted in the physiology of our and other species; and the name of the symptom becomes a name for a particular kind of emotional attitude by means of a mechanism, namely metonymy, that is ubiquitous in the formation of human concepts, especially emotional concepts. Even as a metonymy, however, φρίκη retains many of its fundamental connotations as an automatic, involuntary, instinctive reaction – indeed it is precisely in order to retain connotations of that sort, to conjure up something of the experience of emotion rather than merely labelling it, that one would choose to say φρίσσω rather than φοβοῦµαι.90 That being so, φρίκη the emotion emerges as a particular kind of fear – experientially structured, immediate, instinctive, and much more regularly occurrent than dispositional. The particular kind of fear that it is makes its use especially appropriate in specific sorts of scenario. In many cases, the specifics of these scenarios depend in turn on the nature and development of Greek norms and values, especially where humans’ relations with the divine are concerned, but also in connexion with (for example) the norms and conditions of pity and with the nature of audiences’ responses to poetry, drama, historiography, and biography. This is not just a matter of a single and simple emotion having a range of different elicitors; the specific history of φρίκη as a concept has to take account both of the ways that it is deeply enmeshed in elaborate ritual processes, such as mystic initiation and the swearing of oaths, and of the particular associations of the term that make it especially appropriate both for vivid literary representation and as an element in the aesthetics of literary appreciation. As ever, it is only once one has established the extent to which the emotions of another culture resemble those of one’s own that one can begin to pursue the interesting question of how they differ. BIBLIOGRAPHY Balcombe, J. (2010) Second Nature: The Inner Lives of Animals, New York. Berrettoni, P. (1970) Il lessico tecnico del I e III libro delle epidemie ippocratiche, Annali della Scuola Normale Superiore di Pisa. Lettere, storia e filosofia, ser. 2.39, 27–106, 217–311. Borgeaud, P. (1979) Recherches sur le dieu Pan, Rome. Bouvier, D. (2011) Du frisson (phrikê) d’horreur au frisson poétique: interpretation de quelques emotions entre larmes chauds et sueurs froides chez Platon et Homère, Mètis n.s. 9, 15–35. Burkert, W. (2010) Horror Stories: Zur Begegnung von Biologie, Philologie und Religion, in A. Bierl and W. Braungart (eds.), Gewalt und Opfer: Im Dialog mit Walter Burkert, Berlin, 45– 56. Cairns, D. L. (1993) Aidôs: The Psychology and Ethics of Honour and Shame in Ancient Greek Literature, Oxford. ––– (2004) Pity in the Classical World (review article on Konstan 2001), Hermathena 176, 59–74. Calame, C. (1999) Performative Aspects of the Choral Voice in Greek Tragedy: Civic Identity in Performance, in S. Goldhill and R. Osborne (eds.), Performance Culture and Athenian Democracy, Cambridge, 125–153. 90

Cf. Burkert 2010, 54: ‘Die biologisch-körperliche Grundlage geht in den Metaphern der Texte kaum verloren.’

106

Douglas Cairns

Cameron, A. (1968) The Identity of Oedipus the King: Five Essays on the Oedipus Tyrannus, New York. Chantraine, P. (1968–80), Dictionnaire étymologique de la langue grecque: histoire des mots, Paris. Chapman, H. H. (1998) Spectacle and Theater in Josephus’s Bellum Judaicum, unpublished PhD thesis, Stanford University. ––– (2007) Josephus and the Cannibalism of Mary (BJ 6. 199-219), in J. Marincola (ed.), A Companion to Greek and Roman Historiography, Malden MA, 419–426. Clinton, K. (1992) Myth and Cult: The Iconography of the Eleusinian Mysteries, Stockholm. Damasio, A. R. (1995) Descartes’ Error: Emotion, Reason, and the Human Brain, London. Darwin, C. (1889) [1998] The Expression of the Emotions in Man and Animals, third edition, edited by P. Ekman, Oxford. Eibl-Eibesfeldt, I. (1989) Human Ethology, New York. Eidinow, E. (2011) Luck, Fate, and Fortune: Antiquity and its Legacy, London. Faulkner, A. (ed.) (2008) The Homeric Hymn to Aphrodite, Oxford. Frijda, N. H. (1986) The Emotions, Cambridge. Griffiths, P. (1997) What Emotions Really Are: The Problem of Psychological Categories, Chicago. Gross, D. M. (2006) The Secret History of Emotion: From Aristotle’s Rhetoric to Modern Brain Science, Chicago. Heath, M. (1987) The Poetics of Greek Tragedy, London. Holloway, R. R. (1991) The Archaeology of Ancient Sicily, London. Ideler, J. L. (1841) Physici et medici Graeci minores. Vol. I, Berlin. Keltner, D. and J. Haidt (2003) Approaching Awe, a Moral, Spiritual, and Aesthetic Emotion, Cognition and Emotion 17.2, 297–314. Kövecses, Z. (2000) Metaphor and Emotion, Cambridge. Konstan, D. (2001) Pity Transformed, London. Lakoff, G. (1987) Women, Fire, and Dangerous Things: What Categories Reveal about the Mind, Chicago. Lakoff, G. and M. Johnson (1980) Metaphors We Live By, Chicago. Lateiner, D. A. (2009) Tears and Crying in Hellenic Historiography, in T. Fögen (ed.), Tears in the Greco-Roman World, Berlin, 105–134. Op de Hipt, D. (1972) Adjektive auf - ωδης im Corpus Hippocraticum, Hamburg. Pelling, C. (2002) Plutarch and History, London/Swansea. ––– (2005) Pity in Plutarch, in R. H. Sternberg (ed.), Pity and Power in Ancient Athens, Cambridge, 277–312. Peradotto, J. (1992) The Ideological Mapping of Oedipus Tyrannus, Transactions of the American Philological Association 122, 1–15. Porter, W. H. (ed.) (1937) Plutarch’s Life of Aratus, Dublin and Cork. Radt, S. L. (ed.) (1985) Tragicorum Graecorum Fragmenta. Vol. III. Aeschylus, Göttingen. Richardson, N. J. (ed.) (1974) The Homeric Hymn to Demeter, Oxford. Rutter, N. K. (2012) Artistic Identity: The Case of the ‘Signing Artists’ in Sicily, Quaderni Ticinesi: Numismatica e Antichità Classiche 41, 71–89. Schurtz, D. R., S. Blincoe, R. H. Smith, C. A. J. Powell, D. J. Y. Combs, and S. H. Kim, (2012) Exploring the Social Aspects of Goose Bumps and their Role in Awe and Envy, in Motivation and Emotion 36.2, 205-212. Seaford, R. A. S. (1996) (ed.) Euripides: Bacchae, Warminster. Seltman, C. T. (1955) Greek Coins, London (second edition). Sicurus, D. (1862) Theophili et Stephani Atheniensis de febrium differentia ex Hippocrate et Galeno, Florence. Swain, S. (1989) Plutarch: Chance, Providence, History, American Journal of Philology 110, 272– 302.

A Short History of Shudders

107

Theodoropoulou, M. (2012) The Emotion Seeks to be Expressed: Thoughts from a Linguist’s Point of View, in A. Chaniotis (ed.), Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, Stuttgart, 450–464. Tudeer, L. O. T. (1913) Die Tetradrachmenprägung von Syrakus in der Periode der signierenden Kunstler, Berlin (= Zeitschrift für Numismatik 30, 1–292). Wankel, H. (ed.) (1976) Demosthenes, Rede für Ktesiphon über den Kranz, 2 vols., Heidelberg. Williams, L. E. and J. A. Bargh (2008) Experiencing Physical Warmth Promotes Interpersonal Warmth, Science 322.5901, 606–607. Zaborowski, R. (2002) La crainte et le courage dans l’Iliade et l’Odyssée, Warsaw. Zhong, C.-B. and G. J. Leonardelli (2008) Cold and Lonely: Does Social Exclusion Literally Feel Cold?, Psychological Science 19.9, 838–842. Zink, N. (1962), Griechische Ausdrucksweisen für warm und kalt im seelischen Bereich, Diss. Mainz/Heidelberg.

REFLECTIONS ON THE DISCOURSE OF FEAR IN GREEK SOURCES Maria Patera ἀµορφότατος τὴν ὄψιν εἰµὶ γὰρ Φόβος, πάντων ἐλάχιστον τοῦ καλοῦ µετέχων θεός. Adespota F 873 eds. Kassel/Austin.

1 INTRODUCTION Greek sources mentioning fear are extremely numerous, and terms referring to fear are very diverse.1 Just as in French, this emotion is signified by several words with various connotations: effroi (dread), épouvante (terror), frayeur (fright), horreur (horror), peur (fear), panique (panic), terreur (terror), tremblement (trembling), alarme (alarm), angoisse (anguish), anxiété (anxiety), appréhension (apprehension), crainte (fear), effarouchement (scare), frissonnement (quivering), hantise (obsessive fear), inquiétude (disquiet).2 Therefore the objective here will not be to conduct a lexical study, notwithstanding the immense interest inherent in, and certain utility of, such research.3 Rather, the aim will be to study a certain discourse of fear that, when provoked, is associated with manipulation of the masses and with the deception employed to this end. Another aspect to be examined will be the response that sources recommend against the experience of fear, namely the appeal to reason or shame. This type of thinking appears often in Greek sources, independent of their date or the religious affiliation of their authors. Thus, this study will not be limited to a given time period nor to the usage of a specific vocabulary. We will examine texts that designate fear with various terms, as well as sources from many periods and genres, for it seems that a certain usage of fear persists through several centuries of Greek thought. In view of the number of texts with similar content, we must make a selection: the texts examined are examples among others. 1 2

3

Translated from the French by Michael Anthony Fowler. On the complexity of the categories of fear and on the translation problems that terms from one language pose for another, cf. Borgeaud 2007, 190f., note 4. If we add that an emotion need not always be clearly designated in order to be recognised as such, Greek passages evoking fear become innumerable. For the very complex vocabulary of fear, consult, e.g., the study of Homeric epic by Zaborowski 2002, 71–248.

110

Maria Patera

2 PHOBOS KENOS AND ITS USAGE The Stoic work entitled On the passions defines fear as ‘irrational flinching or flight in the face of an expected evil’.4 There are thirteen types of fear, including shame, anguish, cowardice, superstition, etc., and, above all, ψοφοδέεια, the groundless fear that a poltroon feels.5 It is therefore evident that fear is multiple, has several aspects, and is expressed in myriad terms. It can be collective, like panic, or individual; fear can be religious (fear before the transcendent divine) or superstitious (defined as φόβος τοῦ δαιµονίου or as exaggeration of the honour due to the gods); ‘rational’ in the face of ‘real’ danger and ‘irrational’ in the face of ‘imaginary’ danger (insofar as many philosophers and Christian thinkers considered fear of death as ‘irrational’).6 The sort of irrational fear defined as φόβος κενός, ‘empty, futile, pointless fear’, is called ψοφοδέεια in the Stoics and µορµοί in the plural in Hesychios.7 Equivalent to φόβος κενός is the proverbial expression ἀδεὲς δέος, which applies to those who fear in vain (ἐπὶ τῶν µάτην δεδιότων).8 Φόβος κενός is a topos encountered very often in sources that refer to childish fears treated as irrational when found in adults. The paradox is that, according to the same sources, this much-mocked, groundless fear is of some use: with it nannies make recalcitrant children reasonable and statesmen impose themselves on the people. Sources therefore mention provoked fear and experienced fear, and the ways of using the former and responding to the latter. In this way, Strabo explains the marvellous legends of Greek mythology with the statesmen’s need to impress the people: the founders of states accepted certain things (thunderbolt, aegis, trident, etc.) as bogeymen against simple – literally 4

5 6

7 8

[Andronikos], On the passions 1.2, ed. Glibert-Thirry: φόβος δὲ ἄλογος ἔκκλισις· ἢ φυγὴ ἀπὸ προσδοκωµένου δεινοῦ. This work was erroneously attributed to Andronikos of Rhodes, the head of the Peripatetic school of Athens in the middle of the first century BCE. PseudoPlato, Definitions (415e6 and 8, ed. Souilhé), inspired by Platonic, Aristotelian, and Stoic philosophy, defines fear (φόβος) as ‘the fright of the soul in the expectation of evil’ (ἔκπληξις ψυχῆς ἐπὶ κακοῦ προσδοκίᾳ) and fright (ἔκπληξις) as ‘fear in the expectation of an evil’ (φόβος ἐπὶ προσδοκίᾳ κακοῦ). [Andronikos], On the passions 3.1–13, ed. Glibert-Thirry (ψοφοδέεια is mentioned in 3.8). The terms ‘rational’ and ‘irrational’ are useful here in the sense of ‘does or does not conform to the ever-prevalent view of reason’, and not in the sense of ‘cognitive’ or ‘non-cognitive’. On this point, cf. Nussbaum 1996, 320 n. 4. Furthermore, the notion of ‘reason’ is itself mutable in the writings of ancient philosophers: cf. Sorabji 1996, 330–334. [Andronikos], On the passions 3.8, ed. Glibert-Thirry. Hesychios, s.v. µορµοί. Diogenianus, 1.16.1, eds. Leutsch/Schneidewin. Hence, the terms φόβος and δέος may be equivalent, even though dictionary definitions oppose them in defining φόβος as a sudden fear and δέος as a rational fear. This distinction was also advocated by Ammonius, De adfinium vocabulorum differentia 128, s.v. δέος καὶ φόβος, ed. Nickau, and undoubtedly defined by the Sophists, notably by Prodikos (cf. Plato, Protagoras 358d-e). On this point, cf. also below p. 122 n. 77. In Homeric epic, however, the term δέος designates a complex fear often associated with other types of fear: cf. Zaborowski 2002, 92–94.

Reflections on the Discourse of Fear in Greek Sources

111

‘childish’ – spirits (πρὸς τοὺς νηπιόφρονας).9 Clement of Alexandria accuses philosophers of brandishing the gods as bogeymen and of speaking like chattering old women (µυθολογῶν ὕθλῳ γραϊκῷ). It is likewise drunk old women’s discourses and children’s bogeymen that John Chrysostom likens to the belief that there are souls wandering around the world. As for Proclus, he places beliefs in terrifying beings that punish the condemned in hell on the level of puerile terrors.10 Statesmen and rambling old nannies therefore have something in common: the use of fear to achieve their goals. According to Lucian, legislators use punishment as a bogeyman for their constituents. But bogeymen are not useful to tyrants; those who dwell among people who detest them cannot inspire fear: they must act through coercion. Tyrants need to be cruel, for the more they punish, the more reasons they have to punish. Theodoret of Cyr compares the unreasoned fear that children have of bogeymen to the fear that adults experience before courts and government administration.11 These few excerpts represent only a small fraction of the sources that make use of the rhetoric of groundless and childish fear. Irrespective of its (false) object, it is always effective. As Plato would say, there is in each of us a frightened child, and each child has once been afraid of a bogeyman. As for Dio Chrysostom, he recognises that anyone could have his or her particular source of fear, just as each child fears a specific bogeyman and is accustomed to being afraid of it.12 In a number of the cited passages, the use of these bogeymen is deliberate: infernal punishments and gods are among the objects that incite κενοὶ φόβοι. Let us add that, for Aristotle, fear is not an instinctive aversion but a socially conditioned reaction in which power relations and arguments concerning the attitudes of others play a crucial role.13 Xenophon’s Socrates argues that fear renders men more obedient and more disciplined (εὐπειθεστέρους καὶ εὐτακτοτέρους).14 Yet, sources reveal a marked belittling of the fear experienced by the masses, who are considered childish and gullible. We have seen that one arouses their fear through various means: earthly punishments, weapons, or divine punishments are useful for statesmen. The paradoxographers mention cases in which collective fear provokes a political reaction, such as the births of ‘monsters’ like androgynes, supernatural signs, or ghostly apparitions. All of these incite civic assemblies wherein decisions are made regarding the action to undertake so as to respond best to the

9 10 11 12 13 14

Strabo 1.2.8. For a more radical opinion, attributed to the sophist Critias, according to which the gods were invented in order to prevent crime, cf. Chaniotis 2012, 207. Clement of Alexandria, Protreptic 6.67.1.1–3. John Chrysostom, Homiliae XC in Matthaeum, PG 57.353.38–40; Proclus, In Platonis rem publicam commentarii 180.19. Lucian, Phalaris 1.8.1–11. Theodoret of Cyr, Correspondence 36.1–4. Plato, Phaedo 77e. Dio Chrysostom, Oration 66.20.2. Aristotle, Rhetoric 2.5.1382a–1383a. Cf. Konstan 2006, 133. Xenophon, Memorabilia 3.5.5–6.

112

Maria Patera

possible consequences of the manifested sign.15 Fear (φόβος), Aristotle states, makes people ‘ready to deliberate (βουλευτικούς)’. Indeed, fear is caused by the expectation or impression of impending danger, which implies a cognitive judgment that the adversary’s forces are superior. Yet, in order for people to have fear, they must also have hope of being saved, since nobody takes counsel about hopeless things (οὐδεὶς βουλεύεται περὶ τῶν ἀνελπίστων).16 And, according to the Stoic position, in order to stop being afraid, one must stop hoping.17 3 THE CULT OF PHOBOS For those who are not Stoics, there exist other remedies against fear: first of all, the use of reason in order to control the experience of fear, the operating mode of some strategists. Next, the use of mockery to ridicule fear. This is a more aggressive and pervasive way of rationalising fear by making an appeal to adults’ sense of dignity: they should distinguish themselves from children, who are considered as mentally weak, gullible, and cowardly.18 In reality, mockery is to encourage them to confront their fear by appealing to shame (αἰδώς) or the fear of dishonour (αἰσχύνη) which they should experience from fear. Finally, one can also attempt to ally oneself with fear by honouring it: this is the way of the Spartans, who created a cult of the god Phobos. In the comic fragment cited in the epigraph (p. 109), Phobos is described as extremely ugly in appearance or, rather, formless, a god who participates very little in the good. In epic, Phobos is mentioned both as one of the participants in battle and as a shield motif. He is a son and a companion of Ares, often accompanied by his brother Deimos, who bears another name for fear.19 In Hesiod, they are both sons of Ares and Aphrodite.20 In the Iliad, they 15

16

17 18 19

Examples in Phlegon, Mirabilia 2.63.10; 3.68.12f. ed. Keller. In Polybios, φόβος and terms derived therefrom designate an experienced fear, passively endured, that ultimately leads to failure by paralyzing reasoning and impeding all effective action (cf. Guelfucci 1986, 229). Φόβος has a negative effect on action: for example, the Roman crowd, περίφοβος before the imminence of battle, resorted to the irrational (Polybios 3.112.8): ‘All the oracles that had ever been delivered to them were in men’s mouths, every temple and every house was full of signs and prodigies, so that vows, sacrifices, supplicatory processions, and litanies pervaded the town’ (trans. W. R. Paton). Aristotle, Rhetoric 2.5 1383a5–8. Thus fear contains in itself an objective, the salvation that leads to the action to undertake in order to achieve this objective; as Aristotle says (ibid. 3.5.1382b3f.), the frightened person is necessarily in a state of preparation: cf. Fortenbaugh 2008, 36, 42. According to Philo of Alexandria, Legum allegoriarum libri I–III 2.8 ed. Cohn, the utility of fear is to warn the soul to neglect nothing (µηδενὸς ὀλιγωρεῖν): cf. Winston 2008, 212. Thucydides 7.61.2 speaks of ‘fear’s hope’ (τὴν ἐλπίδα τοῦ φόβου), which really means an anxious expectation, but one which remains an expectation and therefore not devoid of hope (cf. Huart 1968, 136 n. 3). Seneca, Epistulae morales 5.7–9. Golden 1990, 5f., 10. Iliad 4.440. Cf. also Artemidoros, Oneirokritika 2.34.

Reflections on the Discourse of Fear in Greek Sources

113

hitch up Ares’ horses, while in the Shield they drive his chariot.21 As a shield motif, Phobos does not seem to have an anthropomorphic form, but he ‘spread round like a crown’ (ἐστεφάνωται) on Athena’s aegis. A dragon-φόβος (or an ἀδάµαντος … Φόβος, depending on the manuscripts) with bright eyes of fire also decorates the shield of Herakles.22 According to Pausanias, the shield of Agamemnon on the chest of Kypselos was decorated with a lion-headed Phobos.23 With respect to figural representations, Phobos appears in anthropomorphic form in the battle between Herakles and Kyknos, the son of Ares, notably on a black-figure oinochoe from Berlin dating to around 530 BCE, where Phobos, identified by an inscription, drives Ares’ chariot.24 But above all, Phobos received a cult that is well attested at Sparta. Phobos had a sanctuary, which is a small room close to (or situated in) the ephoreion,25 and was considered as the fear that citizens ought to have of the law. He had, therefore, a political value with an internal use: he inspired citizens to be obedient.26 Yet, at Sparta, Phobos was also associated with Phygē (Flight), and thus had 20 21

22 23 24

25

26

Hesiod, Theogony 933–935. According to Stobaios, Anthologion 1.49.44.193, Phobos was the son of Selene. Iliad 15.119; Shield 463. In Nonnos, Dionysiaka 39.216f., Phobos and Deimos assist Ares with navigation. Occasionally, Phobos is mentioned alone (Iliad 9.1f. and 13.299) without Deimos. Note that in the philosophical problem on the divine essence posed by Plutarch, Moralia 763c (Dialogue on love [47] 18), philosophers refuse to recognise the divinity of Phobos, while for Clement of Alexandria, Protreptikos 2.26.4, certain philosophers personify (ἀνειδωλοποιοῦσι) human emotions like fear. Iliad 5.739: Athena. Shield 144, 195: Herakles. Pausanias 5.19.4. Cf. Shapiro 1993, 211 n. 134. Phobos is again represented and identified on a cup from Tarquinia, attributed to the inception of the red-figure technique, as a bearded man driving the chariot of Ares; it has been suggested that Phobos is actually the winged spirit flying over the chariot (ibid. 214). Fear is no longer depicted in the Classical period and his known subsequent representations are two Athenian lamps from the Imperial period which portray a bear with the inscription Phobos (cf. ibid. 215; cf. also Deubner 1902, 258, figs. 1 and 2). According to Robert 1981, 28 n. 6, it is not an incarnation of Phobos personified in the form of a formidable animal, but the bear’s name used in combats in the amphitheatre, where the beast battled a bestiarius specialising in this type of combat: the ursarius. A funerary inscription from Kilikia mentioning Phobos, Ate, and Moira, probably as guardians of the tomb, dates to the Imperial period – more precisely to the second century CE: cf. Deubner 1902, 263f.; Heberdey and Wilhelm 1896, 38, note 94D. It is in the sanctuary of Phobos that the ephor Agylaios found refuge in 227 BCE (Plutarch, Kleomenes 8.3); Plutarch states that the sanctuary, normally closed, was likely open by accident, probably because the city was at war. Kleomenes III, who executed the other ephors, spared Agylaios, perhaps because of the fear inspired by Phobos (cf. Richer 1998b, 219f., 223). Furthermore, according to a hypothesis mentioned and refuted by Wide 1893, 276, the Phoibaion mentioned by Pausanias 3.14.9 and Herodotos 6.61, where the ephebes sacrificed to Enyalios, would actually be a Phobeion, a sanctuary dedicated to Phobos (cf. also Bernert 1949, 312). For a chronology of the use of Phobos at Sparta, cf. below p. 125 n. 97. Richer 1998b, 217, 219f., 236, 247; 1999, 92. According to Mactoux 1993, 282–284, Phobos symbolised the power of the ephors, whose function was to put all transgressions back in or-

114

Maria Patera

a martial value with an external use as well.27 Phobos comprised ‘passion-Fear and action-Fear’, that is, passively experienced fear and actively inflicted rout.28 The cult of Phobos represents an institutional use of provoked fear. The citizens must both respect the law and induce fear in their enemies. It is this ‘serious’ use of provoked fear by statesmen that is much mocked in the sources. In Aeschylus, Phobos is one of the guarantor divinities of the oath to destroy Thebes that the Seven take.29 A component signifying fear also describes divine epithets like those of Athena Φοβεσιστράτη30 and of Zeus Δειµάτιος, honoured by the Romans.31 It is actually alongside Zeus and Athena that Phobos is thanked as a divinity who saved the city in an inscription commemorating a victory of Selinous in the fifth century BCE.32 Alexander, just before the battle of Gaugamela against Darius (October 331), sacrificed to Phobos, just as Theseus had done before engaging in the battle against the Amazons.33 In sacrificing to Phobos on the eve of a battle, one sought to divert from one’s army the force capable of destroying it, as well as to direct this force against the opposing camp. In both cases, those who honoured Phobos did so either upon the order of an oracle (Theseus) or with the

27

28

29 30

31 32

33

der, whether they were cultic or institutional. Fear was the concrete manifestation of the ephors’ power, and with the cult of Phobos the Spartans recognised the legitimacy of this power (cf. also Richer 1998b, 224). Indeed, the cult of Phobos is associated by Plutarch, Kleomenes 9.7, with the attribution of a greater power to the ephors: cf. Richer 1998b, 232, and below, pp. 124f. Richer 1998a, 22. The primitive sense of phobos is also ‘flight’, and it is in this sense that the verb φέβοµαι is used in Homeric epic: to signify a troop fleeing in haste and disorder, plagued by panic-stricken fear: cf. Chantraine 1999, s.v. φέβοµαι (cf. Suda, s.v. φόβος· παρ’ Ὁµήρῳ ἡ φυγή). Cf. also Zaborowski 2002, 127–131, 136, and 2004, 567. Fear is also associated with flight in the Stoic excerpt mentioned above (p. 110). Mactoux 1993, 268. Although the warlike Phobos was linked to the heroic aspect of Homeric war in which the combatant tries to terrorise an adversary by various means (cries, grimaces, gesticulations, etc.), it is at Sparta, ‘dans la cité la plus hoplitique’, that Phobos received a cult, being also necessary in phalanx warfare, as Détienne 1968, 124f., remarked. But, ‘la phalange fait de l’hoplite, comme la cité du citoyen, une unité interchangeable’, and hoplite warfare required ‘une maîtrise entière de soi, un constant contrôle pour se soumettre à une discipline commune’ (cf. Vernant 1962, 59). On self-control, enabling one to control not only one’s own emotions but also those of others, cf. below, p. 131. Aeschylus, Seven against Thebes 45f. Aristophanes, Knights 1177. Cf. also Etymologicon magnum, s.v. Φοβέστρατος Ἀθηνᾶ· ἡ φυγὴν τοῖς πολεµίοις ἐµποιοῦσα. In Galen, On the doctrines of Hippocrates and Plato 3.8.14, ed. De Lacy (= Hesiod F 343, 18, eds. Merkelbach/West = Chrysippos F 908, ed. Arnim), the adjective φοβέστρατος describes the aegis of the goddess. Dionysios of Halikarnassos, Roman Antiquities 6.90.1. Syll.3 1122 (= IG XIV 268; see also SEG XXXIV 970; Arena 1989, no. 53; Manganaro 1995; SEG XLIX 1328). For an attempt to attribute an unidentified temple at Selinous to Phobos, cf. Manni 1975, 195. Plutarch, Alexander 31.9; Theseus 27.2. Cf. Mactoux 1993, 279f. A sacrifice to Phobos before engaging in battle is also mentioned by Appian, Libyka 85 eds. Gabba/Roos/Viereck.

Reflections on the Discourse of Fear in Greek Sources

115

assistance of a soothsayer (Alexander), and were met with success: Alexander was victorious and Theseus’ war was concluded upon a treaty.34 4 PHOBOS AND AIDŌS Another way of reacting to fear is to make an appeal to shame, to the fear of the gaze of others. In this case, fear is intimately linked to courage, especially that of a warrior. According to Xenophon, great αἰδώς (fear of dishonour) is accompanied by great obedience (πειθώ), at least among the Lacedaemonians.35 Here one approaches the Aristotelian definition of shame, ‘a kind of pain or trouble concerning the type of bad things that seem to lead to loss of reputation’ (ἀδοξίαν), an ‘appearance of infamy’ (ἀδοξίας φαντασία).36 The appeal to shame during war constituted, according to a Homeric scholion,37 a Greek trait, whereas ‘barbarians’ made appeals to fear: Hector threatens to kill his men if they attempt to escape from battle.38 This is certainly not a uniquely ‘barbarian’ discursive feature, since in the same Iliad, Agamemnon threatens to dispose of his men as food for the dogs and birds if he sees them retreat from combat.39 On this point, Aristotle commits an interesting error: in discussing military leaders forcing their soldiers to fight violently, he erroneously places Agamemnon’s threat in the mouth of Hector.40 According to Plato, it is fear of dishonour before friends (φίλων φόβος αἰσχύνης) combined with courage that assures victory in war.41 Further34

35

36 37

38 39 40 41

Phobos appears completely armed in a magical papyrus: PGM XIII.529 ed. Preisendanz (cf. also XIII.544 and 586); Dieterich 1891, 87f. In addition, Phobos is the name of one of the twelve Pans in Nonnos, Dionysiaka 14.81. According to Bernert 1949, 313f., the militaristic function of Phobos was taken up by Pan when Phobos lost his function as a divinity and became a spirit present in folklore. This thesis is refuted by Borgeaud 1979, 152, who emphasises that it is exactly in the Classical period that Phobos appears in ritual, with the inscription of Selinous, the Spartan cult, and Alexander’s sacrifice. Yet, in the same period, Pan appears as the cause of military panic; it is therefore difficult to argue that one supplanted the other. Xenophon, Constitution of the Lacedaemonians 2.2. For this meaning of αἰδώς, cf. Richer 1999, 98. On the different meanings of the term (respect, shame, restraint, modesty, honour, timidity, reverence in the sense of respectful fear), cf. Zaborowski 2002, 120–126. Let us add that according to Pseudo-Plato, Definitions 412c8–10, shame (αἰδώς) is, inter alia, the desire to avoid justified reproaches (εὐλάβεια ὀρθοῦ ψόγου). For the fear of ill-repute, cf. also Plato, Euthyphron 12b. Aristotle, Rhetoric 2.6.1383b13–15 and 1384a22. On αἰσχύνη, cf. Cairns 1993, 57f., 71, 182 note 11, 205, 216, 236f., 266, 278, and below, p. 116. Scholia ad Iliadem 13.95a: Ἑλληνικὸν τὸ τῆς αἰδοῦς προβάλλεσθαι· τοῖς δὲ βαρβάροις ὁ Ἕκτωρ οὐ τὸ αἰσχρόν, ἀλλὰ τὸν φόβον. On αἰδώς associated with fear in Homeric epic, cf. Cairns 1993, 69–71. Iliad 12.250. Iliad 2.391–393. Aristotle, Nicomachean Ethics 3.11.1116a38–40. Plato, Laws 1.647b (cf. also 699c); Cairns 1993, 375. According to Pseudo-Plato, Definitions 416a8, disgrace (αἰσχύνη) is fear before expected dishonour (φόβος ἐπὶ προσδοκίᾳ

116

Maria Patera

more, the Spartans say of themselves that they are courageous in battle because they have learned to feel shame before their leaders (αἰδεῖσθαι) and not to fear (φοβεῖσθαι).42 However, Leonidas seems to say that it is the combination of the two sentiments that ensures victory: ‘only those who respect and fear their kings battle against their enemies.’43 According to the Athenian Lykourgos, a Spartan law prescribed death for those who refused their duty to their country, a law on which the orator comments in saying that ‘the fear (φόβος) inspired by fellow citizens would be powerful enough to compel each person to brave the perils of confronting the enemy’.44 And Xenophon distinguishes Sparta from the cities where ‘sufficiently powerful men do not want even to be considered afraid of the authorities’ (οὐδὲ βούλονται δοκεῖν τὰς ἀρχὰς φοβεῖσθαι).45

42

43

44

45

ἀδοξίας). On the relationship between the notions of αἰδώς/αἰσχύνη and φιλία, cf. Cairns 1993, 89–95, 174f., 183, 273–276. Zonaras, s.v. φόβος, introduces a difference between αἰδώς, which is fear before the expectation of reproach (φόβος ἐπὶ προσδοκίᾳ ψόγου), and αἰσχύνη, which is fear before a committed infamy (φόβος ἐπὶ αἰσχρῷ πεπραγµένῳ). Αἰδώς here would be the inhibitory fear that prevents one from committing a misdeed, whereas αἰσχύνη would be the fear accompanying the awareness of a committed crime, the fear of the consequences. This difference is rarely clear in the sources, wherein the two terms are often interchangeable: cf. Cairns 1993, 138f., 236f., 264, 271f., 300f., 365f., 380, 414–423. Even Aristotle, who speaks of numerous forms of shame in Rhetoric (2.6.1383b11–1385a15), often employs the two terms indiscriminately: cf. Aubonnet 1986, 226 note 7 (comment on Aristotle, Politics 7.12.5, 1331a41). Plutarch, Moralia 231e–f [Lakonika Apophthegmata] (cf. also 217a). Yet, when it comes to sending scouts, it is best to dispatch in small numbers men who are a little bold, ‘fear being a formidable watch companion’ (ὁ γὰρ φόβος δεινὸς δοκεῖ συµφύλαξ εἶναι): Xenophon, Hipparchikos 7.7; cf. de Romilly 1956, 124. Thucydides 6.34.9 confirms the utility of fear: ‘the surest preparations are those made in fear’ (τὰς µετὰ φόβου παρασκευὰς ἀσφαλεστάτας; cf. also 2.11.5). This virtue of fear and its usefulness even lead to the conclusion that fear can become a strength and render the weak superior to the strong, an idea that the Korkyreans present to the Athenians in Thucydides 1.36.1: ‘his fear (τὸ µὲν δεδιὸς αὐτοῦ), joined with strength (ἰσχὺν ἔχον), will better frighten his adversaries (τοὺς ἐναντίους µᾶλλον φοβῆσον), and his confidence, which (…) will accompany weakness before a strong enemy, will cause less fear’ (ἀδεέστερον ἐσόµενον); cf. de Romilly 1956, 122f. Plutarch, Moralia 225d [Lakonika Apophthegmata]: µόνοι πρὸς τοὺς πολεµίους µάχονται οἱ τοὺς βασιλέας αἰδούµενοι [καὶ φοβούµενοι]. On Leonidas’ saying and the problems raised by the words καὶ φοβούµενοι, cf. Richer 1999, 98f. Already in the Iliad 15.657f., it is αἰδώς combined with fear (δέος) that keeps the Achaeans from fleeing; thus, the two emotions, in counteracting flight, indirectly produce the effect of courage, as Zaborowski 2002, 76, 103f., observes (for the use of αἰδώς and of the verb αἴδοµαι in Homeric epic, cf. ibid. 100–126). Lykourgos, Against Leokrates 130 (we again encounter the same idea in Isokrates, Panegyrikos 77). According to Aischines, Against Ktesiphon 175, one of Solon’s laws punishes cowards so that soldiers, fearing more the imposed punishments than their enemies, become better fighters. Xenophon, Constitution of the Lacedaemonians 8.2.

Reflections on the Discourse of Fear in Greek Sources

117

5 PHOBOS AND LOGOS Fear is thus closely associated with courage. In Aristotle’s Rhetoric, courage is defined as the inverse of fear and goes hand in hand with hope, which comes with the impression that safety is near (µετὰ φαντασίας ἡ ἐλπὶς τῶν σωτηρίων ὡς ἐγγὺς ὄντων), whereas frightening things are either non-existent or remote.46 Fear is a type of pain and trouble arising from the appearance of a bad thing presumed to be devastating or painful; this thing must appear close and equipped with great destructive power.47 Yet, in the Nicomachean Ethics, courage is not the absence of fear, but the right attitude toward the experience of fear; it thus becomes a habit set by education. Indeed fear, like anger, is an involuntary emotion: ὀργιζόµεθα µὲν καὶ φοβούµεθα ἀπροαιρέτως.48 Only crazy and insensible people like the Celts do not fear terrible things exceeding human limits, like earthquakes and the sea’s waves.49 The courageous man is he who fears and understands what is circumscribed by human limits, but also one who fears with reason (λόγος) and faces it virtuously.50 ‘Comme si le vrai courage se signalait à ce qu’il sait éprouver la terreur pour mieux la domestiquer’, ‘comme si la peur était l’épreuve qualifiante

46

47

48 49

50

Aristotle, Rhetoric 2.5.1383a17–20. For example, one is not afraid of death during one’s life, for death appears distant; nor does one fear becoming stupid or unjust, for one believes that it is within one’s power to prevent such a development (ibid. 1382a23–28). Fear thus implies a relationship with time and space (in terms of remoteness) and φαντασία, imagination or impression: cf. Serghidou 2007, 217f., and notes 4 and 5. Note here the interesting view of emotions expressed by Philo of Alexandria, De mutatione nominum 161–163 ed. Wendland: ‘When one hopes for the good, one’s soul rejoices in anticipation, and, thus, in a sense, experiences joy before joy, contentment before contentment. ... In the same way, the presence of evil elicits grief, and its expectation causes fear. Thus, fear (φόβος) is grief before grief (λύπη πρὸ λύπης), just as hope (ἐλπὶς) is joy before joy’ (χαρὰ πρὸ χαρᾶς). See also Nemesios of Emesa, De nature hominis 17.75.17f. ed. Morani, who defines fear as a future evil and grief as a present evil (προσδοκώµενον µὲν κακὸν φόβος, παρὸν δὲ λύπη). On the relationship to Stoic views, cf. Winston 2008, 205. Aristotle, Rhetoric 2.5.1382a21–25, 1383a17–19. This definition is very close to that of the Stoics, according to which fear is belief in a future, menacing evil: cf. Wisse 1989, 289, note 159. However, in the ensuing discussion, Aristotle also introduces the opinions or beliefs implied by the emotion of fear, not only simple impressions or appearances. For example, the opinion that the injury that one can suffer will be irreparable reinforces fear (ibid. 1382b22– 24). The existence of certain opinions in the subject’s mind can also prevent him or her from feeling terror: cf. Knuutila and Sihvola 1998, 13. Aristotle, Nicomachean Ethics 2.5.1106a2f. Aristotle, Nicomachean Ethics 3.7.1115b24–28. Note, in addition, that Aristotle (On the Soul 1.1.403a19–24) introduces bodily condition as a factor determining the intensity of an emotion: thus, one can be frightened without anything frightening happening, or, inversely, one can be unafraid in the presence of significant dangers (cf. Fortenbaugh 2008, 38). Aristotle, Nicomachean Ethics 3.7.1115b11–13. Cf. Sihvola 1996, 143. In Plato, Republic 429a–430c, the civic courage of city guards depends on the education they receive, and consists of the development of correct opinions with respect to what is to be feared and what is not.

118

Maria Patera

du héros’, remarked Nicole Loraux with regard to the Iliad.51 And even if emotion is in itself irrational, virtuous men experience it rationally.52 Finally, what matters is the manner in which the event is evaluated by the agent, not the event itself. Thus the Spartan Agesilaos does not rout his enemies using Phobos (οὐ φόβῳ τρεψάµενος νίκης ἔτυχεν), but he shows himself unaffected by fear by being cheerful in moments when fear would be the appropriate emotion (εἴθιστο δὲ φοβούµενος µὲν ἱλαρὸς φαίνεσθαι).53 In Seven Against Thebes, Aeschylus shows that one can conquer fear with λόγος. Eteokles is disgusted with the cries of fear of the Theban women, ‘hated by all reasonable people’ (σωφρόνων µισήµατα), and with their fear of war, which is a great evil for their houses and the city, and which risks discouraging the army.54 The herald announces the enemies’ seven champions and Eteokles assigns seven of his own champions to confront them.55 He himself is the final champion and announces that he will meet his own brother, Polyneikes, in battle. In this moment, the choir of Theban women stop fearing the imminent battle, very probably reassured by the distribution of the forces, but feeling henceforth horror at the defilement that may result from fratricide.56 Even though Eteokles describes the women’s fear as irrational, it is not in fact. They fear defeat in battle and are reassured by the explanation of their king’s strategy. This recalls the appeal to logos in order to combat fear. It is the logos used by Eteokles in presenting his strategy in the face of the looming battle that comforts the frightened Theban women. Perikles himself also uses reason to reassure his soldiers, who were frightened by

51

52

53 54 55 56

Loraux 1989, 92 and 94. Courage often consists in the domination of fear in Homeric epic: cf. Zaborowski 2002, 281, 284. Cf. also Plato, Phaedo 68d, which demonstrates that courageous men suffer death out of fear of greater evil; therefore, they are only courageous because of their fear. Cf. Nehamas 1994, 268. Note a further distinction introduced by Aristotle, Nicomachean Ethics 3.8.1117a17–20: the temporal dimension of emotion. True courage is possessed by the one who confronts a sudden danger, whereas the coward flees; conversely, when the situation permits deliberation, a person who lacks courage can choose to confront the danger which he considers as apparent. As for the Stoics, they also recognise ‘two times’ in fear (cf. the title of the article by Serghidou 2007a): immediate fear, unreasoned φόβος before danger or before its alleged proximity, and δέος, reasoned fear having weighed the consequences of action (cf. below p. 122; cf. also Diogenes Laertius 7.112; Plato, Laches 198b: δέος γὰρ εἶναι προσδοκίαν µέλλοντος κακοῦ). These two types of fear correspond to the Stoic stages of the emotional process: a first stage of evaluating the situation directly succeeding the stimulus, and a second stage of deciding the emotion to feel. The Stoics believed that before a sudden and terrifying event, the soul cannot keep from being shaken. But the wise person does not accept these imaginations, these terrifying visions; he or she does not give consent to fear, while common folk surrender to this initial impression: cf. below p. 131; Borgeaud 2007, 212–214. Xenophon, Agesilaos 6.2; 11.2. Aeschylus, Seven Against Thebes 184, 186, and 189f. Aeschylus, Seven Against Thebes 448, 467, 476, 504, 555, 620, 673. Aeschylus, Seven Against Thebes 677–682, 720–726, 790f.

Reflections on the Discourse of Fear in Greek Sources

119

the eclipse of the sun: he unfurls his cloak before the steersman and asks him whether the resultant shadow is truly something remarkable.57 In all cases, fear is cognitive: it depends on the evaluation of danger and, in return, this emotion affects judgment. This is Aristotle’s definition: the emotions are ‘the causes that make men vary in their judgments’ (ἔστι δὲ τὰ πάθη δι’ ὅσα µεταβάλλοντες διαφέρουσι πρὸς τὰς κρίσεις), and fear is provoked by the expectation or impression of an impending danger.58 Φόβοι κενοί result, in reality, from poor evaluation of a non-existent danger, and they affect people’s judgment by rendering them ‘mentally weak’ like children (Strabo’s νηπιόφρονες).59 Finally fears, whether rational or irrational, the means of responding to fears, and the ‘remedies’ for them are the same. The commander of the Athenian navy, Phormion, used an interesting tactic to encourage his men before battle by making them consider the fear that they elicited from the enemy. The terror provoked is thus legitimated by reason.60 It is a brilliant rhetorical tactic of making one’s men imagine the fear experienced by the other in order to incite their self-confidence; making them imagine this fear is entirely contingent on the use of arguments – of logos – which recalls the rationalisation of fear.61 Once again, when Eteokles addresses himself to the women of Thebes in order to reprimand them for their fright62 and then reassures them by explaining his strategy, it is still a statesman – a commander – who speaks to a crowd. Whether this crowd is composed of women, soldiers, or commoners, it is always the logos of one that orchestrates the emotions of the many. Crowds are impressionable, like children. 6 PROVOKING PHOBOS Speaking of provoked fear is speaking of false objects employed for that purpose, a stratagem similar to that used during the famous episode of the war between the Phokians and the Thessalians (shortly after 500 BCE) that Herodotos recounts.63 The Phokians had been pushed back, and their soothsayer, Tellias of Elis, devised the following stratagem (σοφίζεται):

57 58 59

60 61

62 63

Plutarch, Perikles 35.2. Aristotle, Rhetoric 1378a20f. On fear as a cognitive and not instinctive emotion in Aristotle, see Konstan 2007, 417. For example, Aristotle, Nicomachean Ethics 8.6.1149a7–9, believes that fear of mice or cats is so absurd that it becomes pathological (διὰ νόσον), or rather it is likened to bestial cowardice (θηριώδη δειλίαν). For νηπιόφρονες, cf. above p. 111. Thucydides 2.89.5. Cf. Huart 1968, 133. Furthermore, this may imply a sort of ‘transfer of fear’: when fright is transformed into reasoned fear, this change is likely to frighten the adversary; cf., for example, the reasoning of the Korkyreans in Thucydides 1.36.1 (above, p. 116, note 42; Huart 1968, 125). Aeschylus, Seven Against Thebes 238–240. Herodotos 8.27.3–4., translated by A. D. Godley. Cf. also Pausanias 10.1.11.

120

Maria Patera He covered six hundred of the bravest Phokians with gypsum (γυψώσας), themselves and their weapons, and had them attack the Thessalians by night. ... The Thessalian sentinels were the first to see them and to flee for fear, supposing falsely that it was something supernatural (δόξασαι ἄλλο τι εἶναι τέρας), and after the sentinels the whole army fled as well. The Phokians made themselves masters of four thousand dead and their shields.

This interesting stratagem again pertains to childhood: in Kallimachos, Hermes disguises himself as a bogeyman by smearing black ash on his face in order to frighten (µορµύσσεται) a little girl.64 This strange face, daubed with black, has no more recognisable traits than the Phokians whitened with gypsum. It has been thought that Hermes’ ‘soot mask’ contrasts with the Phokians’ ‘gypsum mask’: covering oneself with gypsum can be a tactic to deceive one’s victim (like the Titans masking themselves with gypsum in order not to be recognised by the baby Dionysos, whose throat they were going to slit) and not to frighten the victim.65 However, covering oneself with gypsum evinces, as much as with soot, a wish to deceive one’s victim and to frighten it. The consequences of the elicited terror are, owing to the given circumstances, what differentiates them. In addition to being mistaken about the identity of their adversaries, the Phokians are terrified as is the little girl by Hermes, and in the same way as children generally are – not only by their nannies’ stories about bogeymen, but also by the masks that one shows to frighten them. Indeed an entire series of sources mentions masks with which one menaces children, προσωπεῖα in general, or even the actors’ masks called µορµολυκεῖα or γόργια, which make reference to the bogeymen of children (Mormo) and of adults (Gorgo) respectively. A scholion to Plato’s Gorgias states that with these masks, women frightened children; according to Epictetus, the masks are ‘scary’ (φοβερά) for them.66 According to the Etymologicon Magnum, such masks (µορµολυκεῖα) are fashioned ‘in order to strike fear’ (πρὸς κατάπληξιν).67 Yet, the κατάπληξις is still, according to [Andronikos], On the passions, a type of φόβος that is provoked by a strong imagination/appearance (ἐκ µείζονος φαντασίας).68 And the Phokians covered in gypsum are exactly a frightening apparition, a pretence: they wear a mask that assures them victory. Warriors, like statesmen, also use the ‘tricks of nannies’. 64 65 66 67

68

Kallimachos, Hymn to Artemis [3] 65–71. Ellinger 1993, 175. For Dionysos and the Titans, cf. Eustathius, ad Iliadem 2.735, ed. Van der Valk, vol. I, 519, 1–7. Scholia Platonica, Gorgias 473d. Epiktetos, Conversations II.1.15.3. On mormolukeia, cf. Patera 2005. Etymologicon Magnum, s.v. µορµολυκεῖον. Cf. also Scholia in Theocriti Idyllia 15.40b. On frightening masks used in a ritual of inversion – those of the sanctuary of Orthia at Sparta, where there were youths who used masks to scare the adults in charge and not the reverse –, cf. Mactoux 1993, 288–294. [Andronikos], On the passions 3.6. And κατάπληξις is also used in politics: according to Xenophon, Constitution of the Lacedaemonians 8.3-4, the function of the ephors at Sparta was established ‘in order to strike the citizens with a certain fear that would generate obedience’ (καταπλήξειν τοὺς πολίτας τοῦ ὑπακούειν).

Reflections on the Discourse of Fear in Greek Sources

121

Plutarch explains that in order for children to fear masks no longer, one must turn over the masks and show them to children from all sides. Epictetus advocates this very thing in an excerpt where he rises up against the fear of death: ‘death, what is it? A frightening mask. Turn it over and learn; see, it does not bite’.69 In this type of thinking, it is knowledge that provides the remedy against fear. In Plato, there is nothing more powerful than knowledge (ἐπιστήµη), which can even conquer the quest for pleasure (ἡδονή), whereas the majority of people think that man can have knowledge whilst being governed by their emotions, such as fear.70 Yet, all masks conceal, dissimulate the true nature of those who wear them, and rig reality and the knowledge that one may have of it. Clement of Alexandria speaks of the truth that stands concealed, just as ‘the true face under frightening masks’71 is hidden. This secret dimension is frightening in itself: the mask conceals, but perhaps it only hides emptiness, an innocuous and somewhat ridiculous void, as always when referring to something childish. Still, the masks used to scare are ugly, disgraceful (αἰσχρά), according to a scholion to Aristophanes. And ‘the laughable’, Aristotle says, ‘is a part of the ugly’ (τοῦ αἰσχροῦ ἐστι τὸ γελοῖον µόριον) – ugly like the god Phobos, who plays some role in a comedy, as in this chapter’s epigraph (p. 109).72 In the cited passages, emphasis is placed on the methods used to provoke fear, and on the manner of eradicating it when it is felt. The means are mainly pretences: dissimulations (gypsum, masks) and extravagant stories (bogeymen, divine weapons, punishments in the afterlife). In short, provoked fear is a specialty of the forger (nanny, warrior, statesman, and playwright). Even the philosopher, the lover of truth, in Plato’s Republic must, because of his role as governor of the city, make use of the art of imposture.73 In this case, the science of philosophy serves to control society’s beliefs, which determine the comportment of its members. Linked to the masses, politics belongs to the sphere of δόξα, without which the philosopher would not be able to hold power.74 Yet, these beliefs in terrible things are useful instruments in the hands of governors, as Strabo argues.75 Prodigies, for

69 70 71 72

73 74 75

Plutarch, Moralia 600e (On Exile [44] 5); Epiktetos, Conversations II.1.13–17: θάνατος τί ἐστιν; µορµολύκειον. Στρέψας αὐτὸ κατάµαθε· ἰδοῦ, πῶς οὐ δάκνει. Plato, Protagoras 352b–c, 357c. Clement of Alexandria, Stromateis 2.1.3.5: καθάπερ ὑπὸ τοῖς µορµολυκείοις τὸ πρόσωπον τὸ ἀληθινόν. Scholia in Aristophanis Pacem 474. Aristotle, Poetics 5.1449a33–35. Note that Aristotle summarises here the relationship between shame and laughter, a relationship to which we will return (cf. below, pp. 127f.). Indeed, the term αἰσχρόν means both ‘ugly’ and ‘shameful, dishonourable, defamatory’; and αἰσχύνη, shame and the fear of dishonour, is associated with the fear of ridicule, the fear of the others’ laughing at one’s own expense. Plato, Republic 6.485c3f., 501d2. Nicolaïdou-Kyrianidou 2006, 256f. Note that inflicted terror (φόβος) is not always enough to prevent more strongly experienced emotions from manifesting themselves. Such is the case of the pain and hatred that the citizens felt before the execution of Agis in Plutarch, Agis 21.1.

122

Maria Patera

example, constituted useful instruments in the hands of Greek and Roman statesmen. As Raymond Bloch remarked,76 cette utilisation sans scrupules des craintes collectives n’échappa pas à l’observateur attentif, mais cette prise de conscience des bons esprits n’atténua pas l’efficacité de cette arme de choix qu’offrait la psychologie des foules.

However, mocking the fear of others is a rhetorical topos, typical of bons esprits. The majority of authors take part in it, identifying themselves with the rational reaction in the face of fear – with the use of logos, and not with the emotion itself. In Thucydides, for instance, intelligent and courageous men never feel irrational, affective fear (φόβος); only sometimes do they feel a reasoned, thoughtful apprehension (δέος). Φόβος is reserved for weak characters and minds which are not capable of calmly envisioning a situation and which allow themselves to be distressed by events.77 Thucydides never ascribes this type of fear to Perikles, a man who distinguishes himself by the firmness of his thinking and the strength of his character.78 But a politician can use phobos:79 When one has personal fears (reasoned: δεδιότες), one establishes paralyzing fear (ἔκπληξιν) in the city and makes use of common fear (φόβῳ) in order to obscure one’s objective.

76

77

78 79

Bloch 1963, 38f. We have encountered the power of an alleged prodigy with the Phokians’ strategy (above, pp. 119f.; for other examples cf. above, p. 112, note 15). In a passage from Plutarch, Aratos 31.4–32.4, the attackers are vanquished in the face of what they believe is a divine epiphany. Furthermore, one can pretend to experience a divine epiphany or to be pursued by the divinity in order to provoke fear in one’s audience and thus escape condemnation (cf. Plutarch, Marcellus 20.8–10). For these two passages, see the analysis of Angelos Chaniotis in this volume (pp. 54–57). Huart 1968, 123–126; cf. also de Romilly 1956, 119. Like Thucydides, Polybios (cf. Guelfucci 1986, 229) uses φόβος and its derivatives to signify an irrational, affective, thoughtless fear, whereas δέος and δείδω express thoughtful fear before the eventual consequences of an event, hence a cognitive fear. The suitable response to affective fear is an appeal to reason (cf. also above, p. 110). In Polybios, ‘good fear’, δέος, designates the fear of an evaluated danger, which is preliminary to taking appropriate measures – the kind of fear that saves, for example, Hannibal and his army (3.53.1). This also applies in Thucydides 1.137.2, where Themistocles, the victor of the battle of Salamis, feels a reasoned apprehension (δείσας) that leads exactly to the conception of a salvation plan. Huart 1968, 129: barbarians, lacking reflection, are the ones who often feel this type of fear, and it is Aristophanes, Peace 606, who ascribes it to Perikles. Thucydides 6.36.2. Cf. de Romilly 1956, 120. According to Thucydides 2.65.9, Perikles shocked the Athenians with his words in order to make them afraid (λέγων κατέπλησσεν ἐπὶ τὸ φοβεῖσθαι). In this case, it is a useful fear (cf. Huart 1968, 136 note 2), but elsewhere, good citizens feel fear (φόβῳ) before the violent attacks of the demagogues that deprive the city of good advisors (Thucydides 3.42.4).

Reflections on the Discourse of Fear in Greek Sources

123

7 PHOBOS AND THE LAW The use of the crowds’ fear by statesmen and legislators also feeds the thinking of bons esprits – political thinking primarily. As Lucian would say, tyrants are subject to the necessity of governing by fear and coercion. If they want to be happy, they must make their subjects happy, and not rule with fear, danger, and evil.80 Furthermore, the people are governed by laws or tyrants, and tyrants by their own fears.81 On the other hand, as concerns the subjects of all such governments, it is preferable to obey the law not out of fear of sanctions but by accepting norms in a reasoned manner. This is as Democritus basically states: ‘Abstain from misdeeds not out of fear, but because one ought not to commit them’ (µὴ διὰ φόβον, ἀλλὰ διὰ τὸ δέον ἀπέχεσθαι ἁµαρτηµάτων).82 Clement of Alexandria applies the same sort of thinking to divine law: he distinguishes those who live under the Law, fearing It like the insane (ἀφρόνως), from those who freely choose to believe and experience fear in a sensible manner (ἐµφρόνως).83 It must be added that, beyond its coercive function, the law also has an educational function. When Aristotle analyses the role of sanctions imposed by law, he recognises that ‘the multitude’ (τοὺς πολλούς) submit to fear (φόβῳ) and to punishments (τιµωρίας) rather than to honour (αἰδοῖ), and that discourses (λόγοι) must urge men to virtue through honourable motives, but cannot reach the masses; this is why legislators ought to appeal to the virtue of the few and to use punishments against the rest.84 Aeschylus, through the mouth of Athena, expresses this idea in a more poetic fashion: ‘not to drive fear wholly out of the city. For who among mortals, if he fears nothing, is righteous?’85 Thus, divine sanctions, like human ones, have a preventative function, and punishment has an educational value. This is also what Plato states in the Gorgias:86 80 81

82 83 84 85

86

Xenophon, Hieron 8–11 (cf. Gray 2000, 148f.); Aristotle, Politics 5.9.1315b4–10 (cf. Roberts 2000, 358). For Lucian, cf. above, p. 111 note 11. Philemon F 31, ed. Edmonds. In Thucydides 6.59.2, the tyrant Hippias’ cruelty is also inspired by fear (διὰ φόβου). For Plato, Republic 8.578d–e, tyrants feel the same fear as the slave-owner who finds himself at the mercy of his slaves. The despot’s authority presupposes relationships based on mutual fear: cf. Thalmann 2007, 195, 198. Democritus B 41 eds. Diels/Kranz. Cf. Taylor 2000, 127; Cairns 1993, 364–366. Clement of Alexandria, Paidagogos 1.6.33.3. Aristotle, Nicomachean Ethics 10.10.1179b7–14, and 10.10.1180a6–12. Aeschylus, Eumenides 698f.: καὶ µὴ τὸ δεινὸν πᾶν πόλεως ἔξω βαλεῖν· τίς γὰρ δεδοικὼς µηδὲν ἔνδικος βροτῶν (translated by H. W. Smyth). In the Eumenides (690f.), fear combined with respect (σέβας) keeps citizens away from crime; cf. de Romilly 1958, 111f. Plato, Gorgias 525b, translated by W. R. M. Lamb. In Aristotle, Rhetoric 2.5.1382b24–26, and 2.8.1386a2f. and 27f., this fear of others’ sentences produces pity: everything that provokes our fear elicits our pity when it happens to others. Finally, the majority of occasions of fear are also occasions of pity, and in fear, as in pity, we acknowledge our own vulnerability (cf. Nussbaum 1996, 309, 312). For shared vulnerability in vicissitudes as a condition of pity, cf. the references furnished by Douglas Cairns in this volume (p. 103). On the association of fear with pity as central in tragedy, cf. Halliwell 1986, 170f.; on the subject of tragedy in Aristotle, cf. Schadewaldt 1955. On how fear and pity work in tragedy (fear necessitating identi-

124

Maria Patera Every one under punishment rightly inflicted on him by another should either be made better and profit thereby, or serve as an example to the rest, that others seeing the sufferings he endures may in fear amend themselves (ἵνα ἄλλοι ὁρῶντες πάσχοντα ἃ ἂν πάσχῃ φοβούµενοι βελτίους γίγνωνται).

By the way, ‘he who fears the law will be not be troubled by it’.87 Just as a father’s rules form and train the child,88 ‘the city instructs man’.89 Thus we come to paternal authority, which the commander also enjoys. The comic poet Timokles expresses this idea very clearly: ‘He who feels fear and shame before his father will be a good citizen’.90 However, Aristotle thinks that the power of the law is much greater than that of paternal authority which does not have this irrepressible force resembling need.91 As Menander would say, ‘a father who threatens does not inspire great fear’.92 Thus fear – as a means of education and manipulation – has a utility that even the gods acknowledge, as Athena’s words in Aeschylus demonstrate. What is more, ‘fearing the divine is what wise mortals do’, according to Menander.93 The Christian God himself makes use of fear: according to Theodoret of Cyr, God created beasts and reptiles as children’s bogeymen (µορµολυκεῖα) in order to frighten men and attract them to Him; and he concludes that just as adults (οἱ τέλειοι) despise bogeymen, students of virtue (οἱ τῆς ἀρετῆς τρόφιµοι) do not fear the attacks of beasts.94 This is not without recalling the Socratic maxim ‘one cannot do harm to a good man’, since virtue cannot be touched by external contingencies.95 The Spartans more than all others recognised fear’s utility to the point of honouring it as a divinity. An excerpt from Plutarch both illustrates and summarises the relationship between fear, the divine, and politics:96 Now, the Lacedaemonians have temples of Death, Laughter, and that sort of thing, as well as of Fear. And they pay honours to Fear, not as they do to the powers which they try to avert because they think them baleful, but because they believe that fear is the chief support of their

87 88 89 90 91 92

93 94 95 96

fication with the suffering subject, while pity requires psychological distance, the consciousness of the ‘other’), cf. Rosenbloom 1993, 166f. Menander F 858, ed. Edmonds: νόµον φοβηθεὶς οὐ ταραχθήσῃ νόµῳ (‘he who fears the law will not be troubled by it’). Sophocles, Ajax 548. Simonides F 50, eds. Bergk/Hiller: πόλις ἄνδρα διδάσκει. Timokles F 36, eds. Austin/Kassel: ὅστις φοβεῖται τὸν πατέρα κᾀσχύνεται, οὗτoς πολίτης ἀγαθὸς ἔσται. Aristotle, Nicomachean Ethics 1180a22f. On the educative function of the law, cf. de Romilly 1971, 228–231. Menander F 453, ed. Edmonds: πατὴρ δ’ ἀπειλῶν οὐκ ἔχει µέγαν φόβον. And conversely (F 418), being a father inspires sadness, fear, and anxiety (λύπη, φόβος, φροντίς) without respite. Menander F 1106, ed. Edmonds: φόβος τὰ θεῖα τοῖσι σώφροσιν βροτῶν. Theodoret of Cyr, Quaestiones in Genesim, Patrologia Graeca 80, 97B. Plato, Apology 30c–d, 41c–d. Cf. Nussbaum 1996, 313. Plutarch, Kleomenes 9, translated by B. Perrin, slightly modified.

Reflections on the Discourse of Fear in Greek Sources

125

97

civil polity. For this reason, too, when the ephors enter upon their office, as Aristotle 98 says, they issue a proclamation commanding all men to shave their moustaches, and to obey the laws, that these may not be severe upon them. They insist upon the shaving of the moustache, I think, in order that they may accustom the young men to obedience in the most tri99 fling matters. And the men of old, in my opinion, did not regard bravery as a lack of fear (ἀφοβίαν), but as fear of reproach and dread of disgrace (φόβον ψόγου καὶ δέος ἀδοξίας). For the men who feel most dread of the laws have most courage in facing their enemies; and those shun death least who most fear ill fame. Therefore it has been well said: ‘... for where 100 dread is, there also is reverence’ (ἵνα γὰρ δέος, ἔνθα καὶ αἰδώς). And Homer says: 101 ‘revered art thou by me, dear father-in-law;’ and also: ‘without a word, in dread of their 102 leaders.’ For by the multitude reverence is most apt to be felt towards those whom they also fear. For this reason, too, the Lacedaemonians erected a temple to Fear alongside the mess104 hall of the ephors,103 after they had endowed this magistracy with almost absolute powers.

Here, the author shows clearly the political manipulation of religion (and especially of Phobos),105 recalling Polybios, who considers superstition a fundamental in-

97

98 99

100 101 102 103

104 105

Phobos and other abstractions venerated by the Spartans, like Aidōs and Thanatos, are already mentioned in an elegy of Tyrtaios, written toward 640 BCE and transmitted by the Athenian orator Lykourgos, Against Leokrates 107 (cf. Richer 1998b, 227 note 51). According to Richer 1999, 98f., Phobos was initially intended for enemies, and he only acquired his double value affecting citizens as well later on, perhaps in the first half of the sixth century (the systematic conceptualised use of abstractions in service of the city takes place in this period: cf. Richer 1998b, 230f., who provides a history of abstractions at Sparta); since the beginning of the fifth century, Phobos is also present within the Spartan army as a power intended to ensure unit cohesion, cohesion assured by the fear felt before the leaders (cf. the saying attributed to Leonidas, above, p. 116 note 43). Aristotle, fr. 539, ed. Rose. Richer 1998b, 251–255, relates this order to the status of eromenoi of twenty- to thirty-year old youths, who are entitled to wear a beard but not the moustache reserved for the erastai. Through this order, the ephors recalled the rules of another abstraction honoured at Sparta, Eros, and not those of Phobos, which is associated with the long hair that gives warriors a dreadful appearance. Kypria F 24, ed. Davies. Cf. also Plato, Euthyphron 12b. Cf. Cairns 1993, 372, note 83. Iliad 3.172. Iliad 4.431. This passage is like an illustration of the principle articulated by Aristotle, Politics 7.12.5.1331a40–1331b1, according to which ‘the presence of magistrates in the public eye best inculcates true shame and fear in free men’ (τὴν ἀληθινὴν αἰδῶ καὶ τὸν τῶν ἐλευθέρων φόβον); cf. Richer 1998b, 221. See also Aischines, Against Timarchos 180, who discusses the Spartan gerontes who were surrounded by fear and respect (τις τῶν γερόντων, οὓς ἐκεῖνοι καὶ αἰσχύνονται καὶ δεδίασι). According to Africa 1960, this passage was inspired by the writings of Phylarchos of Athens. On Phobos as a foundation of the Spartan politico-religious system, cf. Richer 1998b, 258. Serghidou 2007, 219f., 223, attributes his cult to an implicit fear of breaking civic cohesion, and more precisely to the threat of the helots. As for Epps 1933, he associates the importance of Phobos with the Spartan’s cowardly ‘nature’, a nature that they would have combatted with the rigour of their education.

126

Maria Patera

strument of government among the Romans.106 Fear is placed at the centre of the city’s government. And as Menelaos says in Sophocles’ Ajax:107 The laws would never be accepted as they should in the city, if fear did not exist, nor could the army be directed wisely, without having fear and shame as a bulwark (µηδὲν φόβου πρόβληµα µηδ’ αἰδοῦς ἔχων).

In addition, Phobos can be used jointly with the law in order to guarantee specific conduct: for example, when Lysander sent the spoils to Sparta after the surrender of Athens in 404 BCE, Gylippos, the person in charge of conveying it, embezzled a part thereof. The Spartans then decided to prohibit gold and silver coins and ‘set up phobos and the law as sentinels’ (τὸν φόβον ἐπέστησαν φύλακα καὶ τὸν νόµον) to prevent the money from being introduced into the citizens’ homes.108 8 PHOBOS AND GELŌS Φόβος κενός, ‘empty fear’, could be taken for an anxiety before the unknown, fear without a determined object, disquiet. In speaking about our modern conception of fear, Jean Delumeau differentiates between fear of something, to which emotions like fear, fright, and terror belong, and anguish, that is, the expectation before undetermined danger, to which worry, anxiety, and melancholy belong.109 Our sources do not make this distinction, however. The majority of comparisons between adult fears and childish fears draw a parallel between an adult’s fear before something specific (enemies, government, punishment) and children’s fears before bogeymen. Yet, a bogeyman does not need to be present in order to inspire fear. It is likely to come, like divine punishments, like death, and that is enough. The fear is of something, whether or not it is based on reality.110 It is this reality, 106 Polybios 6.56.6–12. 107 Sophocles, Ajax 1073–1076. Cairns 1993, 235f., considers this passage, highlighting the utility of αἰδώς and φόβος for the state, as the reflection of contemporary thinking on civic life and the need of justice within the community. 108 Plutarch, Lysandros 17.10. 109 Delumeau 1978, 30. 110 In addition, one could also try to protect oneself from fear on an individual level: for example, an undated amulet from Sicily or Italy with a representation of Anubis was intended to protect against demons and fears (φόβους): IG XIV.2413.8; Bonner 1950, 95. Another example is an amulet representing Harpocrates, whose bearer requests that nothing should frighten him or her (〈µηθ〉ὲ〈ν〉 〈φ〉οβείτω µε): IG XIV.2413.11. For a recipe for making an amulet against various enemies including fears (φόβους), contained in a papyrus of the fourth or fifth century CE, cf. Kenyon 1893, 122 (pap. 124.25). In the Orphic Hymn to Night 3.14, ed. Quandt, the goddess is asked to dismiss the ‘fears that shine in the night’ (φόβους νυχαυγεῖς), while Korybas (39.3, ed. Quandt) is described as ‘he who makes terrible fears cease’ (φόβων ἀποπαύστορα δεινῶν), and Pan (11.7, ed. Quandt) as ‘he who terrifies mortals with fears’ (φόβων ἔκπαγλε βροτείων). According to Dieterich 1891, 89f., the φόβοι of the Orphic hymns, like those of the amulet representing Anubis, were spirits or phantoms analogous to the nocturnal φόβος of Psalm 90 (οὐ φοβηθήσῃ ἀπὸ φόβου νυκτερινοῦ). In our

Reflections on the Discourse of Fear in Greek Sources

127

the reality of danger represented by the object of fear, which is called into question by the sources.111 When the evaluation of the danger is false or overestimated, the φόβος experienced is κενός; the fear is baseless. And precisely therein lies the comic effect of baseless fear ridiculed by authors. According to Demetrios, it is the comic effect arising from empty fear (διὰ κενῆς) experienced by someone before a strap that he takes for a serpent.112 There is actually an illustration of this baseless fear, engendering laughter on a red-figure oinochoe, dated to around 425 BCE, on which a boy holding a satyr-mask pursues another boy who recoils before him.113 It is a children’s game, wherein one is amused by frightening the other with a dreadful object, a game that causes laughter. Its parallel in the adult world is the manipulative use of fear and the derisive attitude towards those who experience it. In comedy, there are numerous examples of baseless fear provoked by a bogeyman, which engenders great laughter in the spectator. A freeing laughter, laughter at one’s own childhood memories of anguish, and laughter at one’s fear – perhaps there is here another explanation for this habit of turning fear into a joke. But derisive laughter is what one emits when one recognises one’s superiority against the infirmity of others, as Thomas Hobbes argues.114 Recall that, beyond the cult of Phobos, the Spartans rendered cult to Laughter, as well. Plutarch mentions Gelōs at the same time as Phobos, and Gelōs could be a companion abstraction (like Nikē accompanying Athena). Indeed, laughter is an instrument of fear: it is through the fear of ridicule that a Spartan maintains normative conduct.115 In the same way, laughter also serves Aidōs, an abstraction also venerated at Sparta:116 it is through fear of dishonour and, in fact, of the laughter of others that a Spartan behaves in accordance with the laws. The laughter of mockery happens at the ex-

111

112 113 114 115

116

opinion, φόβος could designate here both what provokes the emotion and the emotion itself, just as in the same period and later (in the Byzantine period), the terms ἀπάντηµα, συνάντηµα, etc. refer to the encounter with the evil spirit as well as to the spirit itself (for an example of this use, cf. Kaimakis 1976, Περὶ ὄνου 20f.; for Byzantine use, cf. Delatte and Josserand 1934, 212). The Epicureans were apparently the only ones who distinguished the fear of the known from that of the unknown, saying that fear of death always lurks in us (cf. Konstan 2006, 149). But fear of death comes from false opinions, erroneous evaluations, and fear can be battled with reasoned argumentation: cf. Warren 2004, 6–16. Demetrios, On Style 159, ed. Chiron. Mylonas 1975, II 80–82, III pl. 362c. This vase was found in a child’s tomb from the west necropolis of Eleusis. Hobbes 1651, 36 (chapter I, §6). Plutarch, Kleomenes 9.1. In this passage, Plutarch alludes to the sanctuaries dedicated to Phobos and Gelōs, but in Lykourgos 25.4, he speaks only of a small statue of Gelōs. On the political and social implications of laughter in Spartan society, cf. David 1989. Aidōs is the only important female abstraction honoured by the Spartans; represented with a statue (Pausanias 3.20.10f.), Aidōs was an emotion which concerned men as well as women: cf. Richer 1999, 93–96 (following Cairns 1993, passim).

128

Maria Patera

pense of someone and humiliates its victim.117 Thus, through laughter individuals control one another.118 Actually, the notions of Aidōs (as fear, the objects of which are customs and rules) and Phobos resemble one another, at least in the case of Sparta,119 and they serve, with the notion of Gelōs, to strengthen social order.120 The ephor Chilon (around 556/5 or 555/4) is said to have prescribed, ‘When one is strong, one must be gentle in order to be respected by one’s neighbours rather than feared’ (ὅπως οἱ πλησίον αἰδῶνται µᾶλλον ἢ φοβῶνται). Phobos is thus present in society. Plato believes that the legislator should hold fear (φόβος) in great esteem, by calling it respect, fear of public opinion, shame (αἰδώς, αἰσχύνη); in addition, he describes αἰδώς/αἰσχύνη as divine fear (θεῖον φόβον).121 ‘La crainte du déshonneur est ce sur quoi la cité peut le plus utilement compter.’122 The Athenians themselves, in Thucydides, acknowledge that their empire is founded on fear (δέος), but proclaim that they are afraid of no one.123 Furthermore, it is fear (φόβος) of Athenian might that persuades the Lacedaemonians to engage in war: an abstraction venerated at Sparta becomes, according to Thucydides, the principal driver of its foreign policy.124 Still, for Thucydides, fear assures balance in relations between states, and he also believes that ‘the only sure guarantee of an alliance is an equilibrium of fear’ (τὸ δὲ ἀντίπαλον δέος µόνον πιστὸν ἐς ξυµµαχίαν).125 Therefore, equilibrium of forces is tantamount to equilibrium of fear.126 According to Polybios, in Sparta, fear constitutes a factor of political balance:127

117 On the laughter of mockery, cf. Sommerstein 2000, 68. Even the gods can be victims of laughter: in Aristophanes, Frogs 480f., Xanthias tells Dionysos – who soiled his pants out of fear – to get up before a stranger sees him. Here, laughter is addressed not only to an unseemly experience of fear (Dionysos behaves more timidly than his servant), but also to its physical results (in comedy it is common to say that someone farts or defecates out of fear): cf. Henderson 1991, 195f., no. 424; Taillardat 1965, 163, § 309. For another physical symptom of fear, trembling, cf. the article by Douglas Cairns in this volume (pp. 85–105). 118 Cf. Plutarch, Lykourgos 14.5, 17.1, 25.3. On ‘shame cultures’, wherein the rules must be internalised and applied under pain of exclusion by the community, cf. Dodds 1951, 17f. Contra: Cairns 1993, 27–47, who believes that the distinction between ‘shame-culture’ and ‘guiltculture’ is artificial. 119 Richer 1999, 96f. 120 Cf. David 1989, 2, 14, 17; Mactoux 1993, 285; Richer 1999, 107 note 68. 121 Plato, Laws 1.647a, 671d. Cf. Richer 1999, 108 note 78; Cairns 1993, 374f. 122 De Romilly 1958, 113. 123 Thucydides 1.75.3 and 5.111.1 (cf. also 1.76.2, 6.83.4, 7.63.3). Yet, it is because of the fear (φόβῳ) that the Athenians feel towards them that some peoples remain independent (Thucydides, 5.97). 124 Thucydides 1.23.6, 1.33.3, 1.88. Cf. Huart 1968, 135; Richer 1998b, 223 note 28. It is also out of fear that the allied cities gather around Sparta: cf. Thucydides 1.123.1, 2.8.5, 5.11.1. 125 Thucydides 3.11.2. Cf. Huart 1968, 344. 126 Cf. de Romilly 1956, 125. 127 Polybios 6.10.8f.

Reflections on the Discourse of Fear in Greek Sources

129

The monarchy was kept from arrogance through fear of the people (διὰ τὸν ἀπὸ τοῦ δήµου φόβον) and the people, for their part, would not dare to despise the kings out of fear of the gerontes (διὰ τὸν ἀπὸ τῶν γερόντων φόβον).

9 PHOBOS IN THE CITY If fear is such a characteristic of the very life of the city, it is because fear is an emotion intimately linked to discourse, a central element in Greek civic organisation. In his Praise of Helen, Gorgias remarks that the power of discourse can just as much provoke fear as put an end to it.128 A large part of Aristotle’s reflections on the emotions in general and on fear in particular come from his work on rhetoric.129 Of course, πειθώ, the art of persuasion, pertains par excellence to discourse.130 But the tendency to see emotions as closely associated with discourse in general and with rhetoric in particular derives from the manner in which emotions were comprehended in daily life and from people’s attitudes toward them, which rhetorical manuals confirmed and examined in depth.131 Indeed, when it comes to human affairs and not to emotions provoked by objects exceeding our limits, the majority of sources refer to discourse in its relation to fear. Even when fear is caused by a natural phenomenon like an eclipse, one can make it stop by using an argumentative discourse. Legislators, statesmen, warlords, nannies, and playwrights use discourse in order to excite or to quell fear.132 Later, Cicero mentions 128 Gorgias, F 8 and 14 ed. Radermacher. 129 On the manner in which orators should use emotions on the one hand and argumentation on the other in order to persuade their audience in Aristotle, cf. Cooper 1996. Aristotle, On the Soul 3.3.428a22–24, believes that δόξα, opinion or belief, pertains to the sphere of conviction, πίστις (because one cannot have an opinion without believing it), and conviction pertains to the sphere of persuasion, πειθώ, which itself falls under reason (λόγος): cf. Sihvola 1996, 118. According to Sorabji 1996, 317, this passage in Aristotle may have been poorly understood: every opinion is not necessarily formed following external persuasion; it cannot here be rhetorical πειθώ. Aristotle includes persuasion that comes from oneself. 130 Occasionally, πειθώ is a form of persuasion that does not allow contradiction, the free exercise of λόγος, and resembles φόβος. Πειθώ uses φόβος when it is based on presupposed knowledge of divine matters, for example, on the art of divination: ‘Through terms of evil (κακῶν) their wordy arts bring men to know fear (φόβον) chanted in prophetic strains’ (Aeschylus, Agamemnon 1130–1135, trans. W. H. Smyth); ‘He who brags about knowing divine matters knows nothing more than how to persuade (πείθειν) with speech’ (Euripides, F 795, 4f.). In fact, fear of the gods is, along with desire and veneration, one of the modes of relating to the divine: cf. Plutarch, Aristides 6.4. On the fear of the gods, cf. Chaniotis 2012. The power of discourse and persuasion is also underscored by Gorgias, F 11–14 ed. Radermacher, who speaks of bad persuasion, with which one poisons and bewitches the soul (οἱ δὲ πειθοῖ τινι κακῆι τὴν ψυχὴν ἐφαρµάκευσαν καὶ ἐξεγοήτευσαν). 131 Konstan 2007, 417. 132 For the use of theatricality by statesmen and by historians of the Hellenistic period in order to elicit precise emotions from their respective publics, see the study of Angelos Chaniotis in this volume (pp. 63–73).

130

Maria Patera

the orator’s power to instil fear in the heart of people by making use of the dangers that apply either to themselves or to the community.133 Fear is thus associated with several elements of the city: with reasoned discourse, with law, with the education of citizens, and with power relations. This emotion therefore shows itself to be a constitutive part of civic life, largely defining the relationships that men sustain among themselves, but also with civic institutions, the law, and the gods.134 Thus, from the stories of children’s bedrooms to philosophical writings, Greek thinking about fear appears remarkably coherent: the same notions traversed Greeks minds, from the legislator to the nanny. Rhetorical uses of fear are a matter of manipulation, of the people as well as of children. The response to fear, ‘les chemins pour sortir du pays de la peur’,135 is also a matter of discourse. The appeal to rationalisation, with or without the appeal to dignity and honour, is accompanied by laughter. Turning fear into something grotesque and ridiculous is perhaps the most effective way to rid oneself of it; laughter frees us from psychologically stressful situations like those caused by the feeling of fear. All this leads to a ‘mentality’ of fear, conceived as deeply significant to human life and, consequently, to civic life.136 Fear is found at the foundation both of the city and of the constitution of the community and the collective. The masses feel fear as if they were a single man. And yet, it is one single man the one who does not feel fear: the individual who distinguishes himself from others, a man of superior courage and intelligence. In the sources, we also note this distancing in the face of emotion, of which the authors make use in order to distinguish themselves from children and similar groups (crowds, women, old women), but at the same time to identify themselves with the superior man, who demonstrates his superiority through this dissociation from the emotions of the crowd. And this identification with the sagacious, courageous, and virtuous man serves in posing as such a man, in appearing as such before one’s public. Ultimately, it is always the gaze of others that counts and, in the end, distinguishing oneself from cowards is a matter of φόβος ἀδοξίας, fear of losing one’s reputation – whether one is a warrior or a writer.137 133 Cicero, De oratore 2.209. This distinction between fear for oneself and fear for the community lacks an equivalent in Aristotle and the Stoics: cf. Wisse 1989, 288f. 134 On the use of the fear of the gods for the maintenance of a law created by men, cf. Dörrie 1974, 259f.; Chaniotis 2012, 206f., 210f., 221f. See also Aristotle, Metaphysics 12.1.1074b4f. 135 Delumeau 1978, 41. 136 Rosenbloom 1993, 187f., introduces a difference between the functioning of fear at Sparta, which he considers as a tool of social control (Vernant 1962, 63, discusses Phobos at Sparta as fear ‘qui courbe dans l’obéissance tous les citoyens’), and its functioning at Athens as it appears in tragedy. At Athens, fear functioned as an instrument of civic cohesion, an instrument that saved the city at Marathon and Salamis, and was transformed thereafter through tragedy into a mechanism of communal memory; fear combined with shame was essential for keeping the community free, integral, and powerful. 137 Αἰδὼς ἐν ὀφθαλµοῖς, ‘shame resides in the eyes’, says the proverb (cf. Diogenianus, Centuria 1.69, eds. Leutsch/Schneidewin) that Aristotle, Rhetoric 2.6.1384a31–1384b1, explains with the fact that one feels more shame when one finds oneself before the gaze of others.

Reflections on the Discourse of Fear in Greek Sources

131

When it is mentioned, what fear implies is often affective manipulation, inasmuch as it is a powerful catalyst of decision-making.138 The individual who exerts his influence over the crowd possessed by its emotions is the man who controls his emotions and those of others, like the playwright who can stage fear in order ultimately to purify it with the use of reason. As a result, when it comes to fear, we are dealing with control. Control exerted over oneself and others is what defines an individual as superior. What is implemented in the civic and political domain pertains to that of the psychological. In Stoic doctrine, the wise man does not consent to fear, since only the common man surrenders to his initial impressions.139 Wise or not, the superior man distinguishes himself from the masses first through mastery of his own emotions, and then, strengthened by this knowledge, through manipulation of others’ fear.140 It is therefore a matter of managing fear. And it is also a matter of managing the event: one must first recognise the event as a sign of a nearby danger (for example, a prodigy) before reacting against it by inciting the civic assembly.141 On the subject of Homeric warriors, Nicole Loraux writes,142 La peur est la chose du monde la mieux partagée. … Au-delà de l’un et de l’autre, de celui qui l’incarne comme de celui qui l’éprouve, phobos est ce lien qui enchaîne au terrifiant le terrifié.

Let us add that this also applies in peacetime. By and large, the functioning of the city is founded on fear, in its form of government and in the relationship between the crowd and the leaders, who persuade the crowd through fear, or govern it through it, like tyrants. Fear binds the terrifying to the terrified, and it binds together the city which claims, like Athens, to have overcome it.143 Therefore, whether it is used directly or indirectly, fear seems, in the view that the Greeks have of it, to be constitutive of the very condition of the citizen, to varying degrees according to the model of government followed. We should also point out this interesting paradox: that authors, be they philosophers or not, dissociate themselves from the masses’ fear even as they recognise that fear is very useful for good governance of the city. Certainly, and in spite of its ‘constructive’ side, we still prefer the fears that other people feel.

138 Borgeaud 2007, 192. 139 Epiktetos, F 9 ed. Schenkl (apud Aulus Gellius, Attic Nights 19.1.17–20). See also Gill 2006, 251, and above, p. 118 note 52. In fact, for the Stoics, all emotions come from false opinion about what is actually good or bad; this is why the wise man does not experience emotions: cf. Brennan 1998, 47. What the Stoics did not acknowledge is the temporal dimension of the past, the manner in which past events influence the emotions of the present; cf. Nussbaum 2001, 177. 140 Ultimately, the only permitted fear is that before the epiphany of the gods (cf. Borgeaud 2007, 216), and even this fear is sometimes ridiculed (Lucian, Philopseudes 22–24). 141 Cf. Borgeaud 2007, 220. 142 Loraux 1989, 93 and 98. 143 Loraux 1989, 107.

132

Maria Patera

BIBLIOGRAPHY Africa, T. W. (1960) Phylarchus and the Gods: The Religious Views of an Hellenistic Historian, Phoenix 14, 222–227. Arena, R. (1989) Iscrizioni greche arcaiche di Sicilia e Magna Grecia. Iscrizioni di Sicilia I. Iscrizioni di Megara Iblea e Selinunte, Milan. Aubonnet, J. (1986) Aristote, Politique. Tome III, 1. Texte établi et traduit (Collection des Universités de France), Paris. Bernert, E. (1949) Phobos, Realencyclopädie der classischen Altertumswissenschaft 20/1, 309– 318. Bloch, R. (1963) Les prodiges dans l’antiquité classique, Paris. Borgeaud, P. (1979) Recherches sur le dieu Pan, Geneva. ––– (2007) Rites et émotions. Considérations sur les mystères, in Rites et croyances dans les religions du monde romain (Entretiens Hardt sur l’antiquité classique, 53), Vandœuvres/Geneva, 188–222. Brennan, T. (1998) The Old Stoic Theory of Emotions, in Sihvola and Engberg-Pedersen (eds.) 1998, 21–70. Cairns, D. L. (1993) Aidōs. The Psychology and Ethics of Honour and Shame in Ancient Greek Literature, Oxford. Chaniotis, A. (2012) Constructing the Fear of Gods: Epigraphic Evidence from Sanctuaries of Greece and Asia Minor, in id. (ed.), Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, Stuttgart, 205–234. Cooper, J. M. (1996) An Aristotelian Theory of the Emotions, in A. Oksenberg Rorty (ed.), Essays on Aristotle’s Rhetoric, Berkeley/Los Angeles/London, 238–257. David, E. (1989) Laughter in Spartan Society, in A. Powell (ed.), Classical Sparta: Techniques behind her Success, London/Norman, 1–25. Delumeau, J. (1978) La peur en Occident, Paris. Delatte, A. and C. Josserand (1934) Contribution à l’étude de la démonologie byzantine, in Mélanges Bidez I, Annuaire de l’Institut de philologie et d’histoire orientales 2, Brussels, 207– 232. Détienne, M. (1968) La phalange: problèmes et controverses, in J.-P. Vernant (ed.), Problèmes de la guerre en Grèce ancienne, Paris, 119–142. Deubner, L. (1902) Phobos, Mitteilungen des Deutschen Archäologischen Instituts (Athener Abteilung) 27, 253–264. Dieterich, A. (1891) Abraxas. Studien zur Religionsgeschichte des spätern Altertums, Leipzig. Dodds, E. R. (1951) The Greeks and the Irrational, Berkeley/Los Angeles. Dörrie, H. (1974) Polybios über pietas, religio und fides (zu Buch 6, Kap. 56), in Mélanges de philosophie, de littérature et d’histoire offerts à Pierre Boyancé (Collection de l’École française de Rome 22), Rome. Ellinger, P. (1993) La légende nationale phocidienne. Artémis, les situations extrêmes et les récits de guerre d’anéantissement (Bulletin de Correspondance Hellénique, Supplément, 27), Paris. Epps, P. H. (1933) Fear in Spartan Character, Classical Philology 28, 12–29. Fitzgerald, J. Z. (ed.) (2008) Passions and Moral Progress in Greco-Roman Thought, London/New York. Fortenbaugh, W. W. (2008) Aristotle and Theophrastus on the Emotions, in Fitzgerald (ed.) 2008, 29–47. Gill, C. (2006) The Structured Self in Hellenistic and Roman Thought, Oxford. Golden, M. (1990) Children and Childhood in Classical Athens, Baltimore/London. Gray, V. J. (2000) Xenophon and Isocrates, in Rowe et al. (eds.) 2000, 142–154. Guelfucci, M. R. (1986) La peur dans l’œuvre de Polybe, Revue de Philologie 60/2, 227–237. Halliwell, S. (1986) Aristotle’s Poetics, Chicago (second edition: 1998). Heberdey, R. and A. Wilhelm (1896) Reisen in Kilikien (Wiener Denkschriften 44/6), Vienna.

Reflections on the Discourse of Fear in Greek Sources

133

Henderson, J. (1991) The Maculate Muse: Obscene Language in Attic Comedy, New York/Oxford. Hobbes, T. (1651) Leviathan or the Matter, Forme, and Power of a Commonwealth Ecclesiastical and Civill, London. Huart, P. (1968) Le vocabulaire de l’analyse psychologique dans l’œuvre de Thucydide, Paris. Kaimakis, D. (ed.) (1976) Die Kyraniden, Meisenheim am Glan. Kenyon, F. G. (1893) Greek Papyri in the British Museum, London. Knuutila, S. and J. Sihvola (1998) How the Philosophical Analysis of the Emotions was Introduced, in Sihvola and Engberg Pedersen (eds.) 1998, 1–19. Konstan, D. (2006) The Emotions of the Ancient Greeks. Studies in Aristotle and Classical Literature, Toronto/Buffalo/London. ––– (2007) Rhetoric and Emotion, in I. Worthington (ed.), A Companion to Greek Rhetoric, Malden/Oxford/Victoria, 411–425. Loraux, N. (1989) Les expériences de Tirésias. Le féminin et l’homme grec, Paris. Mactoux, M.-M. (1993) Phobos à Sparte, Revue de l’Histoire des Religions 210/3, 259–304. Manganaro, G. (1995) L’elaphos di oro dedicato dai Selinuntini nell’Apollonion (IG XIV, nr. 268), Zeitschrift für Papyrologie und Epigraphik 106, 162–164. Manni, E. (1975) Da Megara Iblea a Selinunte: le Divinità, Kokalos 21, 174–195. Mylonas, G. E. (1975) Τὸ δυτικὸν νεκροταφεῖον τῆς Ἐλευσίνος, Athens. Nehamas, A. (1994) Pity and Fear in the Rhetoric and the Poetics, in D. J. Furley and A. Nehamas (eds.), Aristotle’s Rhetoric. Philosophical Essays, Princeton, 257–282. Nicolaïdou-Kyrianidou, V. (2006) La cité des frères: la polis parfaite de Platon et la Jérusalem du christianisme, in A. Bresson, M.-P. Masson, S. Perentidis, and J. Wilgaux (eds.), Parenté et société dans le monde grec, Bordeaux, 237–269. Nussbaum, M. (1996) Aristotle on Emotions and Rational Persuasion, in A. Oksemberg Rorty (ed.), Essays on Aristotle’s Rhetoric, Berkeley/Los Angeles/London, 303–323. ––– (2001) Upheavals of Thought. The Intelligence of Emotions, Cambridge. Patera, M. (2005) Comment effrayer les enfants: le cas de Mormô/Mormolukê et du mormolukeion, Kernos 18, 371–390. Richer, N. (1998a) Des citoyens maîtres d’eux-mêmes: l’eukosmon de Sparte archaïque et classique, Cahiers du Centre Gustav Glotz 9, 7–36. ––– (1998b) Les Éphores. Études sur l’histoire et sur l’image de Sparte (VIIIe–IIIe siècles avant Jésus-Christ), Paris. ––– (1999) Aidôs at Sparta, in A. Powell and S. Hodkinson (eds.), Sparta: New Perspectives, London/Swansea, 91–115. Robert, L. (1981) Amulettes grecques, Journal des Savants, 3–44. Roberts, J. (2000) Justice and the Polis, in Rowe et al. (eds.) 2000, 344–365. Romilly, J. de (1956) La crainte dans l’œuvre de Thucydide, Classica et Mediaevalia 17, 119–127. ––– (1958) La crainte et l’angoisse dans le théâtre d’Eschyle, Paris (second edition: 2011). ––– (1971) La loi dans la pensée grecque des origines à Aristote, Paris. Rosenbloom, D. (1993) Shouting ‘Fire’ in a Crowded Theater: Phrynichos’s Capture of Miletos and the Politics of Fear in Early Attic Tragedy, Philologus 137, 159–196. Rowe, C., M. Schofield, S. Harrison, and M. Lane (eds.) (2000) The Cambridge History of Greek and Roman Political Thought, Cambridge. Schadewaldt, W. (1955) Furcht und Mitleid?, Hermes 83, 129–171. Serghidou, A. (2007a) Les deux temps de la peur. Crainte immédiate et peur d’asservissement prospectif. Le cas d’Hérodote, in Serghidou (2007b), 217–230. ––– (ed.) (2007b) Fear of Slaves – Fear of Enslavement in the Ancient Mediterranean, Peur de l’esclave – Peur de l’esclavage en Méditerranée ancienne, Discours, représentations, pratiques (Actes du XXIXe colloque du Groupe International de Recherche sur l’Esclavage dans l’Antiquité [GIREA], Rethymnon, 4–7 novembre 2004), Besançon. Shapiro, H. A. (1993) Personifications in Greek Art. The Representation of Abstract Concepts, 600–400 B.C., Zurich.

134

Maria Patera

Sihvola, J. (1996) Emotional Animals: Do Aristotelian Emotions Require Beliefs?, Apeiron 29/2, 105–144. Sihvola, J. and T. Engberg-Pedersen (eds.) (1998) The Emotions in Hellenistic Philosophy, Dordrecht/Boston/London. Sommerstein, A. (2000) Parler du rire chez Aristophane, in M.-L. Desclos (ed.), Le rire des Grecs. Anthropologie du rire en Grèce ancienne, Grenoble, 67–75. Sorabji, R. (1996) Rationality, in M. Frede and G. Striker (eds.), Rationality in Greek Thought, Oxford, 311–334. Taillardat, J. (1965) Les images d’Aristophane. Études de langue et de style, second, revised edition, Paris (first edition: 1962). Taylor, C. C. W. (2000) Democritus, in Rowe et al. (eds.) 2000, 122–129. Thalmann, W. G. (2007) Despotic Authority, Fear, and Ideology of Slavery, in Serghidou (ed.) 2007b, 193–205. Vernant, J.-P. (1962) Les origines de la pensée grecque, Paris (reprint: 1995). Warren, J. (2004) Facing Death. Epicurus and his Critics, Oxford. Wide, S. (1893) Lakonische Kulte, Leipzig (reprint: Stuttgart, 1973). Winston, D. (2008) Philo of Alexandria on the Rational and Irrational Emotions, in Fitzgerald (ed.) 2008, 201–220. Wisse, J. (1989) Ethos and Pathos from Aristotle to Cicero, Amsterdam. Zaborowski, R. (2002) La crainte et le courage dans l’Iliade et l’Odyssée. Contribution lexicographique à la psychologie homérique des sentiments, Warsow. ––– (2004) Le vocabulaire des sentiments dans l’Iliade, Πρακτικὰ ΙΑ΄ Δ∆ιεθνοῦς Συνεδρίου Κλασσικῶν Σπουδῶν (Καβάλα, 24–30 Αὐγούστου 1999), Athens, III 565–575.

EVOKING ANGER THROUGH PITY Portraits of the Vulnerable and Defenceless in Attic Oratory Lene Rubinstein 1 INTRODUCTION The Athenians were very open indeed about the impact that emotional appeals could have on the decisions reached in their popular courts. As is well known, Athenian litigants regularly appealed directly to the emotions of their dicastic audiences, with instructions to the effect that it was the duty of the judges to feel and express anger, pity, resentment, hatred, and gratitude.1 This is not to say, however, that the Athenians regarded the influence of dicastic emotions as unproblematic. In his prosecution of Aischines,2 Demosthenes asserts that Athenian judges are often ‘led astray’ (παράγεσθαι) by feelings of pity (ἔλεος), resentment (φθόνος), and anger (ὀργή). What is particularly interesting about Demosthenes’ criticism in the present passage is that he represents these dicastic emotions themselves as problematic. It is the emotions, rather than the litigants who appeal to them, that are the cause of the judges’ deafness and blindness, and that prevent them from forming a rational assessment of the issues before them. Demosthenes is not by any means the only Athenian litigant to represent dicastic emotions as a threat to the integrity of the decision-making process. There are several passages in the corpus of Athenian forensic oratory where we find litigants making disapproving comments on previous verdicts reached by the courts. In some of these instances, the alleged miscarriage of justice is represented as the result of the judges’ having permitted their emotions to override their rational assessment of the cases they were judging. Sometimes, their emotional response is portrayed as an understandable reaction to an immediate political crisis;3 but more often litigants preferred to explain the past error as the result of successful emotional manipulation by the victorious party and his supporters.4 There is yet another indication that the Athenian attitude to dicastic emotions was ambivalent: 1 2 3 4

On the arousal of hostile emotions in Attic oratory, see Sanders 2012. Demosthenes 19.226–228. Cf. also Demosthenes 18.277f. and 26.17f. E.g., Antiphon 5.69–71. E.g., Demosthenes 37.47, 45.6f., 58.31f. These passages will be further discussed below. Compare also the general statement made in Lysias 19.6, where the speaker claims that when several defendants are tried on the same charge, those whose cases are heard last tend to be acquitted, because ‘once you have stopped being angry, you listen, and you now willingly accept their refutations’.

136

Lene Rubinstein

the courts are sometimes commended for having imposed the most severe penalties on guilty defendants, in spite of the fact that the defendants in question were arguably deserving of sympathy or gratitude or both.5 In these instances, the ability of the judges to resist appeals to their emotions is represented in a positive light, and the present dicastic audience is asked to follow their example – the main message being that, unlike the men previously sentenced, the present defendant is completely undeserving of any dicastic pity or generosity. Recent modern scholarship, including my own, has tended to focus on the passages in which litigants explicitly solicit a particular emotional response from their audiences, with a clearly stated expectation that the feeling thus evoked will inform the judges’ decision in the case at hand. Such open appeals present the advantage that the litigants define and name the type of emotional reaction expected from the audience in response to particular types of behaviour or personal characteristics displayed by the speakers themselves and by their opponents. These appeals undoubtedly provide important evidence for the historian who attempts to identify the triggers of particular emotions in a specific cultural context. On the other hand, there is a real risk that studies that concentrate predominantly on the explicit soliciting of emotional responses may represent only part of a much larger picture. Perhaps partly because the Athenians clearly were ambivalent when it came to the issue of dicastic emotions and the extent to which they should inform the response of the court, Athenian litigants often appear to have exercised a certain degree of caution when openly asking the judges to respond emotionally to the case they were pleading, opting instead for more indirect methods of emotional appeal. In his book Disputes and Democracy, Stephen Johnstone has shown that direct verbal appeals to dicastic pity are employed only relatively infrequently in prosecution speeches both private and public, and that they are totally absent from the extant speeches delivered in diadikasiai in which litigants advanced their inheritance claims.6 Adopting a similar method, I have argued that a similar uneven distribution can be discerned when it comes to explicit calls for the dicasts to feel and display their anger against the litigant’s opponent.7 Only two defendants, both involved in public actions, ask for the judges’ ὀργή against the prosecutors responsible for bringing them to trial on what is allegedly a baseless charge.8 What is more, open appeals to dicastic anger are totally absent from private defence speeches and speeches delivered in diadikasiai. When it comes to speeches delivered by prosecutors, there is likewise a significant difference between the tactics employed in private and public actions. While most prosecutors in public actions instruct their audiences to feel anger, such instructions are issued only infrequently by plaintiffs who had brought their complaints under the heading of a private action. 5 6 7 8

E.g., Aischines 3.195; Demosthenes 19.280f.; Deinarchos 1.14. Johnstone 1999, 111–120. Rubinstein 2004. Andokides 1.24; Hypereides 3 (For Euxenippos) 17.

Evoking Anger through Pity in Attic Oratory

137

It is more than likely that these differences reflect unwritten conventions, and that the litigants’ direct appeals to dicastic emotions were to some extent constrained by the nature of the case and the type of procedure in which they were involved. On the other hand, if court etiquette may have imposed certain limitations on the overt verbal appeals to dicastic emotions in a variety of contexts, this would not have prevented the litigants from attempting to stir up feelings of anger, resentment, fear, jealousy, and pity by indirect means, through their narratives and argumentation. In such cases, the emotions that a litigant hoped to generate in his audience would not be identified directly. The emotional impact of the more subtle approach may in fact have been greater, precisely because the litigant in question refrained from instructing the judges on what they ought to feel. One of the most famous examples of the indirect method having met with success is Demosthenes’ speech For Phormion (36). This was delivered as a supporting speech, which challenged the admissibility of a legal action brought by Apollodoros, the son of Pasion, concerning outstanding debts allegedly owed by Phormion to Pasion’s bank. While the speaker openly asks the judges to show ἔλεος towards Phormion as the victim of a malicious prosecution (36.59), there are no explicit calls anywhere in the speech for the judges to display anger, hatred, or resentment against Apollodoros. Yet, it is certain that Phormion and his supporters succeeded in generating powerful negative feelings in their audience, and that their personal attack on Apollodoros was utterly devastating. As is well known, Apollodoros himself comments on this in his subsequent attack on Stephanos, one of Phormion’s witnesses. In Demosthenes 45.6, he complains that Phormion and his supporters brought the judges into such a state of agitation that they refused even to give him a hearing. Apollodoros does not label the emotions that he claims had swayed the court in Phormion’s favour, however. It is possible that the response was one primarily of hostility against Apollodoros but a hostility that was amplified by feelings of pity for Phormion as a victim of Apollodoros’ litigiousness and greed. Apollodoros would probably not have wanted to draw attention to the latter when he introduced his renewed attack on Phormion through one of his witnesses, with damages assessed at the enormous sum of one talent. Demosthenes’ speech For Phormion shows that the method of mapping out explicit emotional appeals in the surviving forensic speeches provides only a very partial guide to the way in which litigants attempted to arouse the feelings of their audiences. Any comprehensive study of the part played by dicastic emotions in the interaction between Athenian litigants and the judges, and in the legal decision-making process itself, requires an assessment of the indirect methods by which the speakers attempted to play on the feelings of their audiences. This, of course, may seem a trivial observation. However, the approach itself presents some extremely serious methodological problems. When reading a forensic speech, we may be acutely aware that the speaker is trying to hit some emotional buttons in his audience, but we cannot always be sure precisely what those buttons were. Precisely because the indirect emotional appeals do not name the actual feelings that the speakers are trying to evoke, there is considerable danger that our attempts to identify and label them will be informed in part by our own personal

138

Lene Rubinstein

and, to some extent, culturally determined response to the actions and characters as represented in forensic narratives. For a number of years, I have used Apollodoros’ speech Against Neaira as one of the set texts in my undergraduate course on Athenian law, and there is a recurrent pattern of student reactions to this speech, which is to some extent also a gendered response. The story told by Apollodoros is that of a woman who as a little girl was sold into child prostitution – Apollodoros informs us that she was made available to customers even before she had had her first period (59.22). She was subsequently bought by two of her clients, then manumitted and physically abused by Phrynion who was legally responsible for buying her freedom, subjected to humiliating sexual treatment in public, gang-raped at parties, and finally was the object of a settlement by which Phrynion and Stephanos agreed to have sex with her on alternate days. In my experience, female undergraduates especially respond with sympathy – some even with outrage – at Neaira’s plight. At the same time, we all know that this was almost certainly not how Apollodoros intended his dicastic audience to react to his narrative. Neaira is the defendant in this case, and Apollodoros’ aim is to persuade the judges to pass a verdict that will lead to Neaira’s being sold into slavery at a public auction. Chris Carey comments:9 It is in pathos that the strength of this speech resides, and much of the prosecution case is calculated to have an emotional effect. In the case of Neaira, Apollodoros seeks to arouse in the jurors a mixture of contempt and hostility, based primarily on her past life.

Carey’s interpretation of Apollodoros’ emotional strategy is shared by most modern commentators on the speech, in spite of the fact that the only explicit appeal to dicastic emotion of any kind – anger – is directed against Stephanos rather than Neaira. As far as Apollodoros’ narrative is concerned, it is actually surprisingly difficult to explain why Apollodoros expected his audience to react, not with pity or sympathy, but with negative emotions to the unfolding of Neaira’s story. Part of the answer is, undoubtedly, to be found in general Athenian attitudes to female prostitutes. On the other hand, it would be too simplistic to claim that Apollodoros and his audience were expected to view the abuse of Neaira with indifference because of her status as a non-Athenian, a former slave, and a woman. For it is remarkable that Apollodoros himself uses strong negative vocabulary about the treatment of Neaira by Phrynion (ὕβρις, ἀσελγέω, προπηλακίζειν) – and hybris, it must be noted, was a criminal offence even when committed against unfree non-Athenians.10 Nor can we find a satisfactory answer in the focalisation of Neaira’s story as told by Apollodoros. Much of it is related from Neaira’s perspective: we hear about her hopes, her expectations, her disappointment at not being loved, and her deep fear of Phrynion, which led her to look to Stephanos for protection. It is only in §41 of the narrative that Neaira is transformed from pas9 10

Carey 1992, 14. Omitowoju 1997, 10–12, has argued that Apollodoros deliberately refers to hybris only in his account of Neaira’s own account of her treatment to her ‘rescuer’ Stephanos, but see Fisher 2001, 189f.

Evoking Anger through Pity in Attic Oratory

139

sive victim into active perpetrator, and the only other active part assigned to her is when she attended to Phrastor on his sickbed in order to induce him to take back her pregnant daughter, Phano. While there can be no doubt at all from Apollodoros’ highly charged peroration that he expected his speech as a whole to arouse the judges’ hostility towards Neaira and her ‘husband’ Stephanos, it is indeed puzzling that Apollodoros constructed his main narrative mainly from Neaira’s point of view, when he could just as well have chosen more conventional methods of invective as is seen in countless other forensic speeches.11 In the Neaira case, then, it is the context of the speech as a whole and of Apollodoros’ acknowledged political agenda that plainly proves the response of a middle-aged female Scandinavian to Apollodoros’ narrative – and the response of a good many of my students – to be radically different from the one that Apollodoros seems to have expected from his Athenian audience. Furthermore, most modern analyses of Apollodoros’ emotional strategy are also informed by what we think we know about typical male Athenian attitudes to prostitutes, to the unfree, and to women generally. But even if these assumptions are correct (and I shall try to argue that matters are not necessarily as straightforward as that), we still face a problem when it comes to identifying precisely which negative emotions Apollodoros was trying to stir up in his audience – contempt, hatred, anger, disgust, fear, or a combination of all of them – and, equally importantly, identifying the triggers that were expected to produce the desired response as part of Apollodoros’ strategy. In such situations, it is often tempting to resort to Aristotle’s Rhetoric as a guide to what an Athenian litigant may have hoped would trigger a particular emotional reaction from the dicasts. But this text is not without its problems. First, it is far from certain that Aristotle’s list of triggers is an accurate, let alone comprehensive, guide to what may have moved an average Athenian citizen on a dicastic panel. Secondly, as is most evident in Aristotle’s treatment of anger, his analysis does not comment on how to generate anger as a response to behaviour that could not be depicted as directly affecting the judges themselves or the wider community which they represented. I have previously argued that the pattern of open appeals to dicastic anger do appear, by and large, to conform to Aristotle’s analysis.12 Here, however, it will be argued that it is far more difficult to apply it to the more subtle and indirect attempts to stir up negative emotions in the audience. Thirdly, Aristotle’s systematic treatment of emotions discusses each of them separately, without commenting on how litigants might be able to play on several different emotions simultaneously. And there is a further methodological problem that arises if we use Aristotle’s work as a guide and starting point in our attempt to 11

12

This problem has been highlighted and discussed by Kapparis 1999, 46f., who comments that ‘Apollodoros has misplaced the emphasis on her misfortunes, because this can raise sympathy and pity. He could have emphasised instead the moments when fighting for her survival, she used immoral or unlawful means. As a result of this Stephanos, speaking for the defence, would not need to try hard to secure the sympathy of the judges for her.’ Rubinstein 2004, 193f.

140

Lene Rubinstein

identify indirect appeals to emotions in the corpus of Attic oratory. Because the corpus is so vast and so diverse, there is a risk of selecting precisely those passages that appear to conform to Aristotle’s analysis, while passing over those which do not appear to fit. In what follows, I shall experiment with a different approach. The focus will be on invectives, concentrating in particular on the litigants’ portrayal of third parties, that is, people not involved directly in the legal disputes at hand, who are represented as victims of their opponents’ behaviour. The reason for adopting this method is, first, that it provides a way of limiting the material for analysis without selecting in advance the types of behaviour or victims that the litigants expected would trigger a particular kind of emotional response. Secondly, since the contexts in which these third party victims are represented are invariably intended as character assassinations, and therefore intended to generate a hostile response towards the opponent, they may serve to create a preliminary basis for an investigation of the methods by which litigants might use indirect appeals to dicastic pity as a way of amplifying hostile feelings, including feelings of anger, against their opponents. Some types of third party victim that I identified in my trawl through the corpus did match my initial expectations. They belong to categories of individuals that are often represented collectively not only as vulnerable but also as deserving of protection by the community as a whole. Other types, however, were more surprising. Among them are free and unfree non-citizens, prostitutes, a convicted murderer, in other words people with whom the dicasts would probably not be expected to identify, but whose ill-treatment could nevertheless be represented in the context of character assassination. It is not always clear to what extent the litigants were aiming to evoke feelings of dicastic pity through such representations – and sometimes it is clear from the focus and vocabulary used that they were not – but in other instances it seems certain that the speaker hoped that the judges would be moved by the victim’s plight and that this, in turn, might increase their feelings of anger, hatred, or resentment. 2 ANGER AND PITY: ORPHANS, WIDOWS, AND THE ELDERLY It is of course well known that orphans, widows, and the elderly were frequently deployed, especially by defendants in public actions, to stir up feelings of dicastic pity, often in open and explicit emotional appeals. Despite the frequency with which such appeals were employed, they do not appear to have been without effect, and prosecutors often attempted to pre-empt this tactic. One example is Apollodoros’ parting shot in his speech against Nikostratos concerning property that is allegedly liable to confiscation by the state:13 13

Demosthenes 53.29: ἐὰν οὖν ἐνθυµηθῆτε, ὅτι οὐδέποτε ἔσται ἀπορία τῶν ἀµφισβητησόντων ὑµῖν περὶ τῶν ὑµετέρων (ἢ γὰρ ὀρφανοὺς ἢ ἐπικλήρους κατασκευάσαντες ἀξιώσουσιν ἐλεεῖσθαι ὑφ’ ὑµῶν, ἢ γῆρας καὶ ἀπορίας καὶ τροφὰς µητρὶ λέγοντες, καὶ ὀδυρόµενοι

Evoking Anger through Pity in Attic Oratory

141

Thus, if you bear it in mind that there will never be a lack of people involved in disputes over your property (for they will ask to be pitied by you either by lining up orphans or epikleroi, or by mentioning old age and poverty and their mother’s maintenance, and by engaging in lamentation over those things through which they expect that they will especially deceive you they will try to deprive the polis of the debt) – if you then ignore all this and vote for a conviction, you will reach the correct decision.

That defendants might hope to arouse the emotion of pity by this means almost certainly reflects the fact that the interests and protection of these groups were widely perceived as a collective concern. According to Demosthenes in a highly tendentious statement,14 this concern extended even to the dependants of convicted criminals. However, pity was not the only kind of dicastic emotion that could be generated by this tactic. In two instances where litigants were commenting on alleged miscarriages of justice, they claim that the wrong verdict had been reached because the successful litigant had played the ‘orphan card’, and that the emotion that had swayed the judges had been one of anger. In both of these cases the orphans were, at least technically speaking, ‘third party victims’. One suit was a public action for unlawful decree proposal, a graphe paranomon (action for unlawful decree proposal), related in Demosthenes 58.15 The speaker’s father had

14

15

δι’ ὧν µάλιστα ἐλπίζουσιν ἐξαπατήσειν ὑµᾶς, πειράσονται ἀποστερῆσαι τὴν πόλιν τοῦ ὀφλήµατος), ἐὰν οὖν ταῦτα παριδόντες πάντα καταψηφίσησθε, ὀρθῶς βουλεύσεσθε. Demosthenes 27.65: καὶ ὑµεῖς µὲν οὐδὲ τῶν εἰς ὑµᾶς ἁµαρτανόντων ὅταν τινὸς καταψηφίσησθε, οὐ πάντα τὰ ὄντ’ ἀφείλεσθε, ἀλλ᾿ ἢ γυναῖκας ἢ παιδί᾿ αὐτῶν ἐλεήσαντες µέρος τι κἀκείνοις ὑπελίπετε (‘and you yourselves do not, even when you are convicting someone who has committed crimes against you, deprive them of all their possessions, but because you take pity on either their wives or their children, you leave even these men a part of their property’). Demosthenes 58.30–32: τοῦ γὰρ πατρὸς κατηγορῶν, ὦ ἄνδρες δικασταί, ὅτε τὴν τῶν παρανόµων αὐτὸν ἐδίωκε γραφήν, ἔλεγεν ὡς ἐπιβεβουλευµένος ὁ παῖς εἴη περὶ οὗ τὸ ψήφισµα γεγραµµένον ἦν, ἐν ᾧ τὴν σίτησιν ἔγραψεν Χαριδήµῳ ὁ πατὴρ τῷ Ἰσχοµάχου υἱῷ, λέγων ὡς, ἐὰν ἐπανέλθῃ εἰς τὸν πατρῷον οἶκον ὁ παῖς, ἀπολωλεκὼς ἔσται τὴν οὐσίαν ἅπασαν ἣν Αἰσχύλος ὁ ποιησάµενος αὐτὸν υἱὸν ἔδωκεν αὐτῷ, ψευδόµενος· οὐδενὶ γὰρ πώποτε, ὦ ἄνδρες δικασταί, τοῦτο τῶν εἰσποιηθέντων συνέβη. καὶ τούτων πάντων αἴτιον ἔφη Πολύευκτον γεγενῆσθαι τὸν ἔχοντα τὴν µητέρα τοῦ παιδός, βουλόµενον ἔχειν αὐτὸν τὴν τοῦ παιδὸς οὐσίαν. ὀργισθέντων δὲ τῶν δικαστῶν ἐπὶ τοῖς λεγοµένοις, καὶ νοµισάντων αὐτὸ µὲν τὸ ψήφισµα καὶ τὴν δωρεὰν κατὰ τοὺς νόµους εἶναι, τῷ δὲ ὄντι τὸν παῖδα µέλλειν ἀποστερεῖσθαι τῶν χρηµάτων, τῷ µὲν πατρὶ δέκα ταλάντων ἐτίµησαν ὡς µετὰ Πολυεύκτου ταῦτα πράττοντι, τούτῳ δ’ ἐπίστευσαν ὡς δὴ βοηθήσαντι τῷ παιδί (‘for when he was prosecuting my father in the graphe paranomon, he was making the accusation against him that the child, in whose honour the decree had been passed, in which my father had proposed sitesis [public dining rights] for Charidemos, son of Ischomachos, was the victim of a plot. He claimed that if the child returned to his natural father’s oikos he would lose the entire estate which Aischylos, who had adopted him, had given him, and he was lying. For, judges, this has never happened to any adoptee. And he said that Polyeuktos, who was married to the child’s mother, had been responsible for all this, because he himself wished to possess the child’s estate. Since the judges were angered by what was said and believed that while the decree itself and the honorific award were in accordance with the laws, but that in reality the child was about to be deprived of his property, they assessed a pe-

142

Lene Rubinstein

allegedly proposed a decree awarding dining rights in the Prytaneion to Charidemos, natural son of the late Ischomachos, and adopted orphaned son of Aischylos, probably his maternal grandfather. The decree was subsequently attacked as unlawful by the speaker’s opponent, Theokrines, who allegedly claimed that the honour would have meant not only that the orphan would have to return to the oikos (household) of his natural father, but also that he would lose the inheritance of his late adopted father. We must of course be sceptical towards the speaker's dismissal of Theokrines’ accusation, but it is indeed likely that there was a real grey area as far the orphan’s position was concerned.16 Whatever the legal basis of Theokrines’ claim, the speaker may well be right in his assertion that dicastic emotion, anger, had made a significant contribution to Theokrines’ victory, and, what is most important, he clearly expected his audience to regard his explanation as plausible. The other suit was a very complicated mining dispute where the plaintiff, Pantainetos, had asserted that his opponent, Euergos, had entered his house in the presence of the plaintiff’s mother and some epikleroi (heiresses) who were also residing there.17 Again, the emotion claimed to have been responsible for leading the judges to the wrong decision was ὀργή. The speaker, who is contesting the admissibility of a similar mining suit against himself, expects Pantainetos pres-

16 17

nalty of ten talents against my father on the grounds that he had been Polyeuktos’ accomplice, and trusted my opponent in the belief that he had indeed come to the rescue of the child’). See Cox 1998, 90. Demosthenes 37.45–48: βούλοµαι δ᾿ ὑµῖν καὶ δι’ ὧν τοὺς πρότερον δικαστὰς ἐξαπατήσας εἷλε τὸν Εὔεργον εἰπεῖν, ἵν’ εἰδῆθ’ ὅτι καὶ νῦν οὐδὲν οὔτ᾿ ἀναιδείας οὔτε τοῦ ψεύδεσθαι παραλείψει. πρὸς δὲ τούτοις καὶ περὶ ὧν ἐµοὶ δικάζεται νυνί, τὰς αὐτὰς οὔσας ἀπολογίας εὑρήσετε· ὅσπερ ἔλεγχος ἀκριβέστατός ἐστιν ὑπὲρ τοῦ τότ’ ἐκεῖνον σεσυκοφαντῆσθαι. οὗτος γὰρ ᾐτιάσατ᾿ ἐκεῖνον πρὸς ἅπασι τοῖς ἄλλοις ἐλθόντ’ εἰς ἀγρὸν ὡς αὑτὸν ἐπὶ τὰς ἐπικλήρους εἰσελθεῖν καὶ τὴν µητέρα τὴν αὑτοῦ, καὶ τοὺς νόµους ἧκεν ἔχων τοὺς τῶν ἐπικλήρων πρὸς τὸ δικαστήριον... ἡ δ᾿ ὀργὴ ἡ παρὰ τῶν ἐξηπατηµένων ὑπὸ τούτου δικαστῶν, ἐφ᾿ ᾧ τὴν ψῆφον εἶχον πράγµατι, τούτου κατεψηφίσατο. καίτοι τὸν ἐκείνους ἐξηπατηκότα τοὺς δικαστάς, ἆρ’ ὀκνήσειν ὑµᾶς ἐξαπατᾶν οἴεσθε; ἢ πεπιστευκότ᾿ εἰσιέναι τοῖς πράγµασιν, ἀλλ’ οὐ τοῖς λόγοις καὶ τοῖς συνεστῶσιν µεθ᾿ αὑτοῦ µάρτυσιν, τῷ τ᾿ ἀκαθάρτῳ καὶ µιαρῷ Προκλεῖ, τῷ µεγάλῳ τούτῳ, καὶ Στρατοκλεῖ τῷ πιθανωτάτῳ πάντων ἀνθρώπων καὶ πονηροτάτῳ, καὶ τῷ µηδὲν ὑποστελλόµενον µηδ’ αἰσχυνόµενον κλαήσειν καὶ ὀδυρεῖσθαι; (‘I also want to tell you by what means he obtained the conviction of Euergos by having deceived the previous panel of judges, so that you will know that now, too, he will not abstain from any type of shamelessness or lies. Moreover, you will find that the line of defence is the same also in the matter over which he is presently suing me. And this is the most certain proof that Euergos was the victim of sykophancy then. For my opponent accused him, in addition to all the other charges, of going to his farm and entering his house in the presence of the epikleroi and his mother, and he went to court armed with the laws concerning epikleroi. ... The anger felt by the judges who had been deceived by him led to Euergos’ conviction on the actual charge that was the basis of the law suit. Yet, do you think that the man who has succeeded in deceiving those judges will hesitate to deceive you? Or that he will enter court trusting in his actual case, rather than in his speech and the witnesses who have ganged up with him, the impure and defiled Prokles, that big man, and Stratokles, the most persuasive and wicked of all mankind, and in his weeping and lamenting without restraint or shame?’).

Evoking Anger through Pity in Attic Oratory

143

ently to resort to the same tactic (he has apparently repeated the allegation formally in his writ).18 Although the speaker does not inform us directly about the nature of Pant-ainetos’ emotional appeal that succeeded in generating dicastic anger against Euergos, his expectation that Pantainetos will presently resort to ‘wailing and lamentation without restraint or shame’ in order to repeat his success is a strong indication that Pantainetos’ tactic had been one of evoking dicastic pity in order to generate and amplify dicastic anger against the alleged perpetrators. A very instructive passage from Isaios 5 provides an example of how this might be done.19 There can be no doubt at all that the speaker’s aim is first and foremost to generate powerful hostile feelings against Dikaiogenes, the speaker’s maternal uncle by adoption. In fact, in the present case, it is not Dikaiogenes him18 19

Demosthenes 37.33. Isaios 5.10f.: ∆ικαιογένης δὲ πρὸς ἡµᾶς ὡς ἐβούλετο ἀγωνισάµενος τῇ αὐτῇ ἡµέρᾳ ἐξήλασε µὲν τὴν Κηφισοφῶντος τοῦ Παιανιέως θυγατέρα ἐκ τοῦ µέρους, ἀδελφιδῆν οὖσαν ∆ικαιογένους τοῦ καταλιπόντος τὰ χρήµατα, ἀφείλετο δὲ τὴν ∆ηµοκλέους γενοµένην γυναῖκα ἃ ∆ικαιογένης ἀδελφὸς ὢν ἔδωκεν, ἀφείλετο καὶ τὴν Κηφισοδότου µητέρα καὶ αὐτὸν τοῦτον ἅπαντα. Καὶ γὰρ τούτων [τε] ἅµα καὶ ἐπίτροπος καὶ κύριος καὶ ἀντίδικος ἦν, καὶ οὐδὲ κατὰ τὸ ἐλάχιστον µέρος τῆς οἰκειότητος ἐλέου παρ’ αὐτοῦ ἔτυχον, ἀλλ’ ὀρφανοὶ καὶ ἔρηµοι καὶ πένητες γενόµενοι πάντων καὶ τῶν καθ’ ἡµέραν ἐπιτηδείων ἦσαν ἐνδεεῖς. Οὕτως αὐτοὺς ∆ικαιογένης οὑτοσὶ ἐγγυτάτω ὢν γένους ἐπετρόπευεν· ὅς γε, ἃ µὲν ὁ πατὴρ αὐτοῖς Θεόποµπος κατέλιπε, τοῖς τούτων ἐχθροῖς παρέδωκεν, ἃ δὲ ὁ πρὸς µητρὸς θεῖος καὶ ὁ πάππος αὐτοῖς ἔδωκεν, αὐτὸς ἀφείλετο πρὸ δίκης. Καὶ ὃ πάντων δεινότατον, τὴν οἰκίαν αὐτῶν τὴν πατρῴαν, παίδων ὄντων τούτων, πριάµενος καὶ κατασκάψας, τὸν κῆπον ἐποιήσατο πρὸς τῇ αὑτοῦ οἰκίᾳ τῇ ἐν ἄστει. Καὶ λαµβάνων µίσθωσιν ὀγδοήκοντα µνᾶς ἐκ τῶν ∆ικαιογένους τοῦ ἡµετέρου θείου χρηµάτων, τὸν ἐκείνου ἀδελφιδοῦν Κηφισόδοτον τῷ ἑαυτοῦ ἀδελφῷ Ἁρµοδίῳ συνέπεµψεν εἰς Κόρινθον ἀντ’ ἀκολούθου· εἰς τοῦτο ὕβρεως καὶ µιαρίας ἀφίκετο. Καὶ πρὸς τοῖς ἄλλοις κακοῖς ὀνειδίζει καὶ ἐγκαλεῖ αὐτῷ ὅτι ἐµβάδας καὶ τρίβωνα φορεῖ, ὥσπερ ἀδικούµενός τι εἰ ἐµβάδας Κηφισόδοτος φορεῖ, ἀλλ’ οὐκ ἀδικῶν ὅτι ἀφελόµενος αὐτὸν τὰ ὄντα πένητα πεποίηκεν (‘when Dikaiogenes had got the result he wanted in the dispute against us, he evicted on the same day the daughter of Kephisophon of Paiania from her share, although she was the niece of Dikaiogenes who had left the property. And he deprived the one who had been the wife of Demokles of what her brother Dikaiogenes had given her, and he also deprived the mother of Kephisodotos and Kephisodotos himself of everything. And he was simultaneously their guardian and their opponent at law, and they received from him not even the slightest amount of pity on account of their relationship but, orphans and isolated and reduced to poverty as they were, they were in want of even the daily necessities. This was how this man Dikaiogenes discharged his guardianship of them, as their nearest kin. He is the man who gave the property left to them by their father Theopompos to their enemies, while he himself appropriated the property which their maternal uncle and grandfather had given them before having obtained a verdict. What is most terrible of all, he bought their paternal home while they were still children and demolished it and turned it into a garden adjoining his own house in the city. And, although he received rent amounting to 80 mnai from the property of our uncle Dikaiogenes, he sent Dikaiogenes’ nephew to Corinth with his own brother Harmodios instead of a retainer. This was the level of hybris and beastliness that he reached! And to add insult to injury, he reproached and criticised him for wearing slippers and a threadbare cloak, as if he himself was being wronged if Kephisodotos was wearing a threadbare cloak, and as if he was not himself wronging Kephisodotos by depriving him of his property and reducing him to poverty’).

144

Lene Rubinstein

self, but his guarantor Leochares, who is formally on trial after Dikaiogenes had failed to meet his obligation to restore part of his assets to his four adoptive sisters and their descendants. The speaker dwells at length on Dikaiogenes’ appalling treatment of all of his relatives, the speaker included, but the most detailed example pertains to Dikaiogenes’ neglect of his widowed adoptive sister and her orphaned children. The speaker highlights not only Dikaiogenes’ failure to pity the orphans in his charge, but also his deliberate humiliation of the elder orphan, who is taunted with his poverty and sent to Corinth as a retainer to Dikaiogenes’ natural brother in place of a slave. What the speaker describes as most terrible of all is Dikaiogenes decision to purchase and demolish the orphans’ paternal home – as the orphans’ guardian he would have acted as both the vendor and purchaser in this transaction – in order to add a garden to his own residence in the asty. The speaker is certainly aiming to stir up dicastic pity for the orphans’ plight with his emphasis on their isolation, utter destitution, and humiliation. But the strong terms ὕβρις and µιαρία following after the description of the children’s’ suffering reveal that his aim is first and foremost that of stirring up anger. The culmination of his character assassination is Dikaiogenes’ taunting the boy with his poverty, which may have especially angered the poorer members of the dicastic panel. In fact, it comes very close to the kind of kakegoria (slander) which, according to the speaker of Demosthenes 57,20 was against the law, while the rest of Dikaiogenes’ behaviour as depicted by the speaker could be interpreted as amounting to the criminal offence of kakosis orphanou (maltreatment of orphan). It must be noted, though, that the speaker does not go as far as openly applying that label, let alone stating that a prosecution of Dikaiogenes ought to be initiated. No remedy is suggested for the orphans’ suffering at the hands of an unscrupulous guardian, and the speaker is not asking the judges to act on their pity and assist the orphans. His main aim, it seems, is to amplify the dicastic anger directed against Dikaiogenes in the hope that this, in turn, might translate into hostility towards the actual defendant, Dikaiogenes’ guarantor Leochares. 20

Demosthenes 57.30f.: περὶ δὲ τῆς µητρὸς (καὶ γὰρ ταύτην διαβεβλήκασί µου) λέξω, καὶ µάρτυρας ὧν ἂν λέγω, καλῶ. καίτοι, ὦ ἄνδρες Ἀθηναῖοι, οὐ µόνον παρὰ τὸ ψήφισµα τὰ περὶ τὴν ἀγορὰν διέβαλλεν ἡµᾶς Εὐβουλίδης, ἀλλὰ καὶ παρὰ τοὺς νόµους, οἳ κελεύουσιν ἔνοχον εἶναι τῇ κακηγορίᾳ τὸν τὴν ἐργασίαν τὴν ἐν τῇ ἀγορᾷ ἢ τῶν πολιτῶν ἢ τῶν πολιτίδων ὀνειδίζοντά τινι. ἡµεῖς δ᾿ ὁµολογοῦµεν καὶ ταινίας πωλεῖν καὶ ζῆν οὐχ ὅντινα τρόπον βουλόµεθα (‘I shall speak about my mother (for he is also slandering her) and I shall call witnesses to my statements. Yet, Athenians, Euboulides is not only slandering her on the matter of the agora in contravention of the decree but also in contravention of the laws, which prescribe that the one who taunts any of his male or female fellow citizens with their work in the agora shall be liable under the law on slander. We for our part admit both to selling ribbons and to living in a way that we do not desire’). Cf. Deinarchos 1.36 about Demosthenes: οὐδ’ ὁτιοῦν, ἀλλὰ περιῄεις κατασκευάζων λογοποιούς, καὶ παρ᾿ αὑτῷ γράφων ἐπιστολὰς καὶ καταισχύνων τὴν τῆς πόλεως δόξαν, ἐκ τῶν δακτύλων ἀναψάµενος περιεπορεύετο, τρυφῶν ἐν τοῖς τῆς πόλεως κακοῖς, καὶ ἐπὶ φορείου κατακοµιζόµενος τὴν εἰς Πειραιᾶ ὁδόν, καὶ τὰς τῶν πενήτων ἀπορίας ὀνειδίζων. On Demosthenes 57.30f., see further Wallace 1994, 116f.

Evoking Anger through Pity in Attic Oratory

145

In general, male and female orphans, as well as the elderly of both genders, appear frequently as third party victims with obvious pity potential.21 Sometimes, in passages relating to the opponent’s behaviour in connection with armed conflicts (both inside and outside Athens), entire groups of women, children, and old people can be represented collectively as the opponent’s victims.22 However, when it comes to members of these categories appearing individually in the context of invectives, it is interesting to observe that the appearance of women in this role is largely restricted to widows (including aging mothers and widowed mothers-in-law), and that of children to orphans (including fatherless young girls at the mercy of their older brothers). It is extremely rare to find the accusation that an opponent has mistreated his wife,23 and the same is true of accusations of paternal neglect. Among the few examples are Aischines’ taunt against Demosthenes for not having mourned his daughter’s death as he should,24 and a general allegation of incest as a type of crime committed by persons of the same sort as the defendant.25 As for cruelty towards non-orphaned children outside the opponent’s household, their deployment is largely restricted to those whose fathers have been stricken with atimia (disfranchisement) as a result of the opponent’s malicious litigation, and who therefore are almost as vulnerable as if they had really been orphans.26 But on the whole, childhood alone does not seem to have done the trick in the Athenian courtroom, and it is especially when it comes to children that there seems to have been a marked difference between the ancient Athenians and modern Europeans. For many of the latter, an emotional response to the plight of a child does not depend on the child’s being without parental protection (although it does of course help). The expectation that a particular type of individual might be an appropriate instrument for generating an emotional response clearly was not only a question of gender or age but just as much of the individual’s actual vulnerability, legally and socially. And it is therefore unsurprising that the following three categories of men tend to appear as third party victims: prisoners of war, those who were atimoi (disfranchised) as a result of sykophancy or innocent debt to the state, and the eld21 22

23

24 25

26

E.g., Lysias 13.44f., 31.17–19, 31.21–23; Isaios 8.40–42; Demosthenes 24.200–203, 25.83f., 45.63–65, 45.70, 57.58; Aischines 1.99, 1.102–104; Lykourgos 1.39–41. E.g., Lysias 14.45f. (women, children, and aged parents of the Thirty’s victims); Deinarchos 1.24 (Theban women and children); Aischines 3.157 (Theban women, children, and elderly); Demosthenes 19.65f. (women, children, and elderly of Phokis), 19.100f. and 310 (women and children of ‘ruined allies’). A striking exception is Andokides’ lurid account of how his opponent, Kallias, drove his young wife to attempt suicide by striking up a sexual relationship with her mother and fathering a child with her, whom he later tried to disown (Andokides 1.124–128). Aischines 3.77f.; Lysias 14.41. A further possible example of parental misconduct being held up for censure is Demosthenes’ allegation that Phrynon of Rhamnous had sent his under-age son to Philip II’s court, with the insinuation that the child was subjected to sexual abuse (Demosthenes 19.230–233). Of course, Phrynon was not the defendant in this trial, but his allegedly close relation with the defendant Aischines may have served to blacken the latter by association. Demosthenes 19.283f., 21.99, 25.84.

146

Lene Rubinstein

erly. Among the third group, fathers who were neglected or abused represented at the same time an extremely powerful instrument by which a litigant might hope to stir up dicastic anger, as was also the case with old and helpless mothers. To a significant extent, though not quite, there is an overlap between the vulnerable groups just mentioned and the groups whose interests were protected by Athenian criminal legislation. When it came to elderly mothers and fathers, the laws prescribed instant and total atimia for the son who neglected their needs in life and after their death. Social norms combined with actual legislation in this area seem to have been so strong that we even find suggestions that one’s duty towards one’s parents might override one’s obligations in other areas of Athenian legislation. In Demosthenes 24 we come across the octogenarian Antiphon of Krioa, allegedly a victim of filial neglect.27 Here the speaker is making it quite clear that he expects his point to meet with dicastic anger. He also explicitly says that the defendant’s behaviour towards his father ought to cancel out any syngnome (sympathy) that he might have hoped to command from the judges. The argument is an interesting one, as far as Athenian values and priorities are concerned. The defendant, Timo-krates, is accused of having introduced an unsuitable law at the behest of Androtion and his associates in return for money. The speaker claims to expect Timokrates to assert that he had been motivated by pity for his friends, and that this ought to count as a mitigating factor. The speaker anticipates this line of argument as follows. The one mitigating factor that Timokrates could have cited in order to win the judges’ syngnome would have been that he had used the bribes to pay off his father's debt to the state. This he has not done, although the debt is allegedly a trivial sum; instead he prefers to enjoy the fruits of his crimes, as long as 27

Demosthenes 24.200f.: ὃ τοίνυν ἔµοιγε δοκεῖ µάλιστ᾿ ἄξιον ὀργῆς εἶναι, φράσω καὶ οὐκ ἀποτρέψοµαι, ὅτι ταῦτ᾿, ὦ ἄνδρες Ἀθηναῖοι, πράττων ἐπ᾿ ἀργυρίῳ, καὶ προῃρηµένος ὡς ἀληθῶς µισθαρνεῖν, οὐκ εἰς ἃ καὶ συγγνώµην ἀκούσας ἄν τις ἔσχε, ταῦτ᾿ ἀναλίσκει. ταῦτα δ’ ἐστὶν τί; ὁ πατήρ, ὦ ἄνδρες δικασταί, ὁ τούτου τῷ δηµοσίῳ ὀφείλει· καὶ οὐκ ὀνειδίζων ἐκείνῳ λέγω, ἀλλ’ ἀναγκαζόµενος· καὶ οὗτος ὁ χρηστὸς περιορᾷ. καίτοι ὅστις µέλλων κληρονοµήσειν τῆς ἀτιµίας, ἂν ἐκεῖνός τι πάθῃ, µὴ οἴεται δεῖν ἐκτεῖσαι, ἀλλὰ κερδαίνειν, ὃν ἐκεῖνος ζῇ χρόνον, ἀξιοῖ τοῦτο τὸ κέρδος, τίνος ἂν ὑµῖν ἀποσχέσθαι δοκεῖ; καὶ τὸν µὲν πατέρ᾿ οὔτ’ ἐλεεῖς οὔτε δεινά σοι δοκεῖ πάσχειν, εἰ σοῦ λαµβάνοντος καὶ χρηµατιζοµένου ἀπὸ τῶν εἰσφορῶν ὧν εἰσέπραττες, ἀπὸ τῶν ψηφισµάτων ὧν γράφεις, ἀφ’ ὧν εἰσφέρεις νόµων, διὰ µικρὸν ἀργύριον µὴ µετέχει τῆς πόλεως, ἑτέρους δ᾿ ἐλεῆσαί τινας φῄς; (‘now, what seems to me to be especially deserving of anger, I shall say and I shall not refrain from doing so, because, Athenians, committing these deeds in return for money and preferring, frankly, to hire himself out, he has not spent this money on the one thing which a listener might sympathise with. What is this? His father, judges, owes money to the public treasury. And I am not saying this to blame him but because I have to. And this fine man does not care. Yet a man, who stands to inherit the atimia, if anything happens to his father, and does not believe that he ought to pay up but thinks fit to enjoy this profit so long as his father is alive, what do you think he will refrain from? And you neither pity your father nor think that he is suffering a terrible wrong if he is deprived of his citizen rights on account of a paltry sum, while you are making money and profiting from the eisphorai that you exact and from the decrees that you propose, and from the laws that you introduce – yet you say that you took pity on others?’).

Evoking Anger through Pity in Attic Oratory

147

his father stays alive. The speaker comes close to saying that a desperate attempt by a son to come to the aid of his aging father in such circumstances might meet with tolerance, even if the proceeds have been made illegally. The sentiment that crimes motivated by destitution might meet with syngnome is not unparalleled in Attic oratory,28 although its rarity may suggest that the view was controversial. In the present context, however, the syngnome envisaged would be based not only on the perpetrator’s own destitution but, more importantly, on his sense of obligation towards his father that conflicted with his obligation towards the laws of the community. Indeed, it might even be claimed that a son in such a situation faced an actual legal dilemma: his choice would be between breaking the law obliging him to provide gerotrophia (care of the aged parents)29 and breaking other laws in order to meet his obligation. That such a dilemma could result in acquittal is suggested by Deinarchos:30 the speaker relates 28

29

30

Demosthenes 45.67: καὶ µήν, ὦ ἄνδρες Ἀθηναῖοι, µᾶλλον ἄξιον ὀργίλως ἔχειν τοῖς µετ᾿ εὐπορίας πονηροῖς ἢ τοῖς µετ᾿ ἐνδείας. τοῖς µὲν γὰρ ἡ τῆς χρείας ἀνάγκη φέρει τινὰ συγγνώµην παρὰ τοῖς ἀνθρωπίνως λογιζοµένοις· οἱ δ᾿ ἐκ περιουσίας, ὥσπερ οὗτος, πονηροὶ οὐδεµίαν πρόφασιν δικαίαν ἔχοιεν ἂν εἰπεῖν, ἀλλ’ αἰσχροκερδίᾳ καὶ πλεονεξίᾳ καὶ ὕβρει καὶ τῷ τὰς αὑτῶν συστάσεις κυριωτέρας τῶν νόµων ἀξιοῦν εἶναι ταῦτα φανήσονται πράττοντες (‘now, Athenians, it is more fitting to be angrily disposed towards criminals in affluent circumstances than towards those in circumstances of poverty. For the latter, the constraints arising from poverty command some sympathy from those who make their assessment with fellow-feeling. But those who commit crimes from a position of affluence are not able to cite a single legitimate excuse, but it will be obvious that they are committing these acts because of lust for sordid gain and greed and hybris and because they believe their own gangs to be more powerful than the laws’). Cf. Lysias 31.11. The verb is actually used in Demosthenes 24.203: ὃς οὖν τὴν µὲν ἀδελφὴν ἐπ’ ἐξαγωγῇ, φησὶ µὲν ἐκδοῦναι, πέπρακε δὲ τῷ ἔργῳ, τὸν δ’ αὑτοῦ πατέρ’ οὕτω γηροτροφεῖ, κολακεύει δὲ καὶ µισθοῦ γράφει καὶ πολιτεύεται, τοῦτον ὑµεῖς λαβόντες οὐκ ἀποκτενεῖτε; (‘the man, then, who claims to have given his sister in marriage but who in fact has sold her for export, who provides so badly for his old father, and who flatters and makes proposals and exercises his citizen rights in return for pay, will you not execute this man when you have caught him?’). On emotional aspects of gerotrophia in papyri of the Hellenistic and Imperial period, see Kotsifou 2012, 78–81. Deinarchos 1.58f.: Πολύευκτον δὲ τὸν Κυδαντίδην τοῦ δήµου προστάξαντος ζητῆσαι τὴν βουλήν, εἰ συνέρχεται τοῖς φυγάσιν εἰς Μέγαρα, καὶ ζητήσασαν ἀποφῆναι πρὸς ὑµᾶς, ἀπέφηνεν ἡ βουλὴ συνιέναι. κατηγόρους εἵλεσθε κατὰ τὸν νόµον, εἰσῆλθεν εἰς τὸ δικαστήριον, ἀπελύσαθ’ ὑµεῖς, ὁµολογοῦντος τοῦ Πολυεύκτου βαδίζειν εἰς Μέγαρ᾿ ὡς τὸν Νικοφάνην· ἔχειν γὰρ τὴν αὑτοῦ µητέρα τοῦτον. οὐδὲν οὖν ἄτοπον οὐδὲ δεινὸν ἐφαίνεθ᾿ ὑµῖν ποιεῖν, τῷ τῆς µητρὸς ἀνδρὶ διαλεγόµενος ἠτυχηκότι καὶ συνευπορῶν, καθ’ ὅσον δυνατὸς ἦν, ἀπεστερηµένῳ τῆς πατρίδος. αὕτη, ∆ηµόσθενες, τῆς βουλῆς ἡ ἀπόφασις οὐκ ἐξηλέγχθη ψευδὴς οὖσα, ἀληθινῆς δ᾿ αὐτῆς οὔσης ἔδοξε τοῖς δικασταῖς ἀφεῖναι τὸν Πολύευκτον· τὸ µὲν γὰρ ἀληθὲς τῇ βουλῇ προσετάχθη ζητεῖν, τὸ δὲ συγγνώµης ἄξιόν φηµι τὸ δικαστήριον ἔκρινε (‘when the assembly had instructed the Areiopagos to investigate if Polyeuktos had met with the exiles at Megara and to report its findings to you, the council presented the conclusion that he had done so. You elected prosecutors according to the law, he came before the court, you acquitted him, although Polyeuktos admitted that he had gone to Megara to visit Nikophanes. For he was married to Polyeuktos’ mother. Thus he did not appear to you to have done anything monstrous or terrible by talking to his mother’s husband in his misfortune and by helping him to the best of his ability in his exile from his fa-

148

Lene Rubinstein

that, although Polyeuktos admitted to the unlawful act of going to Megara during wartime, he was acquitted because his motive had been to assist his mother’s second husband in his exile – and by implication his mother herself. 3 VILLAINOUS VICTIMS As far as Timokrates’ father is concerned, it is possible that the speaker was able to represent his claim to pity from his son, if not also from the judges, in such clear terms because Timokrates’ father does not seem to have incurred his debt as a result of a serious criminal conviction. The debt was small, and the speaker’s assertion that he does not intend to blame Antiphon suggests that his debt had been incurred for entirely honourable reasons.31 But there are other passages that may suggest that it was not only honourable victims that were deserving of pity, at least from their nearest and dearest. In the context of invective, there is a category of victims who on the one hand are presented as illustrations of the opponent’s unacceptable behaviour, but on the other are themselves objectionable characters or at least characters with whom the dicastic audience were not expected to identify. I shall first discuss some striking examples of vulnerable villains, and next turn to victims whose status as non-citizens, some free and some unfree, set them apart from the judges and the citizen community that they were representing. In two of the three extant prosecution speeches against Aristogeiton, Demosthenes 25 and Deinarchos 2, Aristogeiton’s treatment of his late father, Kydimachos, forms part of the character assassination. According to Demosthenes,32 Aristogeiton abandoned his father who was imprisoned in Eretria. It is clear from this passage that an account of this incident has already been presented to the court by Phaidros, either as a witness called by the main prosecutor, or as a synegoros (supporting speaker). Here, the speaker focuses on Aristogeiton’s refusal not only to bury his father, but also to compensate those who had performed the funeral and whom he allegedly sued. However, Kydimachos himself was a not a

31 32

therland. Demosthenes, this report of the council was not exposed as untrue, but although it was true, the judges decided to acquit Polyeuktos. For the council had been instructed to search for the truth, but I assert that the court decided on what was deserving of syngnome’). Davies 1971, 314, tentatively suggests that Antiphon or his son may have served as trierarch in the 370s and the debt may have been incurred in that connection. Demosthenes 25.54: Ἄξιον δ᾿ ἐστίν, ὦ ἄνδρες Ἀθηναῖοι, καὶ τὰ λοίπ᾿ ἀκοῦσαι· δεινῶν γὰρ ὄντων, οὐ µὲν οὖν ἐχόντων ὑπερβολήν, ὧν ἠκούσατ᾿ ἄρτι λέγοντος Λυκούργου, τὰ λοίπ’ ἐνάµιλλα τούτοις καὶ τῆς αὐτῆς φύσεως εὑρεθήσεται. πρὸς µὲν γὰρ τῷ τὸν πατέρ᾿ ἐν τῷ δεσµωτηρίῳ προδοὺς ἀπελθεῖν ἐξ Ἐρετρίας, ὥσπερ ἠκούσατε Φαίδρου, ἀποθανόνθ᾿ ὁ ἀσεβὴς οὗτος καὶ µιαρὸς οὐκ ἔθαψεν, οὐδὲ τοῖς θάψασι τὴν ταφὴν ἀπέδωκεν, ἀλλὰ καὶ δίκην πρὸς ἔλαχεν (‘you ought also to hear the rest, men of Athens. For what you have just heard from Lykourgos are terrible things, yet not exaggerated, but the rest rivals this and will be found to be of the same nature. For in addition to his betraying his own father in prison and leaving Eretria, as you have heard from Phaidros, this impious and defiled man did not bury him nor reimburse those who did, but he even sued them into the bargain’).

Evoking Anger through Pity in Attic Oratory

149

character with whom the judges were likely to sympathise:33 he had fled Athens after being indicted on a capital charge and convicted, almost certainly in absentia. It is therefore understandable that the speaker here characterises Aristogeiton’s neglect mainly as an offence of asebeia, that is a violation of religious norms. And, as Günther Martin notes,34 the strong term of abuse, miaros (abominable), undoubtedly also has a religious connotation in the present context. What dicastic emotion was the speaker hoping to stir up here? Probably a sense of outrage at Aristogeiton’s act of asebeia combined with disgust at his beastly behaviour, but almost certainly not sympathy, let alone pity, for his abandoned father. The focus on Aristogeiton rather than on his victim is to be expected in light of the speaker’s overall strategy and the nature of the prosecution. The formal charge of the endeixis (denunciation) brought against Aristogeiton was that he was exercising his citizen privileges despite being atimos, and the offence of kakosis goneon (maltreatment of parents) was punishable with atimia.35 Thus, even if Aristogeiton proved able to persuade the judges that he was not a state debtor (on which the formal charge appears to have turned), he might still have found it difficult to convince the court that he deserved to be treated as epitimos (entitled to exercise citizen privileges). But if we turn, next, to the account of the same episode in Deinarchos,36 there is a subtle difference in focus and tone. Here Aristogeiton is represented as allowing his father to live in want of basic necessities and to be denied the customary rites after his death. Aristogeiton’s offence is seen from the point of view of his victim, and the passage bears far more resemblance to the topoi on neglect of vulnerable dependants discussed earlier. That the speaker may have hoped to arouse a 33

34 35 36

Demosthenes 25.77: τί γὰρ ὡς ἀληθῶς ἐρεῖ; ὧν ὁ πατήρ τι πεποίηκε νὴ ∆ία. ἀλλὰ κατεγνώκαθ᾿ ὑµεῖς ἐν τουτοισὶ τοῖς δικαστηρίοις αὐτοῦ θάνατον, ὡς πονηροῦ δηλονότι καὶ ἀξίου τεθνάναι. ἀλλὰ νὴ ∆ία, εἰ ταῦτ’ ἐστὶν αὐτῷ δυσχερῆ τὰ περὶ τὸν πατέρα, εἰς τὸν ἑαυτοῦ βίον καταφεύξεται ὡς σώφρονα καὶ µέτριον. ποῖον; ὃν ποῦ βεβίωκεν; (‘what will he actually say? By Zeus, because of something that his father has done. But you sentenced him to death in these courts, because he was obviously wicked and deserving of the death penalty. But, by Zeus, if the issue of his father is proving tricky for him, he will take refuge in his own way of life as one of self-restraint and moderation. What sort of life? Where has he lived it?’). Martin 2009, 187–189. Hansen 1976, 90f. Deinarchos 2.8: ἀλλ᾿ αὐτὸς ὁ κρινόµενος νὴ ∆ία µέτριος τὸν τρόπον, καὶ προγόνων χρηστῶν, καὶ πόλλ᾿ ὑµᾶς καὶ ἰδίᾳ καὶ δηµοσίᾳ καλὰ εἰργασµένος, ὥστε διὰ ταῦτ’ ἄξιόν ἐστιν αὐτοῦ φείσασθαι; καίτοι τίς ὑµῶν οὐ πολλάκις ἀκήκοεν, ὅτι Κυδιµάχου µὲν τοῦ πατρὸς τοῦ Ἀριστογείτονος θανάτου καταγνωσθέντος καὶ φυγόντος ἐκ ταύτης τῆς πόλεως, ὁ χρηστὸς οὗτος υἱὸς περιεῖδε τὸν αὑτοῦ πατέρα καὶ ζῶντα τῶν ἀναγκαίων σπανίζοντα καὶ τελευτήσαντ᾿ οὐ τυχόντα τῶν νοµίµων, ἅπερ αὐτοῦ πολλάκις κατεµαρτυρεῖτο (‘but the defendant himself, by Zeus, is a moderate character, and of good parentage, and responsible for many benefactions towards you both in private and public life, so should he be spared because of this? Yet, who among you have not heard many times that, when Aristogeiton’s father Kydimachos had been sentenced to death and fled this city, this fine son allowed his father both to be in want of the basic necessities while he was alive and not to obtain the customary rites after his death, as has often been testified against him?’).

150

Lene Rubinstein

certain level of dicastic sympathy with the plight of the old man seems more than likely when we consider the speaker’s argument:37 Are you going to turn a blind eye and feel pity when you are about to cast your vote on Aristogeiton, who did not take pity on his own father when he was reduced to a wretched state by starvation?

The argument here, that dicastic pity should be denied to those who are themselves incapable of pitying others, is the same as that used by Demosthenes in his attack on Meidias over his treatment of Straton.38 What is interesting is that this line of argument works only if the judges agree that even a convicted criminal, who has escaped Athenian justice, should nevertheless be deserving of pity during his suffering and after his death. Although there can be no doubt that the speaker is seeking first and foremost to arouse dicastic anger against Aristogeiton, the representation of the wretched condition of Kydimachos may well have been intended to amplify the hostile emotion that the speaker was trying to generate in his audience. Another example of a vulnerable villain is that of the convicted murderer, Aristarchos son of Moschos.39 Aristarchos’ murder of Nikodemos of Aphidna is represented in horrific detail by Aischines:40 both Nikodemos’ eyes were knocked 37

38

39 40

Deinarchos 2.11: ἔπειτ᾿ εἰρωνεύεσθε πρὸς ὑµᾶς αὐτούς, καὶ περὶ Ἀριστογείτονος µέλλοντες φέρειν τὴν ψῆφον ἐλεεῖτε, ὃς τὸν αὑτοῦ πατέρα κακῶς διατιθέµενον ὑπὸ τοῦ λιµοῦ οὐκ ἠλέησεν; (‘will you then turn a blind eye and feel pity when you are about to cast your votes on Aristogeiton who did not take pity on his own father when he was reduced to a wretched state by hunger?’). Demosthenes 21.99: καὶ τίς ἂν ταῦτ’ ἐλεήσειε δικαίως, ὁρῶν τὰ τοῦδ᾿ οὐκ ἐλεηθένθ᾿ ὑπὸ τούτου, ἃ τῇ τοῦ πατρὸς συµφορᾷ χωρὶς τῶν ἄλλων κακῶν οὐδ᾿ ἐπικουρίαν ἐνοῦσαν ὁρᾷ (‘and who would with justice pity his children, observing that Straton’s children were not pitied by him – the children who, quite apart the rest of their plight, cannot see any possible remedy for their father’s misfortune?’). For an outline of this notorious case and of the relevant evidence, see Fisher 2001, 316–318. Aischines 1.170–172: ∆ηµοσθένης γάρ, ἐπειδὴ τὴν πατρῴαν οὐσίαν ἀνήλωσε, περιῄει περὶ τὴν πόλιν θηρεύων νέους πλουσίους ὀρφανούς, ὧν οἱ µὲν πατέρες ἐτετελευτήκεσαν, αἱ δὲ µητέρες διῴκουν τὰς οὐσίας. Πολλοὺς δ᾿ ὑπερβὰς ἑνὸς τῶν δεινὰ πεπονθότων µνησθήσοµαι. Κατιδὼν γὰρ οἰκίαν πλουσίαν καὶ οὐκ εὐνοµουµένην, ἧς ἡγεµὼν µὲν ἦν γυνὴ µέγα φρονοῦσα καὶ νοῦν οὐκ ἔχουσα, νεανίσκος δὲ ὀρφανὸς ἡµιµανὴς διεχείριζε τὴν οὐσίαν, Ἀρίσταρχος ὁ τοῦ Μόσχου, τούτου προσποιησάµενος ἐραστὴς εἶναι, καὶ τὸ µειράκιον εἰς τὴν φιλανθρωπίαν ταύτην προσκαλεσάµενος, ἐλπίδων κενῶν ἐµπλήσας, ὡς αὐτίκα δὴ µάλα τῶν ῥητόρων πρωτεύσοντα, κατάλογον ἀποφαίνων, τοιούτων εἰσηγητὴς αὐτῷ καὶ διδάσκαλος ἔργων ἐγένετο, ἐξ ὧν ἐκεῖνος µὲν φεύγει τὴν πατρίδα, οὗτος δ᾿ αὐτοῦ τὰ τῆς φυγῆς ἐφόδια προλαβὼν τρία τάλαντα ἀπεστέρηκε, Νικόδηµος δ᾿ ὁ Ἀφιδναῖος ὑπ᾿ Ἀριστάρχου τετελεύτηκε βιαίῳ θανάτῳ, ἐκκοπεὶς ὁ δείλαιος ἀµφοτέρους τοὺς ὀφθαλµοὺς καὶ τὴν γλῶτταν ἀποτµηθείς, ᾗ ἐπαρρησιάζετο πιστεύων τοῖς νόµοις καὶ ὑµῖν (‘for Demosthenes, when he had squandered his paternal inheritance, went about the city hunting down rich young orphans, whose fathers had died and whose mothers were administering their estates. I shall pass over many, but I shall mention one of those who have suffered terribly. For he spotted a wealthy household, not well run, the head of which was an arrogant and senseless woman, while a young orphan, half-mad, administered the estate, Aristarchos, the son of Moschos. He pretended to be in love with him, and he invited the young man to his

Evoking Anger through Pity in Attic Oratory

151

out and his tongue cut off, ‘with which he had been exercising his right to free speech, trusting in you and the laws’. Especially this last sentence must have been designed to evoke a combination of horror and pity for Aristarchos’ victim. But Aristarchos himself is likewise portrayed as a victim of Demosthenes – who, it must be noted, is in fact not a party in the present suit but expected to appear as synegoros for the actual defendant, Timarchos. Aischines takes care to mention that Aristarchos was an orphan in a household controlled by an arrogant and senseless woman; although Aristarchos must already have been of age, his lack of a father’s guidance combined with his own mental deficiencies makes him a defenceless prey of Demosthenes (note the participle θηρεύων at the beginning of the passage). We hear how the young man was duped by Demosthenes’ erotic advances and with empty hopes that Demosthenes’ teaching would make him a leading political figure. The story, like the story of Neaira, is told from the young man’s point of view. And in this case it is very likely that Aischines intends his audience to pity the unfortunate young man, whom Demosthenes not only induced to commit the savage murder but also deprived of a substantial sum of money, which had been meant to sustain him in his exile. Aischines’ emphasis on the vulnerable position of the young orphan in the present passage is in stark contrast to his characterisation of Aristarchos’ household in his speech On the False Embassy:41 there it is ‘happy’, and the emphasis is on pretended love and betrayal of friendship, as is also the case in Deinarchos’ version of the story.42 In the version presented in the

41

42

generosity, filling him with empty hopes that he would soon be the most influential of the rhetores, and he showed him a register, and he became an adviser and teacher on such horrible deeds, which led to his fleeing his fatherland, while Demosthenes procured the money that was to support him in his exile and cheated him out of three talents. As for Nikodemos, he met with a violent death at the hands of Aristarchos, with the poor man having both his eyes knocked out and his tongue cut off, with which he had been exercising his right to free speech, trusting in you and the laws’). Aischines 2.166: εἰσῆλθες εἰς εὐδαιµονοῦσαν οἰκίαν τὴν Ἀριστάρχου τοῦ Μόσχου· ταύτην ἀπώλεσας. Προὔλαβες τρία τάλαντα παρ’ Ἀριστάρχου φεύγοντος· τοῦτον τὰ τῆς φυγῆς ἐφόδια ἀπεστέρησας, οὐκ αἰσχυνθεὶς τὴν φήµην, ἣν προσεποιήσω, ζηλωτὴς εἶναι τῆς ἡλικίας τοῦ µειρακίου. Οὐ γὰρ δὴ τῇ γε ἀληθείᾳ· οὐ προσδέχεται δίκαιος ἔρως πονηρίαν. Ταῦτ᾿ ἐστὶν ὁ προδότης καὶ τὰ τούτοις ὅµοια (‘You entered into a happy household, that of Aristarchos, son of Moschos. You ruined it. You received in trust three talents from Aristarchos when he fled into exile. You robbed him of his means of subsistence in his exile, and you were not ashamed of the common report which you put about that you were an admirer of the youthful beauty of the lad. This was not true. For righteous love is incompatible with wickedness. This, and deeds like this, mark the traitor’). Deinarchos 1.30: οὐκ εἰς µὲν τὴν Ἀριστάρχου οἰκίαν εἰσελθών, βουλεύσας µετ’ ἐκείνου τὸν Νικοδήµῳ θάνατον κατασκευασθέντα, ὃν ἴστε πάντες, ἐξέβαλε τὸν Ἀρίσταρχον ἐπὶ ταῖς αἰσχίσταις αἰτίαις; καὶ τοιούτῳ φίλῳ ∆ηµοσθένει ἐχρήσατο, ὥστε δαίµον᾿ αὑτῷ τοῦτον καὶ τῶν γεγενηµένων συµφορῶν ἡγεµόνα νοµίσαι προσελθεῖν; (‘did he not enter Aristarchos’ home and did he not plot with him the death arranged for Nikodemos, which you all know, and did he not get Aristarchos exiled on the basest charges? This man had such a friend in Demosthenes that he believed that Demosthenes had visited him as an evil spirit and the originator of his misfortune?’).

152

Lene Rubinstein

Timarchos speech, Aristarchos’ vulnerability as an orphan and Demosthenes’ exploitation of him are emphasised, most likely in order to appeal to pity for the victim, despite the dreadful deed that he had perpetrated. By this means, Aischines may have hoped that pity might amplify the hostile emotions directed against Demosthenes, culminating in Aischines’ comparison of Demosthenes to Sokrates, the teacher of Kritias and Alkibiades. It might of course be argued at this point that, no matter how abominable the exploits of Kydimachos and Aristarchos, both men were indisputably of citizen status, and that it was this fact in itself that made it possible for Deinarchos and Aischines to adopt the strategies just discussed. But if the citizen status of these victims may have made it easier for the litigants to generate some fellow-feeling in the judges, whether pity or sympathy or a combination of both, there is plenty of evidence to suggest that feelings of pity did not invariably depend on parity in civic status between pitier and pitied. Some of this evidence is provided by several graphic descriptions of the suffering and devastation that has stricken women, children, and the elderly in other Greek communities referred to above. Space does not permit a proper discussion here, except to note that these descriptions are not significantly different from the descriptions of crises and disasters that had befallen Athens at various points in its history. Likewise, when it came to refugees and prisoners of war, those belonging to other cities appear to have had a good deal of pity-potential in the Athenian courtroom. It could be argued that the shared Greekness of the pitiers and the pitied bound them together. Especially in the case of non-Athenian refugees and war captives, a powerful bond may also have been created by the universal realisation that the route from freedom and citizenship to slavery and humiliation was often terrifyingly short in war, and that stasis might strike any community, anywhere, anytime, with displacement of citizens as an almost inevitable result. However, even slaves, freedmen, and non-citizens who were not identified as victims of such catastrophic events were sometimes included in litigants’ narratives. In some instances there can be little doubt that the narrator was trying to stir up pity or sympathy for the plight of the non-Athenian victim – in others, there is undeniably a good deal of uncertainty over what feelings the litigant was aiming to generate in his audience. The first instance I shall treat very briefly is the celebrated account in Demosthenes 47.55–59 of the brutal attack on an elderly wetnurse by the litigant’s opponents and her subsequent death.43 In this passage, the 43

Demosthenes 47.55–59: πρὸς δὲ τούτοις, ὦ ἄνδρες δικασταί, ἔτυχεν ἡ γυνή µου µετὰ τῶν παιδίων ἀριστῶσα ἐν τῇ αὐλῇ, καὶ µετ’ αὐτῆς τιτθή τις ἐµὴ γενοµένη πρεσβυτέρα, ἄνθρωπος εὔνους καὶ πιστὴ καὶ ἀφειµένη ἐλευθέρα ὑπὸ τοῦ πατρὸς τοῦ ἐµοῦ. συνῴκησεν δὲ ἀνδρί, ἐπειδὴ ἀφείθη ἐλευθέρα· ὡς δὲ οὗτος ἀπέθανεν καὶ αὐτὴ γραῦς ἦν καὶ οὐκ ἦν αὐτὴν ὁ θρέψων, ἐπανῆκεν ὡς ἐµέ. ἀναγκαῖον οὖν ἦν µὴ περιιδεῖν ἐνδεεῖς ὄντας µήτε τιτθὴν γενοµένην µήτε παιδαγωγόν· ἅµα δὲ καὶ τριηραρχῶν ἐξέπλεον, ὥστε καὶ τῇ γυναικὶ βουλοµένῃ ἦν τοιαύτην οἰκουρὸν µετ’ αὐτῆς µε καταλιπεῖν. ἀριστώντων δὲ ἐν τῇ αὐλῇ, ὡς ἐπεισπηδῶσιν οὗτοι καὶ καταλαµβάνουσιν αὐτὰς καὶ ἥρπαζον τὰ σκεύη, αἱ µὲν ἄλλαι θεράπαιναι (ἐν τῷ πύργῳ γὰρ ἦσαν, οὗπερ διαιτῶνται) ὡς ἤκουσαν κραυγῆς, κλείουσι τὸν πύργον, καὶ ἐνταῦθα µὲν οὐκ εἰσῆλθον, τὰ δ᾿ ἐκ τῆς ἄλλης οἰκίας ἐξέφερον

Evoking Anger through Pity in Attic Oratory

153

speaker is clearly adopting a multi-pronged strategy: he shows his opponents’ outrageous intrusion into the speaker’s house in the presence of his female and underage dependants, he describes in graphic detail his opponents’ propensity for violence, and he uses the episode as a demonstration of his opponents’ wickedness (poneria).44 The episode as a whole is crucial for the speaker’s overall attempt to prove his opponent guilty of false witnessing: it is much more than just a part of an emo-

44

σκεύη, ἀπαγορευούσης τῆς γυναικὸς µὴ ἅπτεσθαι αὐτοῖς, καὶ λεγούσης ὅτι αὑτῆς εἴη ἐν τῇ προικὶ τετιµηµένα καὶ ὅτι «τὰ πρόβατα ἔχετε πεντήκοντα καὶ τὸν παῖδα καὶ τὸν ποιµένα, πλείονος ἄξια ἢ κατεδικάσασθε»· ἀπήγγειλε γάρ τις αὐτοῖς τῶν γειτόνων κόψας τὴν θύραν. ἔτι δὲ ἔφη τὸ ἀργύριον αὐτοῖς κείµενον εἶναι ἐπὶ τῇ τραπέζῃ· ἠκηκόει γὰρ ἐµοῦ· «κἂν περιµείνητε», ἔφη, «ἢ µετέλθῃ τις ὑµῶν αὐτόν, ἔχοντες ἄπιτε τὸ ἀργύριον ἤδη· τὰ δὲ σκεύη ἐᾶτε, καὶ µηδὲν τῶν ἐµῶν φέρετε, ἄλλως τε καὶ ἔχοντες ἄξια τῆς καταδίκης». ταῦτα δὲ λεγούσης τῆς γυναικὸς οὐχ ὅπως ἐπέσχον, ἀλλὰ καὶ τῆς τιτθῆς τὸ κυµβίον λαβούσης παρακείµενον αὐτῇ, ἐξ οὗ ἔπινεν, καὶ ἐνθεµένης εἰς τὸν κόλπον, ἵνα µὴ οὗτοι λάβοιεν, ἐπειδὴ εἶδεν ἔνδον ὄντας αὐτούς, κατιδόντες αὐτὴν οὕτω διέθεσαν ἀφαιρούµενοι τὸ κυµβίον Θεόφηµος καὶ Εὔεργος ἁδελφὸς αὐτοῦ οὑτοσί, ὥστε ὕφαιµοι µὲν οἱ βραχίονες καὶ οἱ καρποὶ τῶν χειρῶν αὐτῆς ἐγένοντο ἀποστρεφοµένης τὼ χεῖρε καὶ ἑλκοµένης ὑπὸ τούτων ἀφαιρουµένων τὸ κυµβίον, ἀµυχὰς δ᾿ ἐν τῷ τραχήλῳ εἶχεν ἀγχοµένη, πελιὸν δὲ τὸ στῆθος. εἰς τοῦτο δ’ ἦλθον πονηρίας ὥστε, ἕως ἀφείλοντο τὸ κυµβίον ἐκ τοῦ κόλπου αὐτῆς, οὐκ ἐπαύσαντο ἄγχοντες καὶ τύπτοντες τὴν γραῦν (‘moreover, judges, my wife happened to be lunching with my children in the courtyard, and with her was an old woman who had been my wet-nurse, a loyal and trusty person who had been freed by my father. After she had been freed, she married a man. But when he died and she herself was an old woman and there was no-one who would provide for her, she returned to my household. For I was obliged not to allow either her, who had been my wet-nurse, nor the one who had been my paidagogos to be in want. At the same time I was preparing to sail away as a trierarch so that my wife, too, wanted me to leave such a good housekeeper with her. While they were lunching in the courtyard, when these men burst in and came upon them and engaged in carrying off my household equipment, the rest of the female servants (they happened to be in the tower where they lived) locked the tower when they heard the screams. And these men did not enter there but were carrying out the equipment from the rest of the house, although my wife told them not to seize it and said that it had been assessed as part of her dowry and that “you have fifty sheep and the slave and the shepherd, which are worth more than the sum you were awarded in the court case”. For one of the neighbours had knocked on the door and reported this to them. Next, she said that the money was ready for them at the bank. For she had heard this from me. “And if you wait”, she said, “or if one of you goes to fetch him, then leave with the money. But leave our equipment alone, and don’t take any of my property, especially since you already have the value of the judgement debt”. But although my wife said this, they did not in any way leave off. On the contrary, when my nurse had taken the drinking cup next to her from which she had been drinking, and hidden it in her bosom when she saw that these men were inside the house so that they would not seize it, Theophemos and his brother Euergos spotted her and reduced her to such a state in their attempt to seize the cup that her arms and wrists became bruised, because her hands were forced behind her back and she was being dragged about by these men while they were trying to seize the cup, and she received strangling marks on her throat from being suffocated, and her breast was black and blue. And they reached such a pitch of wickedness that they did not stop strangling and beating the old woman until they had taken the cup from her bosom’). On the significance of graphic descriptions and enargeia for emotional arousal, see Chaniotis 2012, 107–109 (focusing on documentary sources).

154

Lene Rubinstein

tional strategy. But at the same time, there can be no doubt that the speaker is trying to generate pity for the old woman’s plight. He describes her as old, she is trusty and loyal, as a widow she had no one to turn to except for the speaker and his family, and the speaker adds the touching detail that he has left her as a companion for his wife during his absence abroad. If the speaker wanted his audience to respond first and foremost to his opponents’ forced entry into his oikos and the dishonour that this caused him as its kyrios (head of household), the initial presentation of the old woman’s story along with the warm and generous relationship between her and her former master might be seen as superfluous. It is most likely that the issue here is not only the outrage perpetrated against a citizen home and its kyrios. We are invited to see the old woman as a victim subjected to the most appalling and brutal maltreatment without provocation, and as a human being, rather than as a member of a particular status group. Now, it may be that the old woman’s pity potential was partly a result of her being a de facto part of a citizen household, which she had served loyally for at least two generations. However, other non-citizen women also appear occasionally as third party victims. Among them is Zobia, a one-time girlfriend of Aristogeiton.45 She is not exactly portrayed as either respectable or chaste. Nor is her willingness to shield a notorious criminal after his escape from prison testimony to her being law-abiding or loyal to the Athenian community as a whole. Yet in the speaker’s representation in Demosthenes 25.56f., she is depicted as a woman who runs a risk to save her former lover, and Aristogeiton’s treatment of her is 45

Demosthenes 25.56f.: πρὸς δὲ τούτοις τοιούτοις οὖσιν ἕτερον δεινόν, ὦ γῆ καὶ θεοί, πρᾶγµ᾿ ἀκούσεσθε. ὅτε γὰρ τὸ δεσµωτήριον διορύξας ἀπέδρα, τότε πρὸς γυναῖκά τιν ἔρχεται Ζωβίαν ὄνοµα, ᾗ ἐτύγχανεν, ὡς ἔοικε, κεχρηµένος ποτέ· καὶ κρύπτει καὶ διασῴζει τὰς πρώτας ἡµέρας αὐτὸν ἐκείνη, ἃς ἐζήτουν καὶ ἐκήρυττον οἱ ἕνδεκα, καὶ µετὰ ταῦτα δοῦσα δραχµὰς ὀκτὼ ἐφόδιον καὶ χιτωνίσκον καὶ ἱµάτιον ἐξέπεµψεν εἰς Μέγαρα. ταύτην τὴν ἄνθρωπον, τὴν τοιαῦτ’ εὐεργετήσασαν αὐτόν, ὡς πολὺς παρ᾿ ὑµῖν ἔπνει καὶ λαµπρός, µεµφοµένην τι καὶ τούτων ὑποµιµνῄσκουσαν καὶ ἀξιοῦσαν εὖ παθεῖν τὸ µὲν πρῶτον ῥαπίσας καὶ ἀπειλήσας ἀπέπεµψεν ἀπὸ τῆς οἰκίας, ὡς δ᾿ οὐκ ἐπαύεθ᾿ ἡ ἄνθρωπος, ἀλλὰ γυναίου πρᾶγµ᾿ ἐποίει καὶ πρὸς τοὺς γνωρίµους προσιοῦσ᾿ ἐνεκάλει, λαβὼν αὐτὸς αὐτοχειρίᾳ πρὸς τὸ πωλητηρίον τοῦ µετοικίου ἀπήγαγεν· καὶ εἰ µὴ κείµενον αὐτῇ τὸ µετοίκιον ἔτυχεν, ἐπέπρατ᾿ ἂν διὰ τοῦτον, ᾧ τῆς σωτηρίας αὐτὴ αἰτία ἐγεγόνει (‘in addition to these things, bad as they are, you shall hear another terrible fact, by Earth and the gods. For when this man dug his way out of prison and escaped, he went to a woman called Zobia, with whom, it seems, he had once had a sexual relationship. And for the first few days, while the Eleven were searching for him and had made a public call for his arrest, she hid him and kept him safe. After this, she gave him eight drachmai for his journey and a short tunic and a cloak and sent him off to Megara. This woman, the one who had bestowed on him such a great benefaction, he first struck with a stick and sent away from his house with threats because she had voiced a grievance against him and reminded him of these things, when he was putting on magnificent airs in your midst, and asked him to treat her fairly. When the woman did not desist, but tried to resort to a weak woman’s scheme and went and complained about him to some of his acquaintances, he seized her with his own hands and led her under arrest to the poleterion [office for public auctions] for not paying her metic tax. And if her metic tax had happened not to have been paid, she would have been sold at the instigation of this man for whose salvation she had been responsible’).

Evoking Anger through Pity in Attic Oratory

155

characterised as δεινόν from the very beginning. The minute details of how she made provisions for Aristogeiton’s flight to Megara (including by giving him the sum of eight drachmai – two drachmai more than a year’s metoikion that later nearly cost her her freedom) almost certainly aimed to make the listener sympathise with her when she is later beaten, threatened, and ultimately betrayed by the very man she had saved. Of course, it cannot be ruled out that the speaker was trying to generate contempt for the woman herself. This may be suggested by the designations ἄνθρωπος and γύναιον. But neither of these nouns is unambiguously negative: it must be noted that the loyal, old wet-nurse in Demosthenes 47 is designated ἄνθρωπος εὔνους καὶ πιστή.46 Likewise, the designation ἡ ἄνθρωπος is used in Demosthenes’ devastating account of Aischines’ outrageous treatment of an Olynthian captive woman, from which it may be inferred that it is not invariably used to indicate contempt, at least not to the extent that this would have prevented the listeners from feeling sympathy or pity.47 In the latter description, there can be no doubt whatsoever that Demosthenes’ aim is to stir up a powerful cocktail of pity as well as anger, and that the portrait of the woman herself is entirely sympathetic. As for the noun γύναιον, I have failed to find any clearly contemptuous use of this noun 46

47

The speaker’s description of his own relationship with his freed-woman in Demosthenes 47 has been discussed by Zelnik-Abramovitz 2005, 330–332, who interprets the use of the noun anthropos as an indication that the ties of affection may not have been quite so strong as the speaker makes out. But even if this is correct, the important point in the present context is that the speaker’s account emphasises the warmth of their personal relationship, and that he expects this to resonate with his audience. Demosthenes 19.197f.: ἐπειδὴ δ᾿ ἧκον εἰς τὸ πίνειν, εἰσάγει τιν᾿ ᾿Ὀλυνθίαν γυναῖκα, εὐπρεπῆ µέν, ἐλευθέραν δὲ καὶ σώφρονα, ὡς τὸ ἔργον ἐδήλωσεν. ταύτην τὸ µὲν πρῶτον οὑτωσὶ πίνειν ἡσυχῇ καὶ τρώγειν ἠνάγκαζον οὗτοί µοι δοκεῖ, ὡς διηγεῖτ᾿ Ἰατροκλῆς ἐµοὶ τῇ ὑστεραίᾳ· ὡς δὲ προῄει τὸ πρᾶγµα καὶ διεθερµαίνοντο, κατακλίνεσθαι καί τι καὶ ᾄδειν ἐκέλευον. ἀδηµονούσης δὲ τῆς ἀνθρώπου καὶ οὔτ’ ἐθελούσης οὔτ’ ἐπισταµένης, ὕβριν τὸ πρᾶγµ᾿ ἔφασαν οὑτοσὶ καὶ ὁ Φρύνων καὶ οὐκ ἀνεκτὸν εἶναι, τῶν θεοῖς ἐχθρῶν, τῶν ἀλειτηρίων Ὀλυνθίων αἰχµάλωτον οὖσαν τρυφᾶν· καὶ «κάλει παῖδα», καὶ «ἱµᾶντά τις φερέτω». ἧκεν οἰκέτης ἔχων ῥυτῆρα, καὶ πεπωκότων, οἶµαι, καὶ µικρῶν ὄντων τῶν παροξυνόντων, εἰπούσης τι καὶ δακρυσάσης ἐκείνης περιρρήξας τὸν χιτωνίσκον ὁ οἰκέτης ξαίνει κατὰ τοῦ νώτου πολλάς. ἔξω δ’ αὑτῆς οὖσ’ ὑπὸ τοῦ κακοῦ καὶ τοῦ πράγµατος ἡ γυνή, ἀναπηδήσασα προσπίπτει πρὸς τὰ γόνατα τῷ Ἰατροκλεῖ, καὶ τὴν τράπεζαν ἀνατρέπει. καὶ εἰ µὴ ᾿κεῖνος ἀφείλετο, ἀπώλετ᾿ ἂν παροινουµένη (‘when they got to the state of drinking, he brought in an Olynthian woman, pretty and free and chaste, as the episode revealed. It seems to me that they tried at first to force her to drink and eat quietly, as Iatrokles told me the next day. But as the event unfolded and the atmosphere got heated, they ordered her to lie down and even to sing, too. As the woman was anguished and neither would nor could do so, my opponent and Phrynon said that it was outrageous and intolerable that this woman, one of the damned and accursed Olynthian captives, was to be spoiled. And “call a slave” and “someone bring a whip!”. A slave came with a strap and, I think because they were drunk and easily provoked, when she said something and began to cry, the slave tore off her tunic and flogged her back several times. As the woman was beside herself because of the ill-treatment and the event, she jumped up and threw herself in supplication at the knees of Iatrokles and overturned the table. And if he hadn’t taken her away, she would have been killed as the victim of a drunken episode’).

156

Lene Rubinstein

in fifth- or fourth-century literature, be it poetry or prose.48 Thus, it seems more likely that the speaker of Demosthenes 25 is trying to stir up contempt against Aristogeiton, in addition to other hostile emotions, by describing his reliance on a far weaker person in his hour of need and his complete disregard for his obligations to reciprocate when he later found himself in a position to do so. No clear question mark is placed over Zobia’s entitlement to decent treatment from the man she had saved, and the speaker’s condemnation of Aristogeiton’s ingratitude is unequivocal.49 Zobia, like the old wet-nurse, was almost certainly free, and we may well ask if similar reciprocation would have been regarded as a moral obligation if the benefactor had been of slave status. The number of unfree persons appearing as third party victims is very low indeed; when they do, it is often made clear that their status as slaves is contested, and this makes it far harder to interpret the speakers’ agendas. But in one instance there can be no doubt that the speaker wants his audience to regard his third party victim’s slave status as a fact.50 In Aischines’ account of Timarchos’ sexual relationship with Pittalakos, allegedly a state slave, Pittalakos’ status is a key component of Aischines’ character assassination and, more importantly, of Aischines’ attempt to demonstrate that Timarchos had engaged in sex in return for money. That money, not love, was Timarchos’ reason for submitting to various sexual acts perpetrated by Pittalakos is, to Aischines, proved in part by Pittalakos’ unfree status per se. Aischines’ brief account of the relationship between Pittalakos and Timarchos is followed by a much more detailed description of how Timarchos abandoned Pittalakos and formed a new sexual relationship with an Athenian citizen, Hegesandros. If we may find it hard to work out precisely what feelings the speaker of Demosthenes 25 wanted his listeners to have towards Zobia, the case of Pittalakos presents even more serious difficulties. As has been pointed out by several scholars, the tone in Aischines’ account is to a large extent one of sarcasm, with the sarcasm being at Timarchos’ and his associate Hegesandros’ expense. Some of the vocabulary employed by Aischines clearly belongs to the register of comedy rather than oratory (the noun kakodaimon, ‘poor devil’, is used only twice in the entire corpus of Attic oratory,51 and the term triskakodaimon only once, in the present passage). The image of the jilted Pittalakos constantly hovering outside 48

49 50

51

LSJ s.v. γύναιον II, asserts that, although the noun is sometimes used as a term of endearment, it is more frequently attested in a contemptuous sense. However, although the passage cited as an example, Andokides 1.130, juxtaposes them with ‘little children’, contempt is not obvious here. For a similar interpretation, see Rosenbloom 2003, 109f. Modern opinion on the real status of Pittalakos is divided. See e.g., the discussion in Fisher 2001, 190f., who suggests that Pittalakos may in fact have been free at the time when the assault happened, but that Aischines when representing him as a publicly-owned slave exploited the ambiguity surrounding Pittalakos’ status as a former slave in order to highlight the depth of Timarchos’ degradation. Contrast Hunter 2006, 1–8, who presents strong arguments in favour of the view that Pittalakos was indeed a slave. Antiphon 5.43; Demosthenes 19.115.

Evoking Anger through Pity in Attic Oratory

157

the house of Hegesandros is not without a certain resemblance to the behaviour of love-sick young men in New Comedy, although the speaker represents the nature of Timarchos’ sexual relationship as a passive partner with Pittalakos as completely beyond the pale. But things turn really ugly and the tone changes somewhat, once we reach the point where Timarchos, Hegesandros, and some of their friends decide to teach the abandoned Pittalakos a lesson. They smash up Pittalakos’ gambling venue, kill his birds and, having tied the man to a pillar, give him such a flogging that his screams can be heard by the neighbours (just as had been the case when the old nurse was beaten in Demosthenes 47). The treatment of Pittalakos is labelled hybris,52 and Aischines gives a description of how the man sought sanctuary at the altar of the Mother of the Gods with his physical injuries from the flogging on display. When he tries to seek justice through the courts, he is manipulated, deceived, and intimidated. He finally has to give up without having obtained any justice from his tormentors.53 To what extent are we invited to see the episode from Pittalakos’ point of view? It seems that the speaker is telling at least part of the story from his angle: we are told that he was deeply aggrieved at the hybris to which he had been subjected, and we are told in no uncertain terms that he had suffered injustice. To52

53

Aischines 1.62: βαρέως δὲ φέρων τὴν ὕβριν αὐτῶν ὁ ἄνθρωπος, δίκην ἑκατέρῳ αὐτῶν λαγχάνει. Ὅτε δ᾿ ἐδικάζετο, σκέψασθε µεγάλην ῥώµην Ἡγησάνδρου· ἄνθρωπον οὐδὲν αὐτὸν ἠδικηκότα, ἀλλὰ τὸ ἐναντίον ἠδικηµένον, οὐδὲν προσήκοντα αὐτῷ, ἀλλὰ δηµόσιον οἰκέτην τῆς πόλεως, ἦγεν εἰς δουλείαν φάσκων ἑαυτοῦ εἶναι. Ἐν παντὶ δὲ κακοῦ γενόµενος ὁ Πιττάλακος, προσπίπτει ἀνδρὶ καὶ µάλα χρηστῷ. Ἔστι τις Γλαύκων Χολαργεύς· οὗτος αὐτὸν ἀφαιρεῖται εἰς ἐλευθερίαν (‘since the man took their hybris [outrage] to heart, he initiated private actions against each of them. When he was in the process of going to court, consider the great might of Hegesandros. This person, who had not wronged him but who on the contrary had been wronged by him, and who did not belong to him but was a public slave of the polis, this man he appropriated as his slave, claiming that he was his property. Pittalakos who had been landed in a totally awful situation encountered a real man and a very good one, too. He is one Glaukon of Cholargos. He asserted Pittalakos’ freedom’). On the allegation of hybris and on the presumably private action that Pittalakos was attempting to bring, see Fischer 2001, 199f. Aischines 1.63f.: ὡς δὲ παρῄει ἐπὶ τὸ βῆµα τὸ ὑµέτερον ὁ Ἡγήσανδρος, ὅτε καὶ προσεπολέµει Ἀριστοφῶντι τῷ Ἀζηνιεῖ, πρὶν αὐτῷ τὴν αὐτὴν ταύτην ἐν τῷ δήµῳ ἠπείλησεν ἐπαγγελίαν ἐπαγγελεῖν ἥνπερ ἐγὼ Τιµάρχῳ, καὶ ἐπειδὴ Κρωβύλος ὁ ἀδελφὸς αὐτοῦ ἐδηµηγόρει, καὶ ὅλως ἀπετόλµων ὑµῖν οὗτοι περὶ τῶν Ἑλληνικῶν συµβουλεύειν, ἐνταῦθα δὴ καταµεµψάµενος ἑαυτὸν ὁ Πιττάλακος καὶ ἐκλογισάµενος ὅστις ὢν πρὸς οὕστινας ἐπολέµει, εὖ ἐβουλεύσατο – δεῖ γὰρ τἀληθὲς λέγειν –· ἡσυχίαν ἔσχεν, καὶ ἠγάπησεν εἴ τι µὴ προσλάβοι καινὸν κακόν. Ἐνταῦθα δὴ τὴν καλὴν ταύτην νίκην νενικηκὼς ὁ Ἡγήσανδρος ἀκονιτί, εἶχε παρ’ ἑαυτῷ Τίµαρχον τουτονί (‘when Hegesandros was appearing regularly as a speaker in court, when he was also warring against Aristophon of Azenia, before the latter threatened to make the same indictment of him as I am now making against Timarchos, and when his brother Krobylos was addressing the assembly regularly and, in short, these people dared to advise you on the affairs of Greece, at that point Pittalakos acknowledged his weakness and made an assessment of his own position vis-à-vis those whom he was fighting against; he made a wise decision – for the truth must be told – and kept quiet, and he was happy if he could avoid meeting with a new disaster. At this point Hegesandros, who had won this splendid victory without opposition kept this Timarchos here at his house’).

158

Lene Rubinstein

wards the end of the narrative, we are party to Pittalakos’ assessment of his own situation and his reasons for abandoning his cause as hopeless in the face of an opposition that has proved far too strong and unscrupulous for him. On the other hand, the ironic distance and, indeed, the sarcasm running through much of the narrative combined with the initial account of how Pittalakos procured the sexual services of Timarchos and himself subjected Timarchos to hybris may in fact make it harder for the listener to feel pity for Pittalakos in his plight. This is in spite of the very powerful image of him being tied up and flogged and, later, his sitting naked and bruised at the altar in the middle of the Agora surrounded by a crowd of spectators. As pointed out by Nick Fisher,54 we have seriously to consider the possibility that the account of Pittalakos, like the story of Neaira, may have been deployed in order to stir up, in the first instance, feelings of contempt for Pittalakos, which in turn would make the behaviour of the citizen who had submitted to his sexual advances seem all the more despicable and undignified. This impression is further strengthened by the reference to Glaukon of Cholargos, Pittalakos’ saviour, as a ‘real man’ (ἀνδρί). But at the same time, it is implied that a characteristic of true masculinity is the willingness to come to the rescue of the defenceless, however lowly their status. 4 PERFORMANCE The answer to what emotional effect Aischines intended his account to have on his audience may in part elude us because all we have left of his speech is the written word. His intonation, emphasis, and pitch would almost certainly have been crucial to the success of his various indirect emotional appeals, including those that we may now find it hard to identify. It is frustrating that both Aristotle’s Rhetoric and the Rhetoric to Alexander have only tantalisingly little to say about the use of the voice in the context of emotional appeals. It is even more regrettable that we have lost Thrasymachos’ work on pity, which very clearly did discuss the connection between delivery and the arousing of emotions in the audience. In Plato’s Phaidros, Thrasymachos is characterised as a past master of the art of appropriate diction, ὀρθοέπεια, when producing tearful speeches, οἰκτρογόοι λόγοι, drawing on old age and poverty. He was also more than capable of moving large audiences to anger and, likewise, of charming those who have been brought into a state of anger by ‘singing enchantments’.55 54 55

Fisher 2001, 189f. Plato, Phaidros 267C–D: [Socrates] ὀρθοέπειά γέ τις, ὦ παῖ, καὶ ἄλλα πολλὰ καὶ καλά. τῶν γε µὴν οἰκτρογόων ἐπὶ γῆρας καὶ πενίαν ἑλκοµένων λόγων κεκρατηκέναι τέχνῃ µοι φαίνεται τὸ τοῦ Χαλκηδονίου σθένος, ὀργίσαι τε αὖ πολλοὺς ἅµα δεινὸς ἁνὴρ γέγονεν, καὶ πάλιν ὠργισµένοις ἐπᾴδων κηλεῖν, ὡς ἔφη· διαβάλλειν τε καὶ ἀπολύσασθαι διαβολὰς ὁθενδὴ κράτιστος (‘yes, my boy, appropriate diction and many other splendid things. For the strength of the Chalkedonian seems to me to prevail through the art of producing tearful speeches that draw on old age and poverty; moreover, he is a man who is accomplished in making a mass audience angry and, again, to charm people who have been brought into a sta-

Evoking Anger through Pity in Attic Oratory

159

The importance of voice and of other means of expressing one’s feelings for the effect of direct and indirect emotional appeals is very clear indeed from Aischines’ and Demosthenes’ comments on each other’s delivery. In Demosthenes 18,56 Aischines is criticised for raising his voice, shouting and expressing his glee when attacking Demosthenes as the author of Athens’ misfortune – according to Demosthenes, this mode of delivery is totally inappropriate in the context. A truly loyal citizen would have burst into tears at this point, and Aischines’ mismatch of his mode of delivery – which clearly was expressing anger rather than grief and pain – shows him up as a man indifferent to the fate of his own city. If Demosthenes’ characterisation of Aischines’ delivery is accurate, it is indeed possible that Aischines may have attempted to stir up anger by mentioning Athens’ war dead and their bereaved relatives,57 and that he had adopted an angry mode of delivery to achieve that end. The many references to defendants’ emotional pleas for pity at the end of their speeches also often contain references indicating what the prosecutors expected the defendants’ mode of delivery to be like. It is more than probable that similar histrionic techniques were deployed throughout the speeches, with the speakers modulating their delivery in order to engender particular types of emotion. Unfortunately, we can probably never hope fully to reconstruct this aspect of delivery on the basis of the evidence of the forensic speeches, and we cannot know whether some or all of the passages in which the plight of third party victims was described would have been delivered aggressively or tearfully in order to achieve the desired effect. Yet there are passages that tell us something about a range of different methods of delivery, from dead-pan to impassioned. Here, I shall concentrate on the dead-pan, because it may provide a starting point for a discussion of what we have lost and why this loss makes the identification of indirect emotional appeals so very difficult. To my knowledge, the contrast between the litigants’ voices and the voice of the court attendant has not played much of a part in modern discussions of the Athenian forensic stage. Yet, it is very important for our understanding of

56

57

te of anger with his incantations, as he said. And he is most powerful when it comes both to producing character assassinations and to undoing them’). Demosthenes 18.291: πολλὰ τοίνυν, ὦ ἄνδρες Ἀθηναῖοι, καὶ ἄλλα κατηγορηκότος αὐτοῦ καὶ κατεψευσµένου, µάλιστ᾿ ἐθαύµασα πάντων ὅτε τῶν συµβεβηκότων τότε τῇ πόλει µνησθεὶς οὐχ ὡς ἂν εὔνους καὶ δίκαιος πολίτης ἔσχε τὴν γνώµην οὐδ’ ἐδάκρυσεν, οὐδ’ ἔπαθεν τοιοῦτον οὐδὲν τῇ ψυχῇ, ἀλλ᾿ ἐπάρας τὴν φωνὴν καὶ γεγηθὼς καὶ λαρυγγίζων ᾤετο µὲν ἐµοῦ κατηγορεῖν δηλονότι, δεῖγµα δ᾿ ἐξέφερεν καθ’ ἑαυτοῦ ὅτι τοῖς γεγενηµένοις ἀνιαροῖς οὐδὲν ὁµοίως ἔσχε τοῖς ἄλλοις (‘when my opponent was making many various false accusations, I was surprised, most of all, at the fact that he, when he was recalling the disasters that befell our city then, he did not hold the views of a loyal citizen with a sense of justice, nor did he burst into tears, nor did he suffer any such pain in his soul. No, by raising his voice and displaying his glee and shouting he clearly thought he was accusing me. In reality, he was providing a demonstration to his own detriment that his attitude to those irreparable events was totally different from that of other people’). Aischines 3.152–154.

160

Lene Rubinstein

different modes of delivery and their combination with different types of spectacle, which might all work together to generate powerful emotional responses.58 In some instances, the orators clearly adopted a combination of appeal through their own voice, the voice of the court attendant, and the visual impact of the witness’ presence on the bema. One example is Aischines’ account of Timarchos’ treatment of his old, blind uncle Arignotos.59 Because of his disability, Arignotos had never divided his paternal inheritance with his brother, Timarchos’ father, but continued to live in his household until Timarchos’ father died. While Timarchos himself was still an orphan, his guardians made sure that Arignotos received what he needed, but as soon as Timarchos had come of age, he terminated this arrangement; his uncle was now left to fend for himself and was reduced to such utter destitution that he became eligible for the means-tested disability allowance. When this, in turn, was taken from him, Timarchos allegedly did nothing to intervene, although he had had an opportunity to do so. This verbal 58 59

For an overview and discussion the importance of visual effects, including the speaker’s gaze, gestures, and facial expressions, see Hall 2006, 376–381. Aischines 1.102–104: ἦσαν οὗτοι τρεῖς ἀδελφοί, Εὐπόλεµός τε ὁ παιδοτρίβης καὶ Ἀρίζηλος ὁ τούτου πατὴρ καὶ Ἀρίγνωτος, ὃς ἔτι καὶ νῦν ἔστι, πρεσβύτης διεφθαρµένος τοὺς ὀφθαλµούς. Τούτων πρῶτος ἐτελεύτησεν Εὐπόλεµος, ἀνεµήτου τῆς οὐσίας οὔσης, δεύτερος δ᾿ Ἀρίζηλος ὁ Τιµάρχου πατήρ· ὅτε δ᾿ ἔζη, πᾶσαν τὴν οὐσίαν διεχείριζε διὰ τὴν ἀσθένειαν καὶ τὴν συµφορὰν τὴν περὶ τὰ ὄµµατα τοῦ Ἀριγνώτου καὶ διὰ τὸ τετελευτηκέναι τὸν Εὐπόλεµον, καί τι καὶ εἰς τροφὴν συνταξάµενος ἐδίδου τῷ Ἀριγνώτῳ. Ἐπεὶ δὲ καὶ ὁ Ἀρίζηλος ἐτελεύτησεν ὁ Τιµάρχου τουτουὶ πατήρ, τοὺς µὲν πρώτους χρόνους, ἕως παῖς ἦν οὗτος, ἅπαντα τὰ µέτρια ἐγίγνετο παρὰ τῶν ἐπιτρόπων τῷ Ἀριγνώτῳ· ἐπειδὴ δ᾿ ἐνεγράφη Τίµαρχος εἰς τὸ ληξιαρχικὸν γραµµατεῖον καὶ κύριος ἐγένετο τῆς οὐσίας, παρωσάµενος ἄνδρα πρεσβύτην καὶ ἠτυχηκότα, θεῖον ἑαυτοῦ, τήν τε οὐσίαν ἠφάνισε, καὶ τῶν ἐπιτηδείων οὐδὲν ἐδίδου τῷ Ἀριγνώτῳ, ἀλλὰ περιεῖδεν ἐκ τοσαύτης οὐσίας ἐν τοῖς ἀδυνάτοις µισθοφοροῦντα. Καὶ τὸ τελευταῖον, ὃ καὶ δεινότατον, ἀπολειφθέντος τοῦ πρεσβύτου τῆς γιγνοµένης τοῖς ἀδυνάτοις δοκιµασίας, ἱκετηρίαν θέντος εἰς τὴν βουλὴν ὑπὲρ τοῦ µισθοῦ, βουλευτὴς ὢν καὶ προεδρεύων ἐκείνην τὴν ἡµέραν, οὐκ ἠξίωσεν αὐτῷ συνειπεῖν, ἀλλὰ περιεῖδεν ἀπολέσαντα τὸν τῆς πρυτανείας µισθόν. Ὅτι δ᾿ ἀληθῆ λέγω, κάλει µοι Ἀρίγνωτον τὸν Σφήττιον, καὶ τὴν µαρτυρίαν ἀναγίγνωσκε (‘there were three brothers, Eupolemos, the athletic trainer, Arizelos, my opponent’s father, and Arignotos, who is still alive, an old man disabled by blindness. Of these Eupolemos died first, while the estate was undivided, and secondly Arizelos, Timarchos’ father. While he was alive, he was administering the entire estate because of Arignotos’ frailty and unfortunate loss of sight and because Eupolemos was dead, and he gave Arignotos an agreed allowance for his sustenance. When Arizelos, the father of Timarchos here, had died as well, Arignotos was treated decently in every respect by Timarchos’ guardians, while Timarchos was still a child. But when Timarchos had been entered on the lexiarchikon grammateion and had become master of the estate, he thrust aside the old and unfortunate man, his own uncle, and concealed his assets and did not provide for Arignotos’ needs but allowed him, a man belonging to such a vast estate, to draw the disability allowance. And finally and most outrageously, when the old man had been rejected in the scrutiny of the disabled and made a formal supplication to the council over the matter of the allowance, my opponent did not wish to speak on his behalf, although he was a councillor and presiding on that day, but he allowed him to lose the allocation for that prytany. To show that this is true, please call Arignotos of Sphettos and read out the testimony’).

Evoking Anger through Pity in Attic Oratory

161

account in itself very likely had a damaging effect, but the really devastating part of Aischines’ attack was undoubtedly when old Arignotos was called to his bema as a witness. We do not know precisely what testimony he was asked to confirm, but his very appearance must have had enormous pity potential. We should probably envisage the old, blind man being assisted onto the bema in a spectacle of utter helplessness – a more powerful demonstration of Timarchos’ indifference and neglect than any words could possibly have produced. Another example of how the mere presentation of a witness might be used for powerful emotional effect is found in Demosthenes 25.62. The speaker has just finished an account of how Aristogeiton, during his imprisonment, attempted to rob a fellow prisoner, a Tanagraian man, but was found out. A fight ensued, in which Aristogeiton allegedly bit off the Tanagraian’s nose. This narrative is followed by the Tanagraian himself being called to the platform as a witness, after which point the speaker comments simply καλῶν γε ἔργων ὁ ῥήτωρ ἡµῖν γέγονεν (‘what splendid works this rhetor has wrought for you’). We can only guess at the emotional impact of the display, but it is likely that the sight of the noseless man would have met with a mixture of pity and outright disgust. In other instances, however, the presence of the witnesses and the court attendant’s voice were made to speak for themselves. Two of our speeches delivered in private actions, Demosthenes’ For Phormion (36) and Against Euergos and Mnesiboulos (47), do contain references to third party victims as part of their narratives, but these are not elaborated and, as they stand, their emotional impact may indeed have been very limited.60 However, there are further references towards the end of both speeches, which are produced without any narrative at all, but which nevertheless may have had a considerably greater effect. Here, the litigants announce that information will be provided relating to other aspects of their opponents’ antisocial behaviour.61 In both instances, the crimes attributed to the oppo60

61

Demosthenes 36.53 (Apollodoros’ sykophantic attacks on a number of named victims); Demosthenes 47.32 (Theophemos’ apparently baseless litigation against the orphans left by Demochares). Demosthenes 47.82: εἰ δέ τις ἀγνοήσας αὐτοὺς τότε ἀκάκους ἡγήσατο καὶ ἀπράγµονας εἶναι, βούλοµαι ὑµῖν περὶ αὐτῶν τὰς µαρτυρίας ἀναγνῶναι, ἃς µεµαρτυρήκασί µοι οἱ ὑπὸ τούτων ἠδικηµένοι (λόγῳ µὲν γὰρ διηγήσασθαι οὐκ ἂν ἱκανόν µοι γένοιτο τὸ ὕδωρ), ἵν᾿ ἐκ τούτων ἁπάντων σκεψάµενοι, τῶν τε λεχθέντων καὶ τῶν µαρτυρουµένων, ὁσίαν καὶ δικαίαν ὑπὲρ ὑµῶν αὐτῶν τιθῆσθε τὴν ψῆφον. λέγε τὰς µαρτυρίας (‘in case anyone who hasn’t come to know them were to think that my opponents were innocent then and averse to becoming engaged in legal matters, I want to read you the witness statements about them, which all of those who have been wronged by them have confirmed for me (for there would not be enough water for me to narrate this to you), so that you may deliberate on the basis of all of this, both what has been said and what has been testified, and cast a pious and just vote on your own behalf. Read the witness statements’). Demosthenes 36.55f.: καὶ νὴ ∆ί᾿ ἔγωγ᾿, ὦ ἄνδρες Ἀθηναῖοι, νοµίζω πάνθ᾿ ὅσα τοῦ τρόπου τοῦ Φορµίωνός ἐστι σηµεῖα καὶ τῆς τούτου δικαιοσύνης καὶ φιλανθρωπίας, καὶ ταῦτ᾿ εἰς τὸ πρᾶγµ᾿ εἶναι πρὸς ὑµᾶς εἰπεῖν. ὁ µὲν γὰρ περὶ πάντ᾿ ἄδικος τάχ᾿ ἄν, εἰ τύχοι, καὶ τοῦτον ἠδίκει· ὁ δὲ µηδένα µηδὲν ἠδικηκώς, πολλοὺς δ᾿ εὖ πεποιηκὼς ἑκών, ἐκ τίνος εἰκότως ἂν τρόπου τοῦτον µόνον ἠδίκει τῶν πάντων; τούτων τοίνυν τῶν µαρτυριῶν ἀκούσαντες γνώσεσθε τὸν ἑκατέρου τρόπον.

162

Lene Rubinstein

nents have been committed against several victims, who are all asked to come forward and testify. As is well known, these victims would not have addressed the court in their own words, nor would the litigants be able to engage in a more elaborate narrative detailing their sufferings, while the witnesses were on display. Their presence on the bema would have been to the accompaniment of testimony read out by the court attendant. His job would almost certainly not have involved anything other than just reading the documents plainly, loudly, and clearly as instructed. In these two instances, the court would have learnt the details of the offences only from the court attendant’s voice, reinforced by the spectacle of the witnesses on the speaker’s platform. While the words passionlessly recited may have had the effect first and foremost of evoking anger or other types of hostility, the visual impact of the witnesses may very likely have contributed further emotional amplification through a silent appeal to the judges’ pity. There are other examples of invective where the speakers’ voices are replaced by the voice of the court attendant, accompanied by the spectacle of the opponents’ victims on the witness stand. 62 The practice itself makes excellent sense in the context of private legal actions, in which it is most frequently attested. The time allowance for a litigant in a private suit was very limited compared to the slots allocated to litigants in public actions. The generation of an emotional impact through a conventional narrative of the victims’ plight would mean that the litigant would have less time to devote to his own story. However, while the witness statements were read out, the water-clock would be stopped. Thus, the replacement of the speaker’s own narrative by the court-attendant’s voice combined with the visual display of the opponent’s victims was a way of circumventing the time restrictions that would otherwise have limited the extent to which the speaker could afford to dwell on his opponent’s behaviour towards others.

62

ΜΑΡΤΥΡΙΑΙ. Ἴθι δὴ καὶ τὰς κατ᾿ Ἀπολλοδώρου τῆς πονηρίας. ΜΑΡΤΥΡΙΑΙ. Ἆρ᾿ οὖν ὅµοιος οὑτοσί; σκοπεῖτε. λέγε. ΜΑΡΤΥΡΙΑΙ. Ἀνάγνωθι δὴ καὶ ὅσα δηµοσίᾳ χρήσιµος τῇ πόλει γέγονεν οὑτοσί. ΜΑΡΤΥΡΙΑΙ (‘and, by Zeus, I for one, men of Athens, think that it is of relevance to the case to tell you everything that is an indication of Phormion’s character and sense of justice and generosity. For a man who is unjust in all respects has perhaps also engaged in wrongdoing against the opponent. But the man who has never committed any crime against anyone but has deliberately done good to a lot of people, what motivation would he have had for committing crimes against the opponent, alone of all people? When you have heards the witness statements you will know their respective characters’). E.g., Demosthenes 54.36f., where the speaker appears to call witnesses who claim to have suffered burglaries and wanton violence at the hands of men who are expected to testify for his opponent, and Isaios 8.46 where, it seems, a man is called to testify that he has been cukkolded by Diokles, the mastermind behind the present attempt to defraud the speaker of his maternal inheritance. In Demosthenes 50.68 the court attendant is asked to read out the testimony of the former trierarch Euripides, who allegedly had suffered the same treatment by the defendant as Apollodoros claims to have sustained. The fact that, in this speech as well as in Demosthenes 47 and Isaios 8, it is the court attendant’s voice along with the visual display of the opponent’s victims that bring the speaker’s performance to a close strongly suggests that this tactic was expected to have considerable emotional potential.

Evoking Anger through Pity in Attic Oratory

163

And there is yet another consideration that may have made the practice attractive to litigants engaged in this type of litigation, namely the performance-related demands that a successful description of other people’s suffering would have placed on the speaker himself. Forty-six of a total fifty-five surviving speeches delivered in private disputes were written by logographers, professional speechwriters, to be performed by their clients. While there is no apparent difference in the frequency and intensity of emotional appeals, both verbal and indirect, between logographic works and speeches that were performed by the writer in person, the verbal appeals in the former tend to centre on the litigant’s own plight and on the treatment that he and his nearest and dearest have allegedly suffered at the hands of his opponent, but much more rarely on the plight of third party victims. From the point of view of the logographer, it may have been an important consideration that a focus on the client’s own story would have made it easier for the latter to perform the speech naturally and effectively, not least in regard to those passages that were devised to appeal to dicastic emotions of pity, anger, and resentment. As is evident from Demosthenes’ criticism of Aischines’ mode of delivery discussed at the beginning of this section, the effective performance of passages dwelling on the sufferings of others represented more of a challenge. To a logographer, it would be an important question whether an inexperienced client would have been able to meet it, even if he was given detailed and specific instructions on the tone of voice and body language that went far beyond the provision of the scripted speech itself. In such situations, the logrographer’s effective stage-management of props, that is of silent witnesses accompanied by the passionless delivery of the court attendant, may have represented a far safer strategy.63 As has already been mentioned in connection with Aischines’ display of the old, disabled Arignotos and the display of the noseless Tanagraian in Demosthenes 25, more experienced orators also used the court-attendant’s voice, often in combination with narratives of their own. The most imaginative development of the convention, however, is found in a famous passage in Demosthenes’ prosecution speech Against Meidias.64 Here Demosthenes presents the elderly citizen Straton on his own bema after his description of how Straton had been victimised 63

64

Compare the logographers’ apparent reluctance to incorporate quotations from poetry, as highlighted by Hall 2006, 367, who explains this as a result of the challenges of delivery which an inexperienced speaker may not have been able to meet. Demosthenes 21.93–96 (the text of documents, the authenticity of which is disputed, is left out here): ἀλλὰ µὴν ὡς ἀληθῆ λέγω, κάλει µοι τούτων τοὺς µάρτυρας, καὶ τὸν τῶν διαιτητῶν ἀνάγνωθι νόµον. ΜΑΡΤΥΡΕΣ. Λέγε δὴ καὶ τὸν τῶν διαιτητῶν νόµον. ΝΟΜΟΣ. Κάλει δὴ καὶ τὸν Στράτων᾿ αὐτὸν τὸν τὰ τοιαῦτα πεπονθότα· ἑστάναι γὰρ ἐξέσται δήπουθεν αὐτῷ. Silence. Οὗτος, ὦ ἄνδρες Ἀθηναῖοι, πένης µὲν ἴσως ἐστίν, οὐ πονηρὸς δέ γε. οὗτος µέντοι πολίτης ὤν, ἐστρατευµένος ἁπάσας τὰς ἐν ἡλικίᾳ στρατείας καὶ δεινὸν οὐδὲν εἰργασµένος, ἕστηκε νυνὶ σιωπῇ, οὐ µόνον τῶν ἄλλων ἀγαθῶν τῶν κοινῶν ἀπεστερηµένος, ἀλλὰ καὶ τοῦ φθέγξασθαι ἢ ὀδύρασθαι· καὶ οὐδ᾿ εἰ δίκαι᾿ ἢ ἄδικα πέπονθεν, οὐδὲ ταῦτ’ ἔξεστιν αὐτῷ πρὸς ὑµᾶς εἰπεῖν. καὶ ταῦτα πέπονθ’ ὑπὸ Μειδίου καὶ τοῦ Μειδίου πλούτου καὶ τῆς ὑπερηφανίας παρὰ τὴν πενίαν καὶ ἐρηµίαν καὶ τὸ τῶν πολλῶν εἷς εἶναι.

164

Lene Rubinstein

by Meidias. As a result of Meidias’ arrogance, brutality, and thirst for disproportionate revenge, Straton has incurred atimia through Meidias’ clever manipulation of the Athenian legal system. The sentence is irreversible, since Straton’s atimia is by sentence rather than for an unpaid fine that he or his children might some day hope to pay off. Like the speakers just discussed, Demosthenes asks Straton to face the court, after a lengthy impassioned narrative delivered in Demosthenes’ own voice.65 What would normally happen at this point is that the court attendant would read out a piece of written testimony for Straton to confirm. Instead the silence is total. Not only does Straton himself not speak because of his atimia – which is what most modern commentators hav e highlighted in this celebrated example of forensic theatricality. Even the court attendant, too, is silent because of Straton’s atimia. This comes as a shock after his two lengthy recitals of documents in the previous paragraphs. The reality of Straton’s fate comes across all the more starkly when Demosthenes emphasises that he is prevented even from lamenting. The juxtaposition of the verb ὀδύρασθαι with the verb φθέγξασθαι may well suggest the kind of wordless wailing that women and children engaged in during the open and ritualised appeals to pity at the end of public defence speeches. Even this part is denied to Straton, the elderly citizen and loyal soldier. As appeals to pity go, this is a masterstroke on Demosthenes’ part, not least because this variation on an existing practice would have run counter to the audience’s expectations. Performative techniques could thus vary enormously. In many instances, our inability to reconstruct the diction employed by individual litigants when they were describing the suffering of third party victims means that it is very difficult at times to identify the emotional impact that they were aiming for. To return to the problem with the Neaira speech: one reason why Apollodoros felt sure that his account would not move his audience to pity may have been his confidence that his delivery itself would have been a powerful means of preventing any such emotional response. Any investigation of indirect appeals to emotions, be it through the representation of third party victims or through the representations of the speaker’s own actions and attitudes, as well as the treatment meted out to him by his opponent, must take account of this factor – and of the potential uncertainty that surrounds the interpretation of individual passages as a result. On the other hand, when it came to the arousal of anger or resentment fuelled by pity, the range of victims whose plight was described in the context of invective was arguably far broader than what seems to be assumed in rhetorical manuals. Some of the victims who clearly were expected to be the object of pity belonged to extremely problematic categories, and this at least calls for a reassessment of the view that both the emotion of pity and anger depended on the judges’ assessing the victim as deserving and as an individual with whom they were to some extent able to identify. Moreover, the deployment of third party victims in indirect emotional appeals shows that litigants’ attempts to manipulate dicastic 65

For an overview and discussion of the evidence relating to the methods and effectiveness of Demosthenes’ delivery in general, see e.g., Cooper 2004, 151–160.

Evoking Anger through Pity in Attic Oratory

165

emotions went much further than the prescriptions in the extant works by rhetorical theorists, who mostly discussed their deployment in contexts where it was the litigant himself and his opponent who were the direct intended targets of dicastic pity and anger respectively. BIBLIOGRAPHY Carey, C. (1992) Apollodoros Against Neaira [Demosthenes] 59.Greek Orators VI, Warminster. Chaniotis, A. (2012) Moving Stones: The Study of Emotions in Greek Inscription, in A. Chaniotis (ed.), Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, Stuttgart, 91–129. Cooper, C. (2004) Demosthenes, Actor on the Political and Forensic Stage, in C. J. Mackie (ed.) Oral Performance and Its Context, Leiden/Boston, 145–161. Cox, C. A. (1998) Household Interests. Property, Marriage Strategies, and Family Dynamics in Ancient Athens, Princeton. Davies, J. K. (1971) Athenian Propertied Families 600–300 B.C., Oxford. Fisher, N. R. E. (2001) Aeschines Against Timarchos, Oxford. Johnstone, S. (1999) Disputes and Democracy: The Consequences of Litigation in Ancient Athens, Austin. Hall, E. (2006) The Theatrical Cast of Athens. Interactions between Ancient Greek Drama and Society, Oxford. Hansen, M. H. (1976) Apagoge, Endeixis and Ephegesis against Kakourgoi, Atimoi and Pheugontes. A Study in the Athenian Administration of Justice in the Fourth Century B.C., Odense. Hunter, V. (2006) Pittalacus and Eucles: Slaves in the Public Service of Athens, Mouseion 6, 1– 13. Kapparis, K. (1999) Apollodoros Against Neaira [D.59], Berlin. Kotsifou, C. (2012) Emotions and Papyri: Insights into the Theatre of Human Experience in Antiquity, in A. Chaniotis (ed.), Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, Stuttgart, 39–90. Martin, G. (2009) Divine Talk: Religious Argumentation in Demosthenes, Oxford. Omitowoju, R. (1997) Regulating Rape: Soap Operas and Self-Interest in the Athenian Courts, in S. Deacy and K. F. Pierce (eds.) Rape in Antiquity. Sexual Violence in the Greek and Roman Worlds, London, 1–24. Rosenbloom, D. (2003) Aristogeiton Son of Cydimachus and the Scoundrel’s Drama, in J. Davidson and A. Pomeroy (eds.) Theatres of Action. Papers for Chris Dearden (Prudentia Supplement), Auckland, 88–117. Rubinstein, L. (2004) Stirring up Dicastic Anger, in D. L. Cairns and R. A. Knox (eds.), Law, Rhetoric and Comedy in Classical Athens. Essays in Honour of Douglas M. MacDowell, Swansea, 187–203. Sanders, E. (2012) ‘He is a Liar, a Bounder, and a Cad’: the Arousal of Hostile Emotions in Attic Forensic Oratory, in A. Chaniotis (ed.), Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, Stuttgart, 359–387. Wallace, R. W. (1994) The Athenian Laws Against Slander, in G. Thür (ed.), Symposion 1993. Vorträge zur griechischen und hellenistischen Rechtsgeschichte, Cologne/Weimar/Vienna, 109–124. Zelnik-Abramovitz, R. (2005) Not Wholly Free. The Concept of Manumission and the Status of Manumitted Slaves in the Ancient Greek World, Leiden/Boston.

‘NEGATIVE’ EMOTIONS AND GREEK NAMES Nikoletta Kanavou 1 BACKGROUND: NEGATIVE EMOTIONS IN GREEK THOUGHT1 Emotions pervade most aspects of ancient Greek life and culture. Language expresses emotions, and the grammatical category of personal names is no exception. Greek personal names are significant more often than not.2 Although their meaning is a frequent topic of discussion, this is the first systematic study of emotions as semantic sources for onomastic material. It focuses on the relationship between personal names and ‘negative’ emotions – such as anger, envy, fear, hatred, and grief. With respect to the significant classification of personal names, the classical studies are those of Friedrich Bechtel and August Fick.3 These scholars recognise a group of names that derive from the ‘Geistiges Wesen’ – the spiritual or intellectual essence of a person – proposing further division into subcategories such as ‘Intellect’ and ‘Temperament und Character’;4 or ‘Intellect’ and ‘Gemüth’, with subdivision into ‘Temperament’ and ‘Charakter’.5 However, the names collected by Bechtel mostly reflect emotional behaviour and temperamental or character traits rather than emotions per se. For instance the names Ἄγριος (‘fierce’), Χάλεπος (‘ill-tempered’), and Βίαιος (‘violent’) express behaviour that can emerge from anger, without being directly related to ancient Greek designations of the emotion ‘anger’ (on which see further below). It is clear that this study will need to rely on a well-defined set of terms of emotions; assumptions about the meanings of emotional terms are also important for the explanation of significant names deriving from these terms. This study follows the tendency of modern scholarship to regard perceptions of emotions as socially conditioned and not universal. Ancient Greek emotions do not coincide in spirit with emotions found in modern lists of emotions, nor are modern English terms an accurate match for ancient Greek emotional vocabulary.6 The philoso-

1

2 3 4 5 6

I am grateful to the project ‘The Social and Cultural Construction of Emotions: The Greek Paradigm’ for a scholarship that enabled me to research the topic of this paper, and to Angelos Chaniotis for numerous useful comments and suggestions. E.g., Kanavou 2011, 1–4. Bechtel 1898; 1902; 1917; Fick and Bechtel 1894. Bechtel 1917, 498–506. Bechtel 1898, 52–69. E.g., Konstan 2006, 15f.

168

Nikoletta Kanavou

phical analysis of emotion was introduced by Plato,7 but the most substantial discussions of particular emotions were provided by Aristotle in the second book of the Rhetoric, where an awareness of emotions is presented as a prerequisite in the orator’s quest for εὔνοια (‘goodwill’),8 and by the Stoics.9 Aristotelian and Stoic categorisations are all based on evaluation, but on a generic level Aristotle operates with pleasure and pain, while the Stoics distinguish four γένη (pleasure, pain, fear, desire) with numerous subspecies.10 For reasons that will become clear below, Aristotle’s classification provides a more appropriate methodological basis for the present study. Apart from Aristotle’s terms, synonymous designations used by other authors, both prose writers (not just philosophers) and poets, will be sought out and investigated as possible sources for personal names. Terms discussed here will not include the vocabulary associated with emotional behaviour (for example, lamentation – associated with grief; terms of aggression and violence – associated with anger; etc.), which is vast and requires research on a much larger scale than the space here allows. Despite the philosophers’ help, the identification and interpretation of terms of emotion in the Greek texts are not always straightforward, let alone making a distinction between ‘positive’ and ‘negative’ emotions. Admittedly, such a theoretical distinction, which features frequently in modern scholarly discussions,11 is absent from the Rhetoric. Aristotle is more concerned with the justification of emotions in the instances where they appear than with positive or negative valence.12 But this study will take as its premise that in ancient Greek texts, including the Rhetoric, certain emotions are widely associated with pain, unpleasantness or even vice. Aristotle admits that emotions (πάθη)13 can cause pain (λύπη) or pleasure (ἡδονή),14 while he assigns to many emotions (for instance, anger) an ambivalent nature and a potential connection with λύπη. Aristotle presents emotions with a discernible negative aspect as interconnected.15 The presentation of certain emotions further entails an implicitly negative moral judgement: hate is incurable, while envy characterises bad people.16 For our purposes, ‘negative’ 7

8 9 10 11 12 13 14 15 16

See Knuuttila 2004, 5–24, for a survey of Plato’s views, especially in the Republic 429c–d, 430a–b and Philebos 47e; lists of emotions are in Laws 191d; Symposion 207e; Theaitetos 156b. Notably Longinos (22.1) speaks of πολλὰ … καὶ ἀναρίθµητα πάθη (καὶ οὐδ᾿ ἂν εἰπεῖν τις ὁπόσα δύναιτο), mentioning anger, fear, indignation, and jealousy as examples. See also Zaborowski 2012. This is one of three means of persuasion, the other two being φρόνησις (‘good sense’) and ἀρετή (‘virtue’); see Rhetoric 1378a7–10. Preserved in Diogenes Laertios 7.111–116; cf. Ps-Andronikos, On the emotions 1.1–5. For a comparison between the Aristotelian and the Stoic treatment of emotions, see Striker 1996; further on the Stoic categories, Sorabji 2000, 29–54. See most recently Tappolet et al. 2011. Konstan 2006, 33f.; cf. Kristjánsson 2007, 49–66. On the term (literally ‘passions’), see Konstan 2006, 3f. Rhetoric 1378a22–25. Rhetoric 1382a2f.: anger connected with hatred; 1382a36f.: fear connected with anger. Cf. Plutarch’s parallel treatment of hatred and envy in his essay On envy and hate. Rhetoric 1382a, 1388a.

‘Negative’ Emotions and Greek Names

169

emotions include both those that are evaluated as negative in a moral sense (such as hate and envy) and those that are psychologically negative, that, in other words, involve a disagreeable feeling, even if they are not morally negative (such as shame and pity). Both of these categories induce λύπη in their Aristotelian context: this designation is used here as a guide to defining ‘negative’ emotions, while there is otherwise no intention to deny the complexities associated with construing ‘positivity’ and ‘negativity’ in emotions.17 The collection of onomastic material attached to the various terms of negative emotions will largely depend on the Lexicon of Greek Personal names (LGPN), available in the form of printed regional volumes, as well as a searchable online database. 18 LGPN is now the richest source for documentary names, with vast geographical coverage19 and a chronological range that covers the whole of Greek antiquity up to the early Byzantine period. It is somewhat inconsistent in its treatment of literary names (relevant entries are sometimes included, sometimes not), for which additional sources are used here.20 This chapter further explores the reason behind the choice and use of personal names that relate to negative emotional terms. In real life, personal names are generally expected to express desirable characteristics (often the parents’ wishes for a child’s future and abilities) and to have a positive sound.21 Indeed modern anthropological research suggests that a name may affect a person’s self-image and development.22 This makes the origin of names with a negative meaning hard to explain: they may have been attributed apotropaic powers – that is, they were possibly thought to scare away evil, probably the kind of evil that derives from φθόνος (‘envy’), by making children appear unattractive;23 or they may originate 17

18

19

20 21 22 23

For a taste of the many and varied factors that influence this distinction, see Tarnopolsky 2010, 174f. See also Ben-Ze’ev (2000, 56f.; cf. 103), who notes that, though emotions are often associated with both pleasure and pain, ‘typical emotions have an overall positive or negative value’. A sense of an overall evaluation can be gleaned from some Aristotelian descriptions, certainly of envy vs. jealousy (Rhetoric 1388a36–43). On modern theoretical views of the evaluative aspects of emotion, see also Lyons 1980, 70–91 (esp. 90f.); for further bibliography, see Konstan 2006, 21. http://www.lgpn.ox.ac.uk/. Its electronic search engines allow a wide range of searches for emotion-related semantic particles. Notably the new .xml search displays full lexicon entries (http://clas-lgpn2.classics.ox.ac.uk/). One notable exception is Egypt, the only body of LGPN material that is not yet computerised. The main source is thus still Preisigke 1922, supplemented by Foraboschi 1967; more recent additions are found in the Heidelberg papyrology project Wörterlisten (http://www.zaw.uniheidelberg.de/hps/pap/WL/WL.pdf). Name searches are also facilitated by the search engine available on www.papyri.info. E.g. von Kamptz 1982 [1958] for Homeric names. Kanavou 2011, 2 note 5; Masson 2001, 226. As Golden notes (1986, 246), ‘initially at least the choice of a name tells us more about parents than about children.’ See Golden 1986, 246–252, with references. Golden uses anthropological literature as a frame of reference for examining naming practices in Classical Athens. On phthonos and the evil eye, see Rakoczy 1996, who collects examples from a broad selection of Greek literature. He notes that an apotropaic function is often associated with strange and laughable objects and names (ibid. 173). On apotropaic names, see Hobson 1989,

170

Nikoletta Kanavou

in nicknames.24 The answer lies in the particular circumstances of the naming, which may be difficult to reconstruct.25 This is not to imply that every name gave a discernible meaning; most did, but some (the so-called ‘irrational compounds’) are based on combinations of elements that do not appear meaningful.26 A possible factor in the choice of apparently ‘meaningless’ names, as well as of many significant ones, is ‘linkage’: members of the same family, especially of fathers and sons, used composite names that shared a common element, that is, the same word or two words of similar meanings – for instance, Hippo-krates, son of Hippias, or Orgilios (from orge), son of Oxycholios (from cholos; see below p. 175). This creates a bond of partial homonymy, is a signpost of tradition, and classifies people by marking their position within a community or kinship group27 – we shall see that such elements include emotional terms. Onomastic evidence is far more abundant for ancient Greek men than for women, whose social role was by nature very limited, especially in classical antiquity.28 Literary names of negative meaning constitute a different issue: names of fictional characters are often chosen or invented ‘to supplement physical, mental and behavioural descriptions’ of these characters,29 while contexts usually offer justification. A closer look at these names will contribute to a fuller understanding of literature’s psychological preoccupations: like most significant names, character names that express emotions are interconnected with literary plots (in particular with psychological plot themes); readers/audiences may perceive such names as symbols/encapsulations of these themes, perhaps even as triggers of particular aspects of the plot.30

24 25

26 27

28

29 30

163f. (‘derogatory-protective’ names is an alternative term, as they were believed to protect children by dispelling the danger posed by evil forces); Masson 2001, 260–263; Dasen 2011, 304. Perhaps the most discussed such names are these from κόπρος (‘dung’), the so-called ‘kopronyms’, such as Κοπρεύς, Κόπρων, and Κοπρία. On nicknames in Greek historical and literary onomastics, see e.g., Ghiron-Bistagne 1988. Hobson 1989 provides a useful theoretical framework for this issue, taking into account both the official function of names on documents and their possible social and anthropological value as forms of address. She rightly notes that the study of this latter aspect in ancient texts is at a disadvantage compared to anthropological study in contemporary societies. On these, see Morpurgo-Davies 2000, 18f., 21; Masson 1990, I, 88f.; II, 404f.; 2001, 226. See further Golden 1990, 25 (Classical Athens) and 1986, 257–269 (Athens and Sparta); he calls such names ‘linked names’; they become more common after the Archaic period and especially after 400 BCE. On the connection between names and group affiliations (esp. family membership and family continuity), see also Alford’s classic anthropological study (1988, 54f.). Cf. Solin 1990, on names of closely related persons denoting similar concepts. In Attica, known men outnumber known women by 10:1 (LGPN II, vi note 4). Hence we know less about the naming of women in real life, but it may have followed the same principles used for men (see Kanavou 2011, 21). Golden 1990, 25 notes that, though the number of women with known patronymics is much smaller, there are enough indications that they tended not to be linked by name with their fathers (but some were named after grandmothers, just as grandsons were often named after grandfathers). Thus Lindauer 2009, 79. He rightly points out (ibid. 82) the difficulty of distinguishing ‘positive’ and ‘negative’ emotions in literature. Cf. Lindauer 2009, 82–86.

‘Negative’ Emotions and Greek Names

171

2 METHODOLOGICAL ISSUES The study of this paper’s topic is bound to run into some complications, some of which are common to onomastic studies, while some are relevant to the particular type of names discussed here. Before proceeding with the names, it is necessary to warn the reader about certain methodological problems that we shall encounter on the way. The first of these problems concerns emotion-related words that produce personal names. In some cases they can have a variety of meanings, not directly relevant to emotions. One example are terms such as ὀργή and θυµός and their range of meanings. We shall need to make as much use as possible of contexts and other external indications in order to determine the meaning of the relevant names. Part of our material consists of names of doubtful Hellenicity; these are usually names from the cities of Asia Minor and Egypt. From the Hellenistic period onwards the use of Greek or Greek-sounding names for members of indigenous populations became increasingly common, but the continuation of indigenous traditions resulted in mixed onomastic material that is at times hard to sort out. Some names may be partially or totally Hellenised. In any case, etymology is sometimes elusive. Names whose derivation from a negative emotion is not certain will not form part of the main discussion (but their existence will be acknowledged in the footnotes). In many cases etymology will lead us to seemingly clear meanings. However, perceptions of the meaning of personal names deriving from emotional terms cannot have been static, but were subject to the influence of the changing philosophical and cultural climate. For example, Aristotle recognised that emotions express character and associated some of them with ἦθος χρηστόν (good moral character), others with its opposite (φαυλότης). However, this does not always relate them to virtues and vices.31 Some of his emotional terms are later used with a shift of focus onto the moral side; most characteristically, they are found in several early Christian writings as κακίαι, also called πάθη, the Aristotelian term for the emotions.32 Names that are semantically related to a negative feeling or vice are troubling as to the conditions and implications of their use. A further methodological difficulty emerges from the fact that, as will be shown below, negative emotions are often used for the construction of names with the privative alpha, which have positive meanings, as Alypos (a-lypos), that is, free from λύπη (‘grief’), and Acholios (a-cholos), that is, free from χόλος (‘anger’). The cultural significance of such names becomes clear only when they are examined together with names that express the same significance but in a positive way. Both types of names clearly function as wish-names. For example, names 31 32

Cf. Fortenbaugh 2008, 41. Cf. Paul’s catalogues of vice, used in his Letters in the context of moral exhortation; e.g., Romans 1.29–31, 13.13; 1 Corinthians 5.9–11, 6.9f.; see especially his remarks on anger in the Ephesians 4.26f. In the second century CE Hermas notes that θυµός and ὀργή can be transformed into ἁµαρτία: Hermas, Pastor 34.4f.

172

Nikoletta Kanavou

wishing someone to be free from grief (e.g., Alypos) are analogous to those wishing someone to be happy (e.g., Chaireas). Similarly, Acholios is analogous to Praylios (‘meek, gentle’). Examples of such parallel names will be given in the course of the discussion, but a full treatment of these lies outside the scope of the present study, which focuses on the reception of negative emotions as reflected in the onomastic material. With these observations in mind, we now proceed to the survey of ‘emotional’ names. A basic aim of this survey is to add to the existing knowledge both on name formation and on the reception of emotions in everyday life; more specifically, it should inform us about the comparative value that name-givers (ancient Greek society) attributed to particular emotions. 3 THE NAMES Personal names attested in both documentary and literary sources are arranged and discussed in connection with the ‘negative’ emotion to which they are attached. Each subsection begins with a justification of the inclusion of the selected emotions in this study. 3.1 Anger The Aristotelian term for anger is ὀργή, and the emotion is introduced as ὄρεξις µετὰ λύπης (‘a longing accompanied by pain’).33 Although there is also a certain pleasure in this emotion (derived mainly from the desire to avenge slight), its association with pain suggests a distinct negative aspect, corroborated by comments of other ancient authors.34 The above seem to allow the assumption that personal names attached to ὀργή could have a negative sound. Before discussing the names, it is necessary to explore the semantics of ὀργή further. The word, first attested in the Homeric Hymn to Demeter (205), originally had the neutral meaning ‘character’, ‘disposition’, and acquired the more specialised sense of ‘anger’ in the Classical period. Orge can also signify intense emotion in a generic sense, ‘passion’ (cf. the verb ὀργάω ‘be filled with passion’).35 Related vocabulary includes compound adjectives ἄνοργος – Hellenistic ἀνόργητος – (‘not angry’), δύσοργος (‘bad-tempered’), εὐόργητος and εὔοργος

33

34

35

Rhetoric 1378a; at 1379a, Aristotle further describes anger as a by-product of a painful physical or psychological state, and as a reaction to hurtful and offensive behaviour of others; cf. Nicomachean Ethics 1117a5–15: καὶ οἱ ἄνθρωποι δὴ ὀργιζόµενοι µὲν ἀλγοῦσι. All translations of passages from the Rhetoric used here are by Freese 1926 (Loeb). E.g., Antiphon, 5.72, who notes that anger clouds one’s judgement; in the Nicomachean Ethics 1109a20–30, Aristotle presents anger as a potential vice, inasmuch as it exceeds the µεσότης. See further Konstan 2006, 69–76. Chantraine 1999, s.v. 1 ὀργή.

‘Negative’ Emotions and Greek Names

173

(‘good-tempered’, ‘free from passion’), but none of these has left its trace in onomastics.36 Names from ὀργή are few and rare and come from the periphery of the Greek world. There are a couple of ᾿Οργίλιος, one from Ionia (Magnesia, around 300 CE, LGPN Va) and one from the Roman province Moesia Inferior (Odessos/Varna in north Bulgaria, fifth-sixth century CE, SGLIBulg 96). The Magnesian Orgilios had two names, Telephos Orgilios, an interesting combination if we remember the origin of the hero Telephos from Asia Minor and his role in myth and literature.37 However, it may also be that Telephos was the Greek name adopted by someone whose original name was not Greek.38 The latter (and later) Orgilios is the son of an Oxycholios (see below on names with cholos, another anger-related term) – perhaps a deliberate semantic pairing.39 The similarity of ᾿Οργίλιος to the adjective ὀργίλος ‘anger-prone’, ‘passionate’, certainly allows the possibility that this name evoked an emotion-related sense, but it is hard to tell whether it is anger or (more generically) passion. A slave name such as Ὀργή40 invites imaginative explanations: it may refer to the slave’s bad temper or it may commemorate a particular event, perhaps related to a former owner’s angry behaviour.41 Aside from Aristotle’s main term orge, there is θυµός. The term covers a rather broad semantic range (‘soul’, ‘life’, ‘spirit’, ‘courage’, and ‘anger’) and has 36

37 38 39 40

41

The rare personal names in -orgos (Φιλέοργος, Τίµοργος; see also Bechtel 1917, 161f.) are related to ϝεργ- and not to ὀργή (though Chantraine 1999, s.v. 1 ὀργή notes that ὀργάω is also used of fertile soil and plants (‘swell with produce’), and deduces that there may be ‘des interferences’ between this group and the compounds or derivatives of ϝεργ-). About half of the name’s fifty-six attestations (LGPN) are from Asia Minor. Cf. Colvin 2004, 62–64, on the use of Greek mythological names in Lycia. A case of linked names, see above note 27. Solin 1996, 565. On the significance of slave names, which often express common notions of the character and external appearance of slaves, see Kanavou 2011, 197–202 (with bibliography). The following names are probably not relevant to orge (for references see LGPN): 1) ᾿Οργεύς (genitive ᾿Οργέος), a prominent Thasian of the early fifth century BCE, father of Antipatros. The name must allude to ὄργια ‘secret rites’, ‘the rites of Bakchos, orgies’ and denote ‘priestly or orgiastic functions’, associated with the man’s place of origin (thus R. W. Macan’s comment on Herodotos 7.118); prominent Thasian coin types indeed reflect the orgiastic worship of the Thracian Bakchos. See Picard 1982 on the types of Thracian coinage of the Classical period: two out of three main series (Silenos and nymph, bearded Dionysos) are relevant to Dionysian worship. 2) ᾿Οργιάλης or ᾿Οργιαλεύς (genitive Ὀργιάλεος), father of ∆ελφίνιος (Classical period), from Sinope, an area surrounded by an indigenous population. This name too is more likely derived from orgia. Orgia were celebrated for Apollo Delphinios in Miletos, the mother-city of Sinope; cf. the son’s name, ∆ελφίνιος. 3) ῎Οργασθυς, a citizen of Amastris buried in Athens (second century BCE). Robert 1963, 449–453, associated it with a family of Paphlagonian names from the mountain name Olgassys (cf. Ολίγασυς, Όλγασις; Orgasthys may be the product of phonetic variation), but its meaning could have been naturalised into Greek, a possibility increased by the fact that the man’s father had a good Greek name (Agathokles). On naturalisation of foreign names into Greek, see Zgusta 1965, 95–97; on the factor of phonetic similarity between Greek and indigenous onomastic elements, which probably assisted transposition, see also Colvin 2004, 65–67.

174

Nikoletta Kanavou

an important place in Plato’s theory of the soul as one of the soul’s three parts; the Platonic thymos is hard to pin down, but seems to be associated with a fighting spirit and noble passion. The Stoics apparently used it for a type of orge – more specifically, the early stages of anger.42 The term is used for anger by Aristotle in the relevant chapter of the Rhetoric that discusses this emotion – notably right before a quotation from the Iliad that illustrates the pleasure associated with anger.43 Although some of the vocabulary that relates to thymos expresses the meaning ‘anger’ (θυµόω/όοµαι ‘I anger/am angered’, θυµικός ‘angry’, θύµωµα ‘anger’), most of the thymos compound adjectives do not, and this is where personal names usually come from. The LGPN database contains 73 different names with the element -thym-, but none of these seems to have anything to do with anger.44 A simple name Θῦµος occurs once in Hellenistic Macedonia (LGPN IV) and in Egypt.45 It is also attested as a slave name,46 whose significance may be similar to that of the already mentioned Orge. The similar name Θύµων has two Hellenistic attestations (Arkadia and Megara, LGPN IIIa-b). Another term for anger is the epic χόλος. Its meaning is not identical to ὀργή; it includes a sense of bitterness (‘bitter anger’, LSJ), and its semantics seem to be more overtly associated with unpleasantness than those of orge.47 Χόλος is related to χολή ‘bile’. It is this latter term that is echoed in compounds with this root, according to Pierre Chantraine,48 who notes however that these ‘peuvent comporter des acceptions morales’ which are akin to χόλος. This appears to be true for at least some of the relevant personal names. One of them is Ὀξυχόλιος, reminiscent of the adjective ὀξύχολος ‘quick to anger’. The adjective is ancient (attested as early as Solon, 13.26). Oxycholios belongs to the group of names in -ios, which are very common from the second century CE and extremely common in Late Antiquity; they became common under Roman influence (-ius). The personal name only has a few late attestations, from Thessaly of the Byzantine period (LGPN IIIb), Philippi in Macedonia (a 42 43

44

45 46 47 48

Diogenes Laertios, 7.114. Rhetoric 1378b5. Cf. a further quotation from the Iliad (2.196) at 1379a5, where θυµός refers to the disposition of kings, who ‘are naturally prone to anger’ (thus Kirk 1985, 136 ad loc.; ἀγανακτοῦσι for Aristotle) or ‘have a proud heart’ – despite the Aristotelian interpretation, the example bears witness to the semantic ambiguity of θυµός. These include the common compounds Εὔθυµος, Πρόθυµος and Θυµοχάρης (cf. the adjectives εὔθυµος ‘well-disposed, kind’ or ‘cheerful’; πρόθυµος ‘willing, eager’); also Ἀγάθυµος, Θυµοκλῆς, Θυµοκράτης, the hapax Θυµοµένης (LGPN IIa, in a casualty list of 409 BCE without information on his parentage), and one Θυµολέων from Egypt (WL). There are also occasional Ηellenistic attestations of Θυµωΐδας/ης (LGPN I, Rhodes and Delos; Preisigke: Egypt), apparently deriving from the adjective θυµώδης = θυµοειδής, which can mean ‘courageous’, ‘passionate’, but also ‘hot-tempered’, ‘angry’. However, the fact that θυµοειδής is offered by Plato as an opposite of both πρᾳεῖα φύσις (Republic 375c) and ὀργίλος (ibid. 411c) suggests a certain versatility of meaning. Foraboschi 1967. There is one doubtful late attestation from Syria (SEG XXXVII 1452). Solin 1996, 583. On the Greek terminology of anger, see further Harris 2001, 50–68; Konstan 2006, 41–76. Chantraine 1999, s.v. χόλος.

‘Negative’ Emotions and Greek Names

175

Jew, third or fourth century CE, LGPN IV), and Aphrodisias (another Jew). This last occurrence is on one of two inscriptions citing numerous names of Jews that have been noted to express positive character traits.49 However, the negative associations of oxycholos are hard to ignore. The Oxycholios encountered earlier as the father of an Orgilios (see p. 173) must allude to ‘anger’. Ὀξυχολία appears personified as a woman clothed in black in Hermas.50 The date of the attestations of the personal name might in some cases encourage the allusion to ὀξυχολία as a sin. The name must in any case originate in a nickname. One Χόλος is found in Egypt.51 Χόλος is further found in Monika Hasitzka’s list of names from Coptic documentary sources;52 she takes it as a Greek name. There are also some rare names from ἄχολος ‘lacking gall’, metaphorically: ‘gentle’, ‘allaying bile or anger’ (LSJ). The same sense is positively expressed in names such as Πραΰλος/Πραΰλιος (from πρᾶος ‘calm’) and ῎Ηπιος (‘mild’). ᾿Αχόλιος seems to occur mainly in Asia Minor, not earlier than the Imperial period: there are two Ionian occurrences, three from Caria, and isolated attestations in Lydia, Phrygia, and Egypt.53 It is attested for Jews.54 It was also the name of a bishop of Thessalonike of the fourth century, and perhaps came to indicate Christianity. The form ᾿Αχόλις has a couple of attestations in Lydia and Phrygia. The feminine ᾿Αχολίς is attested once in Thessalonike of the Imperial period (LGPN IV). Another epic term for anger is µῆνις, ‘wrath’. According to Diogenes Laertios, 55 the Stoics counted it among the emotions and commented on its relationship to orge and thymos: they defined µῆνις as orge that lasts long and has become malicious, citing Iliad 1.81f. as an example (though the terms used there are χόλος and κότος). Unlike the other two anger-related emotions, menis has not produced any personal names.

49 50

51

52 53 54

55

Reynolds and Tannenbaum 1987, 103; Chaniotis 2002, 217f., 229, 234, suggests the meaning ‘spirited’ (229 note 71); see also below, note 54. Pastor 92.3.5. Other personifications in this context are Λύπη and Μῖσος, also discussed above as negative emotions. The personification of vices and virtues is found in earlier popular philosophy, and is also practised in Hellenistic Judaism, where vices (‘jealousy’, ‘wrath’, ‘grief’, ‘bitterness’, ‘quarrelsomeness’) are variously presented as sins, passions, or demons. See further Mussies 1981, esp. 315–319 (with reference mostly to the gnostic treatise On the origin of the world). Cf. Philo, De Cherubim 43, where ἀρεταί (as a general term and not as specific virtues) appear as personifications. Preisigke 1922; cf. Wörterlisten. Note also the following doubtful names: Χολλᾶς, single attestation from Ionia (Priene, Hellenistic period LGPN Va); from Egypt: Χολ(λ)ῶς and Χολῶτις (Preisigke 1922; cf. Wörterlisten); Χολλαῦθις, Χολωµῆις (Foraboschi 1967 – possible hybrid names?; cf. below, note 89). www.onb.ac.at/files/kopt_namen.pdf (version of January 2007). Ionia: LGPN Va; Caria: LGPN Vb; Lydia: Sardis VII.1.83; Phrygia: Judeich 1898, no. 79; MAMA VI List 149, 168; Egypt: see Preisigke 1922; Foraboschi 1967; Wörterlisten. One of the attestations is on the already mentioned Aphrodisian inscriptions, on which see Chaniotis 2002, 217f., 229, who associates the name with moral values (in particular ‘willingness to behave in a good manner’). Diogenes Laertios, 7.114.

176

Nikoletta Kanavou

3.2 Hatred and Enmity The Aristotelian terms are ἔχθρα (ἔχθος), ‘enmity’, ‘hostility’, and µισεῖν, ‘to hate’. These are placed together in the Rhetoric treatise, which does not appear to distinguish between the two terms.56 Other uses, however, indicate that, though closely related, they are not identical in sense (it is µῖσος that expresses ‘la haine proprement dite’).57 The negative weight of these emotions is implied by Aristotle’s opening phrase in the relevant section: ‘The causes that produce enmity are anger, spitefulness, slander.’ Enmity is indeed treated as a relative of anger58 as they both involve hostile behaviour and provoke pain, but their function differs both in quality and degree: µῖσος is unrelenting and more malicious than ὀργή.59 Plutarch in On envy and hate60 notes that the intention of hate is to injure. Failure to locate personal names attached to µισέω or στυγέω or other synonymous terms comes therefore as no surprise. The comic character Βδελυκλέων of Aristophanes’ Wasps (from βδελύττοµαι ‘to detest’, ‘to feel disgust at’) is a unique exception – though, as David Konstan notes, ‘the core idea [of bdelyttomai and related terms] is that of being revolted by someone or something’,61 which is related but not identical to µισέω. Through the name, revulsion against Kleon rises to the status of one of two primary comic emotions in this particular play (the other being ‘love for Kleon’, expressed in the antithetic name Philo-kleon). Significantly, the negative emotion wins (as the plot moves away from Kleon to merrier subjects), earning justification from the vile nature of its object. Ἔχθρα and ἔχθος have hardly produced any names; one possible explanation of their use is that they may evoke a sense of enmity inherent in the fear or respect inspired by power.62 The only certain cases are that of Ἔχθρων in Dodona, third century BCE (LGPN IIIa) and Ἐχθάτιος, with three Hellenistic attestations from Thebes (LGPN IIIb). In a military list from Hellenistic Kyrene the form Ἐχθατίαν appears, apparently a masculine name. His son also has a name related to emotions: Αὐτόφιλος (‘the one who loves himself’).63

56 57 58 59 60 61 62 63

Rhetoric 1382a. Cf. Grimaldi 1988 ad loc. Chantraine 1999, s.v. ἔχθος. Similarly in Politics 1312b25–34 it is said that hatred must include anger. For a modern discussion of the relationship between these two emotions, see Ben-Ze’ev 2000, 393f. E.g. Rhetoric 1382a9-10. See further Konstan 2006, 185–200. Moralia 536e–538e. Konstan 2006, 198f. Cf. the Athenians’ observation in the Melian dialogue (Thucydides 5.95) that the power of Athens arouses ἔχθρα. LGPN I; SGDI 4835, line 13 = CIG 5146, line 11.

‘Negative’ Emotions and Greek Names

177

3.3 Envy According to Aristotle,64 envy (φθόνος) is an emotion found in bad people (τὸ δὲ φθονεῖν φαῦλον καὶ φαύλων); notably it is the only emotion that he treats as ‘unqualifiedly negative’.65 Plutarch compares it to hate in his relevant treatise (On envy and hate). It is not surprising that the term has a meagre presence in onomastics, and apart from one occurrence of the noun as personal name, the rest of the names involved express the opposite of φθόνος. Phthonos may also refer to envy personified: references from the Classical period suggest a malicious spirit,66 usually affecting people at their peak of youth and glory. A fragment of the fourth-century BCE comic poet Timokles67 mentions Phthonos as most unholy to the living, while a tragic fragment calls him κάκιστος κἀδικώτατος θεός68 – but there is no evidence of cult. We might infer that any use of the personal name had an apotropaic purpose. Φθόνος has a single attestation from Hellenistic Rhodes (LGPN I, IG XII.1.364); Hermodokos, son of Phthonos, is unfortunately all the text we have, and gives us no hint as to the reason behind the name’s choice. There is a Φθόναρος from Egypt (close in sound to φθονερός, ‘envious’), and a fragmentary name with Φθον-, also from Egypt.69 A character named Phthonos appears in the epilogue to Kallimachos’ Hymn to Apollo: he personifies the envy that Kallimachos ascribed to his antagonists. A few names, all post-Classical, are attached to the adjective ἄφθονος, whose semantic field includes ‘free from envy’, ‘ungrudging’, ‘unenvied’, ‘provoking no envy’, but also ‘plentiful’ (common in Attic, but also in earlier literature), which takes a step away from an emotion-related meaning. It is hard to decide whether a number of rare names (Ἀφθονᾶς, Ἀφθονία, Ἀφθονίδας, Ἀφθοννώ, Ἄφθονος) suggest one sense or the other.70 The commonest name in the φθόνος group, however, is the widespread Ἀφθόνητος/Αφθόνειτος from the adjective ἀφθόνητος (cf. also Ἀφθονήτα/Ἀφθονείτα),71 which seems only to occur with the meaning ‘unenvied’ or ‘free from envy’.72 There are 93 occurrences, none earlier than the Hellenistic period. It is close to the meaning of the common name Ἀβάσκαντος.73 64 65 66 67 68

69 70 71 72 73

Rhetoric 1388a. Thus Konstan 2006, 113 (and 111–128 for a detailed analysis). Plato, Phaidros 247a; Sophocles, Philoktetes 776–778; Euripides, Trojan Women 766–769. PCG 33 edd. Kassel/Austin. TrGF I² 210, fr. 2 (Hippothoon); cf. TrGF II fr. 6b (adespoton). See Bernert 1941b (on Phthonos as a deity) and Gisler 1997. A memorable depiction of the personified Envy, ugly, sickly, and clothed in black, is found in Lucian’s description of Apelles’ painting Diabole (Slander 6–8). Preisigke 1922. The names Ἀφθονία and Ἀφθόνιος (not found elsewhere) are attested for Egypt (Preisigke 1922; Foraboschi 1967, Wörterlisten). LGPN IIIb. For ει = η in Thessalian and Boiotian, see Buck 1955, 25. In early poetry: Aeschylus, Agamemnon 939; Pindar, Olympian 11.7; it resurfaces in Eustathios and lexicography. Phthonos is discussed by Konstan 2006, 111–128, together with another Aristotelian emotion: τὸ νεµεσᾶν, ‘indignation’ (or, in Aristotle’s definition in the Rhetoric 1837a8–9, ‘feeling pain

178

Nikoletta Kanavou

3.4 Fear Fear can be evaluated as a negative emotion: Aristotle introduces it as λύπη τις and as a reaction to certain σηµεῖα, among which are anger and enmity.74 Most names here are built with the privative alpha and denote the absence of fear; this is no coincidence and underlines the negative value of this emotion. These names are of course synonymous with names denoting concepts of courage and bravery (such as Euandros and the numerous names from tharrhos/tharsos). Some names in the φόβος-group75 may reflect the notion of fear as a weapon in the hands of those who provoke it in others. A simple name such as Φόβερος from φοβερός, ‘fearful, terrible’,76 with a single attestation in Aigiale on Amorgos (Imperial period, LGPN I), probably implies the one who provokes fear. With Φόβος, however, it is not easy to tell. As with Phthonos, Phobos’ presence in onomastics may be associated with a personification of the same name. Phobos appears in the Iliad as personified fear and retreat.77 Hesiod, who mentions him as a son of Ares,78 suggests a divinity. The seven heroes who attack Thebes swear by him.79 There is also evidence of a cult.80 But Phobos’ status as a personal name is weak. According to Plutarch, citing Charon of Lampsakos as his source, Φόβος was a leader of the Phokaians in Ionia;81 this is the only occurrence of the name. However, Polyainos gives a different name, Φόξος (presumably from the adjec-

74 75 76 77 78 79 80

81

at someone who appears to be succeeding undeservedly’ – a morally positive alternative to phthonos). The noun nemesis is the more familiar form (Nicomachean Ethics 1108a35: νέµεσις δὲ µεσότης φθόνου καὶ ἐπιχαιρεκακίας), and it is included in the Stoic lists of emotions ([Andronikos], On the emotions 1.2.1: νέµεσις δὲ λύπη ἐπὶ ἐπαιροµένοις παρὰ τὸ προσῆκον). But the noun also designates a personification and deity (Hesiod, Theogony 223), connected with concepts of punishment and retribution. The cult of Nemesis was prominent, especially in Rhamnous in Attica and Smyrna; she also received widespread private worship; see Hornum 1993; Lichocka 2004. The large number of attested nemes- names (which are particularly common in Egypt, see Lichocka 2004, 93–98) was clearly inspired by the personification/goddess, not from the emotion that Aristotle describes. Aristotle presents another emotion, pity (ἔλεος) as the opposite of to nemesan – psychologically negative mainly in the form of self-pity, in that it entails the realisation that one is not immune to the various causes of suffering (Rhetoric 1385b14; see further Konstan 2001, 10f.). The term seems to be absent from onomastics. Rhetoric 1382a24, 37. On which see also Bechtel 1917, 455. A passive meaning (‘afraid’, ‘timid’) is unusual (LSJ). 4.439f., 13.299f., 15.119f.; Hesiod, Aspis 195f., 463f. A comic adespoton (PCG 873 eds. Kassel/Austin) describes him as ugly or shapeless (ἀµορφότατος τὴν ὄψιν). Theogony 933–936. Aeschylus, Seven against Thebes 45. E.g., Plutarch, Theseus 27. John Boardman (LIMC VII s.v. Phobos) notes that he received cult in Sparta and before battle as a war-god; see further Mactoux 1993 and Stafford 2001, 113f. However, Bernert’s view of Phobos as a ‘real and active deity in common Greek belief’ (Bernert 1941a) may be somewhat exaggerated. On the cult of Phobos, see the discussion by Maria Patera in this volume (pp. 112–115). Plutarch, Moralia 255a–e; Charon of Lampsakos, FGrH 262 F 7a.

‘Negative’ Emotions and Greek Names

179

tive φοξός ‘with pointed head’)82 – an attested name, certainly a rare one, but found as early as the Classical period (Thespiai in Boiotia, LGPN IIIb). Another personal name from this root, the slave name Phobianus, is not certain either.83 Among names with phobos and the privative alpha, the commonest is Ἀφόβητος (twelve attestations, mostly from Attica and Macedonia, from the fourth century BCE onwards). Aphobetos was the name of a brother of the orator Aeschines (LGPN II no. 1), whose father was named Ἀτρόµητος – a case of ‘linked’ names. Other names include Ἀφοβητίδας (derived from Ἀφόβητος),84 Ἀφοβία, Ἀφόβιος, Ἄφοβος. Aphobos is mentioned by Olivier Masson85 as an example of the well-known tendency to bestow favourable or auspicious names on children: an Aphobos should never know fear! One of the oldest compound names with φόβος is certainly the Homeric ∆ηίφοβος, a son of Priam. Hans von Kamptz86 convincingly explained it as a compound with δήιος (epic for δάιος), ‘hostile’. Φοβέω in Homer usually means ‘put to flight’, thus the name should suggest someone who terrifies the enemy (cf. δηιάλωτος, ‘taken captive by the enemy’). The name’s first component should also be seen as a linking element, a link with the name of another son of Priam: ∆ηιοπίτης.87 There is a single attestation of the name ᾿Επίφοβος (Thera, Archaic period, LGPN I) from the adjective ‘frightful’; the noun ἐπίφοβος (‘who puts to flight’) occurs in Aeschylus.88 Some rare names are attached to the poetic words τρέµω, ‘I tremble with fear’, and τρόµος. They are nearly always formed with the privative alpha, and they mean absence of fear. There is one exception only:89 the appellation ὁ 82 83 84

85 86 87

88 89

Polyainos, 8.37; on the name, see Bechtel 1898, 21. Solin 1996, 569, tentatively connects it with φόβος but suggests Phoebianus as a more likely possibility (despite Aphobianus). Aphobetos has four attestations from Asia Minor (Hellenistic and Imperial period), while Aphobetidas only has a single attestation from Kyme in Aiolis, fourth or third century BCE (LGPN Va); see further Masson 1990, II 527: the name replaced an original reading Φοβητιδα. Masson 2001, 226. 1982 (1958), 73 (cf. 225: ‘Jage in die Flucht’). Iliad 11.420. On the etymology of this name, see von Kamptz 1982 (1958), 85f., who notes that ὄπις (‘regard’, ‘respect’, but also ‘vengeance’, ‘punishment’), allows more than one meaning: ‘der nach dem Feinde ausschaut’, ‘der am Feinde Rache nimmt.’ Notably the word is only found in Homer in the sense ‘divine punishment’ (Iliad 16.388; Odyssey 14.82, 20.215). A simple name Ὀπίτης is found in a list of Achaians slaughtered by Hektor (Iliad 11.301). Agamemnon 1152. There is one Θεοφόβιος from Late Antiquity (‘a prominent Lazian’; PLRE III s.v.) – but fear of god, that is, piety, is hardly a negative emotion. It is doubtful that a group of Τρεµ- names, apparently exclusive to Egypt, should belong here: Τρεµαῶς, Τρεµενῆσις, Τρεµαντίνοος; also Τρεµονσηθ- and two fragmentary names starting with Τρεµ- (Foraboschi 1967). Wörterlisten: Τρεµόνησις, Τρεµονσῆµις, Τρεµταοῦς, Τρεµῦνις, Τρεµχµοῦν, Τρεµψάεις, Τρερῆσις. The significance of names in this group is unclear. Hybrid names are a familiar phenomenon in Egyptian sources, but note also that Greek transcriptions of Egyptian names were not standardised, which adds to the difficulty of interpreting their meanings.

180

Nikoletta Kanavou

τρέσας for Aristodamos the Spartan who avoided Thermopylai according to Herodotos (7.231) – but this is treated by editors at most as a nickname (it is invariably printed with a small initial); it is perhaps indeed too shameful to acquire the status of a personal name. On the other hand, there are solid attestations of names with the sense ‘fearless’. Ἀτρόµητος has one Ionian and four Attic attestations from the Archaic to the Hellenistic period; in the Classical period, an Ἀτρόµητος was the father of an Aphobetos (LGPN II no. 3; the family of the orator Aeschines90). There is also one Ἀτρέσιος from Troas in the Imperial period (LGPN Va) and seven Ἀτρεστίδης/Ἀτρεστίδας (LGPN IIIa-b). The name Ἀτρεµαίων is recorded on a Cypriot syllabic inscription of the fourth century BCE.91 The heroic name Ἀτρεύς belongs to this group. A further poetic word that has given rise to personal names is τάρβος, ‘terror’. Ἄταρβος is the commonest, with nine attestations, mostly from Attica of the Classical period. Ἀταρβίων and Ἀταρβίδης only have Attic attestations (six and one respectively, mostly from the Classical period).92 Again, we observe here that the use of the privative alpha gives these names a positive meaning: those who are free of terror. 3.5 Shame Αἰσχύνη (shame) is another emotion that is associated with λύπη in Aristotle.93 It is induced by indecent behaviour (shame is an emotional reaction to bad deeds, present, past or future), hence it can be seen as a painful emotion, although it is clearly not a vice (its opposite, ἀναισχυντία ‘shamelessness’ is, briefly mentioned in the Rhetoric).94 However, αἰσχύνη may also denote ‘disfigurement or ugliness’ 95 (cf. αἶσχος ‘shame, disgrace’, but also ‘ugliness, deformity’). 96 It seems that the numerous names with the component aisch- (for instance, Αἴσχρων, Αἰσχίνης, Αἰσχύλος) are attached to this sense rather than ‘shame’.97 90

91 92

93 94

95 96 97

See also Demosthenes’ accusation (On the Crown 130) that Aischines altered his father’s name from Τρόµης to Atrometos once he embarked on a career in public life (he allegedly also changed his mother’s name from the rather inauspicious Empousa to Glaukothea) – a suspicious claim: Tromes is otherwise entirely unattested. The form of the name is a-te-re-ma-i-o-ne. See further Masson 2001, 303 who notes that the text lists ‘souscripteurs ou associés’. It is doubtful that the following names are relevant to τάρβος: Τάρβης in Macedonia (fourth century BCE; LGPN IV s.v.; SEG XLVI 804, line 3), the son of a Σεδέλης, a non-Greek name; Τάρβας in Paphos (fourth century BCE, LGPN I). Rhetoric 1383b15. 1385a15f. On the emotion of shame, see further Tarnopolsky 2010, esp. 172–196, who summarises traditional negative constructions of this emotion but makes a case for the positive character of shame, especially in the context of democratic politics. DGE I² s.v. αἰσχύνη (e.g., Plato, Laws 878c); αἰσχύνω is used in Homer in a physical sense (‘to disfigure’, Iliad 18.24), as well as in the sense ‘to dishonour’ (DGE I² s.v. αἰσχύνω). LSJ; in this latter sense also in Aristotle, Rhetoric 1386a11, 1405b7. Masson 2001, 317 n. 8, following Louis Robert, saw the nature of these names as ‘apotropaic’, ‘qui viserait à détourner l’attention des esprits du mal, jaloux et nuisibles, en

‘Negative’ Emotions and Greek Names

181

3.6 Grief / Sorrow Grief does not feature among the emotions discussed in Aristotle’s Rhetoric.98 The commonest Greek term for it, however, λύπη, accompanies the definition of each and every emotion that he construes as potentially ‘negative’. As mentioned in the introductory section of this study, λύπη was defined as a class of emotions by the Stoics (one of four classes of main or most universal emotions99), and its subspecies include some of the emotional terms that appear in this study: ἔλεος, φθόνος, ἄχθος, ἀνία, ὀδύνη. This implies a similar realisation that lype is present in many emotions, but while Aristotle presents it as an ‘element’, the Stoics define it as a ‘class or category name’. It is included in the present study because it clearly designated psychological suffering – an emotional state.100 Names with the component -lup- are, on the whole, not very common. The commonest is Ἄλυπος, with seventy attestations mostly from the Hellenistic and Imperial periods,101 and a wide distribution. Nearly all other attested names are also formed with the privative alpha: Ἀλύπαντος, Ἀλύπατος, Ἀλύπη, Ἀλύπης, Ἀλύπητος, Ἀλυπία, Ἀλυπιανή, Ἀλυπιανός, Ἀλυπιάς, Ἀλύπιος, Ἀλύπις, Ἀλυπίς, Ἀλυπις, Ἀλυπίων, Ἀλυπώ. Most have single attestations, and the commonest of the group after Alypos, Alypetos, is found a total of fifteen times. They all have more or less the same meaning: ‘without pain’, ‘causing no grief’. Thus they are equivalent in sense to names expressing notions of joy (such as the numerous chair-/char- compounds) and confirm that the sense of lype was largely (and unsurprisingly) undesirable in names. The date of the bulk of names with the privative alpha also indicates an influence of the Stoic ideal of apatheia, ‘lack of emotion’.102 The only name that goes against this trend is Λυπήτη, which however only occurs once (Athens, fourth century BCE).103 There are also some interesting compounds: Εὔλυπος (Athens, third century BCE), Τιµόλυπος (Kalapodi, Classical period, LGPN IIIb)104 – both only have single attestations; and Παυσίλυπος (three Hellenistic attestations from various places), reminiscent of a poetic adjective meaning ‘ending pain’.105 A common term for ‘grief’ in ancient Greek literature is πένθος; it has produced few and rare personal names, mostly related to myths and legends. There

98 99 100 101 102

103 104 105

face d’un enfant qualifié de “laideron”.’ Reference to moral failure is unlikely – it is too negative to be implied in a personal name (cf. Thompson 2007, 683). For an explanation of this omission, see Konstan 2006, 244–248. Lists preserved in Diogenes Laertios, 7.111–116. On grief as a basic and universal human emotion, see Nussbaum 2001, 169. One exception is the name of a sculptor from Sikyon of the Classical period (LGPN IIIa). Cf. the frequent use of alypos as a positive state at the time of death on funerary inscriptions of the Hellenistic and Imperial periods (e.g., SEG XVII 754; XX 412). On the Stoic ideal of freedom from passion and the ‘therapy’ of emotions, see further Knuuttila 2004, 66–80. One might add the name Λυπρός (from the adjective ‘wretched’, ‘sorry’) from Egypt (Foraboschi 1967). One wonders whether Τιµόλυκος is a possible reading – a better attested, though equally irrational name. Sophokles, TrGF IV fr. 425; Euripides, Bacchae 772.

182

Nikoletta Kanavou

are only two attested historical names: Πενθίλος, for a son of Orestes, progenitor of the Lesbian royal house of the Penthilids, and leader of the Aeolic colonisation;106 and another Penthilid.107 The name occurs again in Lesbos in the Hellenistic period, perhaps in remembrance of the Archaic Penthilos. There is one Πενθύλος from Paphos.108 In comparison, the term is found more frequently in literary and mythological names. There is of course Πενθεύς, whose legend was famously dramatised in Euripides’ Bacchae and whose name was believed already in antiquity to reflect his sad fate; as the tragic poet Chairemon (fourth century BCE) put it: Πενθεὺς ἐσοµένης συµφορᾶς ἐπώνυµος.109 There is also the Homeric Μεγαπένθης, son of Menelaos, whose name reflects an attribute of his father, who is ‘leidgeprüft’;110 it is not unusual for sons and daughters who are not as important in myth as their parents to bear names that reflect qualities of the parents.111 An echo of penthos is further heard in the name of the Amazon Penthesileia, famous for her involvement in the Trojan myth – she too had a sad end, as she died at the hands of Achilles according to the Aithiopis. The events are summarised by Proklos and narrated in detail by Quintus Smyrnaeus, where we encounter a number of puns; the Trojans are said to grieve for her, and one of the words used is πένθος: ἀκηχέµενοι µεγάλῳ περὶ πένθεϊ θυµόν (1.632); similarly for her father, the god Ares: Ἄρεϊ δ᾿ ἔµπεσε πένθος ὑπὸ φρένας ἀµφὶ θυγατρὸς (1.675).112 Finally, in a later and very different context, we encounter another Μεγαπένθης (‘Great woe’), son of Λακύδης, a rich tyrant, a fictitious character in Lucian’s Kataplous, one of the Underworld Dialogues.113 The name clearly expresses the psychological state of the character, a newly dead person who resents his fate and who wishes to be allowed back on earth.114 Friedrich Bechtel115 recognised a group of names attached to ἀνία (‘distress’), a term found in the Stoic lists. The word already appears in Homer, al106 Pausanias, 3.2.1. 107 Aristotle, Politics 1311b29. The entry Πενθίλη(α) as a female name in LGPN I (citing Sappho fr. 71.3 V. and Alkaios fr. 75. 10 V.; 302b V.) is incorrect. The fragmentary state of the poems allows no definite reading, but this is either the tyrant’s name (Penthilos) or an adjective from it. 108 Herodotos 7.195; Bechtel 1917, 369. 109 TrGF I² 71, fr. 4 110 Odyssey 4.11; 15.100, 103, 122; cf. the Homeric epithets νεοπενθής, πολυπενθής; von Kamptz 1982 [1958], 32. 111 Cf. the names of Agamemnon’s daughters (Chrysothemis, Laodike, Iphianassa – all reflect qualities of a king) and of Odysseus’ son Telemachos (a possible meaning is ‘whose father fought far away’). 112 See now Fantuzzi 2012, 267–286. 113 The Downward Journey or The Tyrant 8–13, 25. 114 The significant character of the name is underlined by the presence of two more ‘speaking’ names for the other main characters, the two who died on the same day as Megapenthes: the Cynic philosopher Kyniskos (the relevance of the name is obvious) and Mikyllos, a penniless cobbler (hypocoristic form of µικός ‘small’, a name that expresses the social insignificance of the character; notably this is a dialogue that comments on social contrast). 115 Bechtel 1917, 57.

‘Negative’ Emotions and Greek Names

183

though there are no relevant Homeric personal names. The group includes Παυσανίας, a common name, with a sense relevant to ‘ending distress’, and Ἀνάνιος/Ἀνανίας. The latter is only attested for an early iambographer from Ionia (sixth century BCE?), who was important enough to be remembered in the Byzantine period as one of the three ‘most distinguished iambic writers’, besides Archilochos and Hipponax.116 We may wonder whether his name too, like Archilochos’,117 sounded appropriate: it must mean ‘free from distress’ (a synonym of Alypos) – a gift perhaps of iambic poetry to the poet and his audience.118 There is also Λυρανίας (fourteen Hellenistic attestations, most of which come from Oropos on Euboia and from Aegean islands), which has the sound of an ‘irrational compound’ (its first component is λύρα, ‘lyre’).119 The simple name Ἀνίας has some scattered Hellenistic, Ιmperial, and late attestations. A poetic word for grief is ἄχος. There is no clear or definite connection between it and historical names starting with Ἀχ- (the commonest are Ἀχιλλεύς and Ἀχαιός), although a strong case has been made for a poetic exploitation of an Achilles-achos connection in the Iliad.120 However, as is often the case with heroic names, the use of Achilles’ name in the historical period is not necessarily linked with etymology, but rather implies desired comparison with the hero in question.121 Further terms for ‘sadness’, ‘grief’, are not absent (ταλαιπωρία, ἄλγος, ἄχθος, ὀδύνη and the poetic words δύη, πῆµα, οἰζύς), but they do not seem to have produced personal names.122 4 CONCLUSIONS All negative emotions that provide the basis of this study have produced personal names. The use of these names was shown to adhere to the usual criteria of namegiving (repetition of components in names that run in families, apotropaic function), while the exploitation of such names in literature follows known practices that accompany ‘speaking’ names (word-puns, etc.). As for documentary names, 116 Tzetzes, On Lykophron 2.18, ed. Scheer. 117 Leader of a lochos (company of soldiers) – a frequently encountered view. 118 The Biblical name Ananias may suggest a Semitic connection, but the name is of uncertain etymology. Even if originally non-Greek, its sound allowed naturalisation in the Greek language; cf. above, note 41. A similar name, also from the Greek periphery, is Anianos (Imperial Bithynia: LGPN Va; Egypt: Foraboschi 1967 and Wörterlisten). 119 Cf. the rare name Lyrippos (Eretria, third century BCE, LGPN I). 120 See Nagy 1979. 121 Cf. Masson 2001, 227: ‘un Ἀχιλλεύς devait être un nouvel Achille’. This is what Zgusta 1965, 92f., considered a ‘secondary name.’ 122 The following foreign names may possess a phonetic similarity to algos but are clearly irrelevant: Ἄλγαλσος (Livy 44.30.13 [Lat. Algalsus]), Ἀλγεµας? (Imperial Thrace, LGPN IV), Ἀλγοῦµις (Imperial Mysia, LGPN Va) and Αλγανις (Karia, Zgusta 1964, §44). It is unclear whether a Zeisodynon or Zeizodynon (Φλ. Ποντικὸς ὁ κα(ὶ) Ζειζ̣οδύνων, LGPN IV) bears a relevance to odyne.

184

Nikoletta Kanavou

while in several cases the choice of a name can be explained, in many cases we know too little about the lives of individuals to be able to determine the factors that led to the choice of particular names, or to detect whether such names, if given at birth, indeed exercised an influence in character formation. Similarly, ‘emotional’ names are hard to associate with social stratification, as information on social standing is usually too vague and the classificatory function of personal names too subtle. Many names are simply too scarcely attested for us to draw any safe conclusions about local or temporal relevance. There is no doubt that the Greeks were aware of the meaning of names. A case in point is the practice of giving semantically related names to related persons. Both the names of Oxycholios and his son Orgilios (p. 173) allude to anger. Aphobetos’ name signified lack of fear just as did that of his son Atrometos. His second son, Aeschines, also had a name with emotional overtones. Family histories are important for the understanding of name-giving practices but they are often elusive. Despite these limitations, a number of observations emerge. The most important conclusion is that words that express negative emotions usually do not produce names that express negative emotions but, instead, names that reflect the wish that the bearer of the name will be free of feeling a negative emotion. As already noted (p. 171), a large proportion of names are formed with the privative alpha: Ἀλύπαντος, Ἀλύπατος, Ἀλύπη, Ἀλύπης, Ἀλύπητος, Ἀλυπία, Ἀλυπιανή, Ἀλυπιανός, Ἀλυπιάς, Ἀλύπιος, Ἀλύπις, Ἀλυπίς, Ἀλυπις, Ἀλυπίων, Ἀλυπώ (free from grief), Ἀφόβητος, Ἀφοβητίδας, Ἀφοβία, Ἀφόβιος, Ἄφοβος, Ἀτρόµητος, Ἀτρέσιος, Ἀτρεστίδης/Ἀτρεστίδας, Ἀτρεµαίων, Ἄταρβος, Ἀταρβίων, and Ἀταρβίδης (free of fear and terror). The privative alpha cancels out the negative sense and expresses a positive expectation. Apart from the privative alpha, we also encounter uses of emotional terms in compounds that produce a positive sense (for instance, Pausanias). Also a compound name may express the wish that an enemy or an opponent shall feel the negative emotion (for instance Ἐπίφοβος, ‘the one who arouses fear in others’). One may, therefore, observe, that names deriving from negative emotions almost always have a positive sense, either as apotropaic names (protecting the bearer from the negative emotion) or underscoring the fact that a person was free of the negative emotion or wishing that a person should arouse the negative emotion in his opponents. This is in keeping with the general tendency to avoid negative associations in personal names. Recent studies in ancient Greek emotions highlight the fact that the Greeks were acutely aware of the rivalry and the power game inherent in many aspects of their social life.123 Personal names were intended to raise an individual’s status in the eyes of others, not to create room for insult or injury – the choice of nicknames on the other hand was intended to inflict injury when not meant as a joke. It is also no coincidence that names from emotions such as hatred and enmity are greatly outnumbered by names signifying the 123 On competitive emotions in social and philosophical contexts, see the collection of essays edited by Konstan and Rutter (eds.) 2003.

‘Negative’ Emotions and Greek Names

185

opposite – especially love and friendship; one only needs to think of the numerous -φιλο- compounds.124 This seems to be the case also with psychologically negative emotions. For instance, Aristotle’s opposite of fear, θαρραλέον (θάρρος), seems at a quick glance to have produced more personal names (more than eighty in total) than phobos. A second observation is that morally negative emotions (hate, envy) are less frequently represented in the formation of names than psychologically negative emotions (grief). Hate is almost non-existent, and nearly all names composed with φθόνος are formed with the privative alpha. Also psychologically negative emotions are represented in our discussion of onomastics by a greater variety of terms (that in their turn produce names) than morally negative emotions. For example grief is represented by three different terms (potentially more, if we add ἄχος), and so are anger and fear.125 Onomastics thus confirm the low esteem in which morally negative emotions were held, as opposed to psychologically negative ones, which were apparently better received. Names further enhance our awareness of the subtlety and versatility of emotions, as in the case of grief in its various contexts (literary, philosophical, and historical). The behaviour of emotion-related vocabulary presents an obstacle to quantification: numbers of suffixed and compound names from emotional terms are easy to provide, but with some of the terms it is not easy to discern whether derivations have ‘emotional’ meanings or not. For instance, it cannot be assumed either that all a-phthonos names are relevant to ‘envy’. Nevertheless, we may identify some cases of chronological concentration, most strikingly of names associated with the absence of lype: of the 123 occurrences of names starting with Alyp- held in the LGPN database, only seven (that is, around 5%) are dated earlier than the third century BCE. As noted earlier, the dramatic increase in popularity of these names in the Hellenistic and Imperial periods presumably reflects the influence of the Stoic apatheia.126 An interesting phenomenon is a semantic shift in names: in some cases we suspect that meanings of names gradually moved towards the moral side; the name of the Jew Oxycholios (an original nickname?) mentioned earlier (pp. 174f.) perhaps alluded to a vice and not to an emotion. Only a few female names can be found. They are mostly feminised versions of masculine forms. An exception is Orge: the name coincides with a feminine noun that constitutes an emotional term (‘anger’). Naming women after abstract nouns was a common practice. These nouns usually reflect roles and character

124 For such names, see Solin 2001. On attributes of praise deriving composed with φιλο-, see the remarks of Onno van Nijf in this volume (pp. 358–361). 125 The existence of numerous terms also suggests the importance of particular emotions in Greek literature and culture (according to the ethnolinguistic premise that language reflects the significance of aspects of a people’s cultural reality). 126 Cf. the observations of Chaniotis 2012, 121f., on the chronological distribution of names deriving from ἐλπίς (‘hope’).

186

Nikoletta Kanavou

traits expected in women.127 Like some masculine names, feminine names from unpleasant emotions may originate in nicknames. In addition to female names, those of slaves are an interesting separate phenomenon. It must be assumed that, like most slave names, these with an emotionrelated sense do not inform us about generation and family histories, as they were given by their masters. In addition to this, it is worth noting that although a number of names that connote negative sides of character and temperament (e.g., Βίαιος, Ἰταµή, Τρόµης) appear among the Athenian slave names studied by Charilaos Fragiadakis,128 we do not find any names deriving from negative emotions. Thymos and Orge are identical with words denoting negative emotions, probably deriving from personifications of emotions.129 Some of our emotional terms also designate divinities: personifications and deifications of emotions, such as Phobos and Phthonos.130 Personal names from these terms were probably formed from the relevant divine names, in which case they were thought to place children under the protection of the respective deities. The function of such names emerges as a point of intersection between the social and the religious. The presence of emotional terms in names from areas on the fringes of the Greek world deserves to be explored further as cultural indicator. Names symbolically express an individual’s cultural self and contribute to personal identities in a destabilising environment, such as was the world of Asia Minor. Greek names, used extensively from the Hellenistic period by members of the indigenous populations, clearly provide a vehicle for crossing boundaries between categories of geographical origin, kinship etc. Despite the difficulties concerning the interpretation of particular names, this study confirms the semantic field of emotions as a source of ancient Greek personal names and shows the potential of personal names to contribute to our understanding of the nature and function of Greek emotions. BIBLIOGRAPHY Alford, R. D. (1988) Naming and Identity: a Cross-Cultural Study of Personal Naming Practices, New Haven. Bechtel, F. (1898) Griechische Personennamen aus Spitznamen, Berlin. ––– (1902) Die attischen Frauennamen nach ihrem Systeme dargestellt, Göttingen. ––– (1917) Die historischen Personennamen des Griechischen bis zur Kaiserzeit, Halle. 127 See Golden 1986, 250–252, who sees such names as a form of ‘socialisation of girls’, in other words as a way for society to teach them their position and duties, as well as a means of depersonalising and objectifying women. 128 Fragiadakis 1988, 210. 129 The earlier mentioned Cholos is another such case. On personifications, see also above, note 50. 130 On theophoric names, see Parker 2000. Personal names such as Phthonos etc. should fall under the heading ‘Lesser gods/abstractions’ (ibid. 56, 58), though Parker’s list (for Attica) does not include the deities involved in the present study.

‘Negative’ Emotions and Greek Names

187

Ben-Ze’ev, A. (2000) The Subtlety of Emotions, Cambridge, Mass. Bernert, E. (1941a) Phobos, in RE XXXIX, 309–318. ––– (1941b) Phthonos, in RE XXXIX, 961–964. Buck, C. D. (1955) The Greek Dialects: Grammar, Selected Inscriptions, Glossary, Chicago. Chaniotis, A. (2002) The Jews of Aphrodisias: New Evidence and Old Problems, SCI 21, 209– 242. ––– (2012) Moving Stones: The Study of Emotions in Greek Inscriptions, in A. Chaniotis (ed.), Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, Stuttgart, 91–129. Chantraine, P. (1999) Dictionnaire étymologique de la langue grecque, Paris (new ed., with supplement). Colvin, S. (2004) Names in Hellenistic and Roman Lycia, in S. Colvin (ed.), The Greco-Roman East: Politics, Culture, and Society, Oxford, 44–84. Dasen, V. (2011) Childbirth and Infancy in Greek and Roman Antiquity, in B. Rawson (ed.), A Companion to Families in the Greek and Roman Worlds, Oxford, 291–314. DGE I²: F. R. Adrados et al. (eds.), Diccionario Griego-Español, vol. 1, Madrid 2008 (second edition). Fantuzzi, M. (2012) Achilles in Love. Intertextual Studies, Oxford. Fick, A. and F. Bechtel (1894) Die griechischen Personennamen nach ihrer Bildung erklärt und systematisch geordnet, Göttingen. Fitzgerald, J. T. (ed.) (2008) Passions and Moral Progress in Greco-Roman Thought, London/New York. Foraboschi, D. (1967) Onomasticon alterum papyrologicum: supplemento al Namenbuch di F. Preisigke, Milan. Fortenbaugh, W. W. (2008) Aristotle and Theophrastus on the Emotions, in Fitzgerald (ed.) 2008, 29–47. Fragiadakis, C. (1988) Die attischen Sklavennamen von der spätarchaischen Epoche bis in die römische Kaiserzeit: eine historische und soziologische Untersuchung, Athens. Freese, J. H. (1926) Aristotle: The Art of Rhetoric, London/Cambridge, Mass. (Loeb Classical Library). Ghiron-Bistagne, P. (1988) Le nom et le surnom dans l’onomastique grecque: étude de literature et d’épigraphie, in S. Gély (ed.), Sens et pouvoirs de la nomination dans les cultures hellénique et romaine, Montpellier, 5–19. Gisler, J.-R. (1997) Phthonos, in LIMC VIII.1, 992–996. Golden, M. (1990) Children and Childhood in Classical Athens, Baltimore/London. ––– (1986) Names and Naming at Athens: Three Studies, Échos du Monde Classique 30, 245– 269. Grimaldi, W. M. A. (1988) Aristotle, Rhetoric II: a Commentary, New York. Harris, W. V. (2001) Restraining Rage: the Ideology of Anger Control in Classical Antiquity, Cambridge, Mass./London. Hobson, D. (1989) Naming Practices in Roman Egypt, Bulletin of the American Society of Papyrologists 26, 157–174. Hornum, M. B. (1993) Nemesis, the Roman State, and the Games, Leiden/New York/Cologne. Judeich, W. (1898) Inschriften, in C. Humann, C. Cichorius, W. Judeich, and F. Winter, Altertümer von Hierapolis, Berlin, IV, 67–202. Kamptz, H. von (1982) [1958] Homerische Personennamen: sprachwissenschaftliche und historische Klassifikation, Göttingen. Kanavou, N. (2011) Aristophanes’ Comedy of Names: a Study of Speaking Names in Aristophanes, Berlin/New York. Kirk, G. S. (1985) The Iliad: a Commentary. Volume 1. Books 1–4, Cambridge. Kittel, G., G. Friedrich, and G. W. Bromiley (eds.) (1985) Theological Dictionary of the New Testament. Abridged in One Volume, Michigan.

188

Nikoletta Kanavou

Knuuttila, S. (2004) Emotions in Ancient and Medieval Philosophy, Oxford/New York. Konstan, D. (2001) Pity Transformed, London. ––– (2006) The Emotions of the Ancient Greeks, Toronto. Konstan, D. and N. K. Rutter (eds.) (2003) Envy, Spite, and Jealousy: the Rivalrous Emotions in Ancient Greece, Edinburgh. Kristjánsson, K. (2007) Aristotle, Emotions, and Education, Aldershot. Lichocka, B. (2004) Némésis en Egypte romaine, Mainz. LIMC: Lexicon Iconographicum Mythologiae Classicae, Zurich1981–. Lindauer, M. S. (2009) Psyche and the Literary Muses: the Contribution of Literary Content to Scientific Psychology, Amsterdam/Philadelphia. LSJ: H. G. Liddell, R. Scott, and H. S. Jones (eds.), A Greek-English, Oxford 1940 Lexicon (ninth edition); Supplements, Oxford 1968 and 1996. Lyons, W. (1980) Emotion, Cambridge. Macan, R. W. (1908) Herodotus: the Seventh, Eighth, and Ninth Books, Oxford. Mactoux, M.-M. (1993) Phobos à Sparte, Revue de l’Histoire de Religion 210/3, 259–304. Masson, O. (1990) Onomastica Graeca Selecta I-II, edited by C. Dobias and L. Dubois, Nanterre. ––– (2001) Onomastica Graeca Selecta III, edited by C. Dobias and L. Dubois, Geneva. Morpurgo-Davies, A. (2000) Personal Names and Linguistic Continuity, in S. Hornblower and E. Matthews (eds.), Greek Personal Names: Their Value as Evidence, Oxford, 15–39. Munteanu, D. L. (2012) Tragic Pathos: Pity and Fear in Greek Philosophy and Tragedy, Cambridge. Mussies, G. (1981) Catalogues of Sins and Virtues Personified (NHC II,5), in R. van den Broek and M. J. Vermaseren (eds.), Studies in Gnosticism and Hellenistic Religions, Leiden, 315– 335. Nagy, G. (1979) The Best of the Achaeans, Baltimore/London. Nussbaum, M. C. (2001) Upheavals of Thought: the Intelligence of Emotions, Cambridge. Parker, R. (2000) Theophoric Names and the History of Greek Religion, in S. Hornblower and E. Matthews (eds.), Greek Personal Names: their Value as Evidence, Oxford, 53–79. Picard, O. (1982) Monnayage thasien du Ve siècle av. Jésus-Christ, Comptes Rendus de l’Académie des Inscriptions et Belles Lettres, 412–424. PLRE: A. Jones, J. Martindale, and J. Morris, Prosopography of the Later Roman Empire, Cambridge 1971–1992. Preisigke, F. (1922) Namenbuch, Heidelberg. Rakoczy, T. (1996) Böser Blick, Macht des Auges und Neid der Götter: eine Untersuchung zur Kraft des Blickes in der griechischen Literatur, Tübingen. Reynolds, J. and R. Tannenbaum (1987) Jews and Godfearers at Aphrodisias, Cambridge. Robert, L. (1963) Noms indigènes dans l’Asie-Mineure greco-romaine, Paris. Solin, H. (1990) Namenpaare: eine Studie zur römischen Namengebung, Helsinki. ––– (1996) Die stadtrömischen Sklavennamen: ein Namenbuch. II.Teil: griechische Namen, Stuttgart. ––– (2001) Zur Geschichte der Namensippe φίλος in der antiken Anthroponymie, in M. Peachin (ed.), Aspects of Friendship in the Graeco-Roman World, Ann Arbor, 51–62. Sorabji, R. (2000) Emotion and Peace of Mind: from Stoic Agitation to Christian Temptation, Oxford. Stafford, E. (2001) Worshipping Virtues: Personification and the Divine in Ancient Greece, Swansea. Striker, G. (1996) Emotions in Context: Aristotle’s Treatment of the Passions in the Rhetoric and his Moral Psychology, in A. O. Rorty (ed.), Essays in Aristotle’s Rhetoric, Berkeley, 286– 302. Tappolet, C. et al. (eds.) (2011) Les ombres de l’âme: penser les émotions négatives, Geneva. Tarnopolsky, C. H. (2010) Prudes, Perverts, and Tyrants: Plato’s Gorgias and the Politics of Shame, Princeton/Oxford.

‘Negative’ Emotions and Greek Names

189

Thompson, A. (2007) Ancient Greek Personal Names, in A.-F. Christidis (ed.), A History of Ancient Greek: from the Beginnings to Late Antiquity, Cambridge, 677–692. Threatte, L. (1980) The Grammar of Attic Inscriptions. Volume 1. Phonology, Berlin. TrGF: B. Snell, R. Kannicht, and S. Radt, Tragicorum Graecorum Fragmenta, Göttingen 1971– 2004. Voigt, E.-M. (1971) Sappho et Alceus. Fragmenta, Amsterdam. Wörterlisten: http://www.zaw. uni-heidelberg.de/hps/pap/WL/WL.pdf. Zaborowski, R. (2012) Some Remarks on Plato on Emotions, Mirabilia 15/2, 141–170. Zgusta, L. (1964) Kleinasiatische Personennamen, Prague. ––– (1965) Some Principles of Work in the Field of the Indigenous Anthroponomy of Asia Minor, AION(Ling.) 6, 89–99.

IS PISTIS/FIDES EXPERIENCED AS AN EMOTION IN THE LATE ROMAN REPUBLIC, EARLY PRINCIPATE, AND EARLY CHURCH? Teresa Morgan 1 INTRODUCTION Πίστις in Greek, fides in Latin, and their cognates have an elastic range of meanings, centring on ‘trust’ but ranging through ‘trustworthiness’, ‘faithfulness’, ‘good faith’ and ‘honesty’ to more technical meanings such as ‘credit’, ‘pledge’, ‘security’, or ‘protection’ in law; ‘argument’ and ‘proof’ in rhetoric and philosophy; ‘credence’, ‘belief’, and what Christians, in particular, come to translate as ‘faith’. The range of contexts – social, political, legal, commercial, intellectual, and religious – in which the pistis/fides lexicon is deployed is well understood and explored by students of Greek and Roman culture.1 Perhaps surprisingly, however, little attention has been given to the cognitiveaffective nature of pistis/fides, or where it is located in those who experience and express it. Is trust or belief, faithfulness, or good faith a process of thought or an emotion, or both, or does its location depend on circumstances?2 This question is of interest to students of the Graeco-Roman world not least because pistis/fides plays such an important role in Greek and Roman society and culture. Few qualities are more closely involved in the creation, maintenance, and articulation of relationships, individual and corporate, at every level. Exploring how pistis/fides is understood may therefore add to our understanding both of Greek and Roman mentalité and of the way in which ancient societies are understood to function.3 In this essay I shall have something to say about cognition, but I focus on the aspect

1

2 3

E.g. Fraenkel 1916; Heinze 1929; Calderone 1968; Ramelli 2002; Deissmann-Merten 1965; Gruen 1984; Hellegouarc’h 1972, 23–61; Freyburger 2002 (on politics); Holder 1999 (on the military); Pringsheim 1931; Carcaterra 1984; Nörr 1991 (on law); Nock 1925; Dumézil 1969; Piccaluga 1981; Freyburger 1986; King 2003; Scheid 2005 (on religion); on philosophy see below. Trust and religious belief are also extensively discussed by sociologists, philosophers, students of religion, and, increasingly, psychologists. Pistis/fides is also regularly understood as an action and/or a relationship, meanings which are well recognised and which I do not pursue here. The idea that emotions help to create, maintain, and articulate societies is not, of course, new: see notably, for this period, Kaster 2005.

192

Teresa Morgan

of pistis/fides that has received least attention: the extent to which it is experienced as an emotion.4 The emotional aspect of pistis/fides, if any, in the world of the Late Roman Republic and Early Principate, is interesting not only in itself but also because this period sees the rise of Christianity, to which pistis is already central in the writings of the New Testament. Modern Christians tend to assume that faith is intrinsically emotional: it is something one feels, which involves the heart as well as the mind and action.5 This assumption is readily retrojected onto early Christianity, but if pistis/fides is not treated as an emotion in Graeco-Roman culture more widely, this retrojection needs to be carefully scrutinised and may prove hard to defend. In contrast, classicists seem generally to assume that pistis/fides is cognitive, relational, and active but not strongly emotional.6 They may be right, but the assumption bears investigating. We shall begin, therefore, by exploring the possible emotional content of pistis/fides in Greek and Latin writings of the period in general, and then consider the implications of our findings for the study of the early Church.7 To assess the emotional content of a term such as pistis or fides, as distinct from cognitive or other meanings, in any individual text, is often far from easy. Take, for example, a letter from Cicero to his freedman Tiro in which Cicero quotes Tiro as having written that he is ‘faithfully attending to [or ‘observing’]’ Cicero’s health (valetudini fideliter inserviendo).8 What does fideliter mean here? Does it imply loyalty of heart, mind, or action on Tiro’s part? Is he taking a friendly interest in Cicero’s health, procuring some doctor or medicine for him perhaps, or expressing his devotion to his former master? There seems to be no way of telling. Nor is this difficulty confined to modern readers. Cicero goes on to criticise Tiro’s use of fideliter because he does not understand it himself. Where does ‘faithfully’ come from here? The home territory of the word is in matters of duty, but it makes many excursions into other fields.

Cicero does not suggest that Tiro’s fides might have an emotional aspect, but at the end of the letter he describes himself as reciprocally anxious for Tiro’s health 4

5

6

7

8

For convenience, I use the nouns to stand for their cognates too. Confidence (θάρσος, fiducia), however, is both clearly regarded as an emotion by ancient writers and well studied by modern scholars, so, though it is related to pistis/fides, I omit it here. Though this assumption has not been much taken up by scholars of religious psychology (e.g., Corrigan 2008; Riis and Woodhead 2010) nor of the New Testament (e.g., Matthews 1980; Roberts 1992; Elliott 2005). Clearest in studies of rhetorical persuasion (Kennedy 1963, 1969, and 1972; Winterbottom 1980; Russell 1983; Worthington 1994) and religious belief (e.g., Nock 1925; Yunis 1988; Scheid 2003, 18 and 20; Versnel 2011, 544f.). This also has implications for the study of Greek and Roman religions, which I shall not consider here because before discussing whether pistis/fides has an emotional aspect in GraecoRoman religions, one must establish whether it plays a significant role in them at all, a complex project in its own right. Ad familiares 16.17.1.

Was pistis/fides experienced as an Emotion?

193

(sollicitat), using a word with such clear emotional content that we might feel justified in inferring that Tiro’s fides also has an emotional dimension. We cannot, however, be sure. This letter is unusual in that Cicero admits that he does not understand Tiro’s use of fideliter, but the opacity of the word to us is discouragingly typical. Close readings of passages in which the pistis/fides lexicon occurs, with a view to teasing out its emotional content, are all too apt to produce ambiguous results. In what follows, therefore, I take a different approach, designed to establish whether pistis/fides is described in the same terms as recognised emotions and/or commonly associated with them. We begin by investigating where in persons ancient authors locate pistis/fides and whether it is described as operating in the same way as (other) emotions. We then turn to passages where pistis/fides is linked to qualities which clearly are regarded as emotions, before examining its connections with the language of cognition and of virtue, in case pistis/fides is better categorised as a virtue or an operation of the intellect than as an emotion. I shall focus on sources from the first century BCE to the end of the second century CE. This period offers a particularly large, rich, and diverse range of material on which to base our analysis; it also forms the immediate cultural context of early Christianity. Within this period, I shall concentrate on literary texts, as attesting the widest surviving range of understandings and uses of pistis/fides language. These texts are composed by many different authors and belong to many genres, and in putting them side by side I am inevitably glossing over their individuality to some degree. The justification for this is twofold: I have not, on the whole, been able to detect significant differences in the cognitive-affective treatment of pistis/fides by different authors or genres in this period, and in most cases, the affectivity of pistis/fides is not the main focus of a particular text’s or author’s interest, so it is less likely to be being treated in a way distinctive to that author or text. I take pistis and fides as sufficiently close in meaning to be discussed together. There are some differences between them (more shades of legal meaning in Latin texts, for instance, and of philosophical meaning in Greek ones), but they are convergent enough in most contexts to be treated as equivalents. In defining ‘emotion’, I follow lines laid down in a number of recent studies of ancient emotions.9 I take an emotion to be something which is often identified in Greek as to pathos, in Latin as adfectus, in English as ‘affection’ or ‘passion’ as well as ‘emotion’. It involves feeling, but also cognition and action, and we should not assume that it is identified or experienced in the same way in every culture.

9

E.g., Annas 1992, 103–120; Kaster 2005, 1–8; Graver 2007, 1–13; Konstan 2007, 3–40; Fitzgerald 2008, 1–5.

194

Teresa Morgan

2 PISTIS/FIDES AS AN EMOTION One thing about pistis/fides is clear: it does not feature in any of the famous lists or accounts of the emotions in Greek or Latin literature. Plato’s Philebus and Republic, Aristotle’s Rhetoric and Nicomachean Ethics, Cicero’s On Rhetoric and Tusculan Disputations, Quintilian’s Education of an Orator, Plutarch’s Progress in Virtue, and the surviving fragments of the Stoics, Epicureans, Peripatetics, Academics, and Cynics, are unanimous in ignoring pistis/fides in their discussions of emotion. Nor has pistis/fides attracted the attention of modern scholars of emotions in Graeco-Roman history or philosophy, nor (in its primary meaning of ‘trust’), as far as I can discover, in any of the scientific disciplines in which emotions have been studied since the mid-nineteenth century. The message seems to be unambiguous: trust is probably not an emotion in scientific terms, and pistis/fides is not an emotion in Graeco-Roman thinking. It is therefore unexpected and striking that when we look at passages from Greek and Latin literature in which pistis and fides are invoked, they are described in ways which are often indistinguishable from the ways in which emotions are described, they share with the emotions a range of psychosomatic locations, and they are regularly and closely associated with qualities which clearly are understood as emotions. This is true both in texts that are influenced by philosophy and those that are not, or not obviously. This is not the place to discuss philosophical accounts of the emotions in detail, but since many Greek and Roman authors of this period are influenced by philosophy to a greater or lesser degree, a brief summary of some key points may be helpful. Plato’s principal, brief discussion of emotions occurs in the Philebus, where he touches on the difference between various kinds of pleasure and pain.10 Some involve only the body (like the pain of an itch and the pleasure of scratching it); some involve both body and soul (psychē) (such as feeling hungry and anticipating being fed); some involve only the soul (such as anger, fear, love, and grief). All these can be called pathē, but what we call emotions and later authors study as emotions are the ones located in the soul.11 Plato also mentions in the Philebus the doctrine of the soul which he develops more fully in the Republic: the soul is tripartite, its rational (logistikon) part ideally ruling its ‘spirited’ (thumoeidēs) and ‘appetitive’ (epithumētikon) parts, in the last of which emotions are located, alongside other non-rational processes.12 Aristotle touches on emotions in a number of works, but principally in Rhetoric 2.1–2. Here, as elsewhere, Aristotle distinguishes between cognitive, physical, 10 11 12

Philebus 36c6–50a9. Fortenbaugh 2008, 30f. Res publica 434a–441e, 580d–583a. Among Platonists of the early Principate, Plutarch follows Plato, arguing that passions need to be domesticated by reason (Wright 2008). Philo, however, seems to follow the Stoics more closely, arguing that the religious sage overcomes all (other) emotions through the strength of his (rational) love of God (Winston 2008; cf. Dillon 1977, especially 151–153).

Was pistis/fides experienced as an Emotion?

195

behavioural, and ‘psychic’ elements of emotions, the last occurring in the heart or soul. Like Plato, he thinks that reason should rule the emotions but need not eradicate them.13 Epicureans and, above all, Stoics, develop the most systematic interest among ancient philosophers in emotions and where in a person they are located.14 The two schools share an account of the soul (psychē), which divides it into pneuma, ‘spirit’, and nous, ‘mind’, the latter rendered in Latin as animus or mens. The mind is the seat of all mental states, including the emotions, and is located in the chest or the heart.15 To complicate this picture slightly, animus in standard literary Latin can be translated as either ‘mind’ or ‘heart’. Catullus may be drawing on philosophical ideas current among Roman litterati, or he may be reflecting ordinary Latin usage when, addressing Cornelius in Poem 102, he locates his fides in his animus: If any secret was ever entrusted by a faithful friend, the trustworthiness of whose heart/mind was deeply known (cuius sit penitus nota fides animi), you will find that I too am consecrated by their rite, Cornelius …

Tibullus (1.6.75–76) locates his lover Delia’s fidelity in her mens: Don’t be chaste through savage fear, but through your faithful mind (mente fideli); may mutual love keep you safe when I am away.

Cicero, explaining in Pro Marcello (14) why he followed Pompey to Pharsalia, avers that he is always on the side of peace and maintaining civil society, but that sometimes a private obligation overrides political prudence: I followed an individual out of private, not public obligation, and the faithful memory of a grateful mind/heart (grati animi fidelis memoria) was so strong in me that, not out of any greed or even hope, but in full knowledge and understanding of what I was doing, I rushed towards a voluntary doom.

Seneca the Younger, speaking as a Stoic, also locates fides at times in the animus. In On Benefits (3.15.1f.) he wishes that commercial agreements did not need to be hedged about with legal documents, but could be left to ‘fides … and a mind/heart (animus) that cultivates justice’. Elsewhere, Seneca associates fides with the breast (pectus), location of the physical heart and seat of feelings, especially of love, across a wide range of literary works. In his letter of consolation to Helvia (19.1), he tells Helvia that her greatest source of consolation is her sister, who has fidelissimum tibi pectus (‘a heart most faithful to you’). In Letter 88.29, reflecting on various virtues, he asserts that ‘trust is the most holy good in the human heart’ 13

14

15

Grimaldi 1988, ad loc.; Fortenbaugh 2002, 9–12, 23–44; Fortenbaugh 2008, 31–38; cf. Verbeke 1990, 157, on Aristotle’s Nichomachean Ethics. Comparing Plato and Aristotle see Knuuttila 2004, 5–31. E.g., Wisse 1989; Donini 1995; Brennan 2005, 82–113; Graver 2007 (on Stoics); Graver 2002 (on Cicero); Asmis 1990; Cooper 1999; Sanders 2008 (on Epicureans); Hankinson 1971 (on Galen and the Stoics); Long 2002 (on Epictetus); cf. Dillon 1983 (on later Platonists). On Cynic and Neopythagorean ideas about passions see Aune 2008 and Thom 2008. Long and Sedley 1987, I, 70f., II, 319–322; Everson 2005; Long 2005.

196

Teresa Morgan

(fides sanctissimum humani pectoris bonum est). Fides, he continues in highly emotional terms, cries, ‘Burn me, cut me, kill me – I shall not betray [my trust?]…’. The Distichs of Cato, derived substantially from Greek gnomic sayings and aimed probably at a popular audience of the early Imperial period, locate fides in the heart proper, cor. When a man simulates friendship with words but is not a faithful friend in his heart (nec corde est fidus amicus), do the same to him: so art is cheated by art.

In Precepts of Statecraft (821a–b), Plutarch observes that: Nothing makes a man easy to handle and willingly gentle towards another man but trust (pistis) in his goodwill and belief (doxa) in his nobility and justice. Which is why Demosthenes is right when he proclaims that mistrust (apistia) is the best defence cities have against tyrants, because the part of the soul (psychē) with which we trust is the easiest to capture.

Pistis could be cognitive here and parallel to doxa, but when Plutarch talks about the part of the soul with which we trust being the easiest to capture, it is tempting to see him as talking as a Platonist, about the irrational, emotional part of the soul which needs to be ruled by reason. In all these passages, varied though they are in language and genre, pistis/fides is an interior process and the various parts of our interior that host it are all also seats of emotion. This does not prove that pistis/fides is an emotion, not least since the same parts (at least according to philosophers) host thought. It may, however, be significant that we cannot always, and perhaps even contemporary readers could not always, distinguish references to cognition and emotion in words like animus, mens, and psychē. Sometimes, in the thinking of Late Republican and Early Imperial authors, mental processes and emotions may simply not have been segregated. The context in which pistis/fides language occurs in most of these examples is also suggestive. In some passages it is linked with love or friendship, or with fear or grief – all qualities which are well attested as emotions in themselves, or qualities with significant emotional aspects.16 In others it is part of a flow of rhetoric which seems calculated to arouse emotion. This takes us to our next group of examples, in which pistis/fides language appears in close association with language that is clearly emotional. 3 PISTIS/FIDES AND LINKED EMOTIONS There is a strong connection in both Greek and Latin between pistis/fides and the language of love and friendship: philia and erōs in Greek, amicitia and amor in Latin. Hundreds of passages in Greek and Latin literature link friendship, love, and trust, in public and private life, family life, and erotic relationships. 16

I follow Konstan 1997 in finding emotional content in friendship.

Was pistis/fides experienced as an Emotion?

197

Trust is regularly described as a characteristic of established relationships of all kinds. In particular, Greek and Latin literature abound with references to trust, good faith, and trustworthiness between friends – covering all the many meanings of friendship from political alliance to patronage, from military comradeship to relations between equals in private life.17 Πιστός, fidelis, πιστότατος, and fidelissimus are such common adjectives qualifying ‘friend’ as to be clichéd.18 Men planning political coups share their plans with those of mutual pistis/fides and philia/amicitia.19 Fides, according to Cicero, is the basis of that stable constancy which we look for in friendship.20 In worrying times in 59 BCE, he tells Atticus in a letter how much he relies on Atticus’ love and faithfulness (credibile non est … quantum in amore et fide ponam).21 In Plutarch’s Life of Alexander (19.2.4), when the suspicion is raised that Alexander’s doctor Philip is trying to poison him, Philip’s pistis in his philia with Alexander and Alexander’s pistis towards Philip ensure that Alexander lets Philip cure him. Mithridates, by contrast, in Appian’s Roman History (12.111) complains of the ‘deadly poison’ which is to be found in every king’s house: the untrustworthiness (apistia) of those who should be most attached to him: his children, friends, and army. Sometimes trust is said to be the basis of love or friendship, especially in relations between states, where pistis or fides in the sense of an agreement to form an alliance is often said to lead to friendly relations.22 In other contexts, love or friendship leads to trust. According to Plutarch (Dialogue on Love 769a), physical union is the beginning of philia and pistis between husbands and wives. King David’s friend, Hunshai, making overtures to David’s son Absalom in Josephus’ Jewish Antiquities (7.211f.), promises him that if Absalom regards him as his friend, he will show the same pistis and goodwill towards him.

17 18

19 20 21 22

Though pistis/fides between clients and patrons is rare (despite Badian 1958, 2–10); two examples are Juvenal 9.82 and Valerius Maximus 9.11.6. E.g. (among many other examples) Dionysius of Halicarnassus, Antiquitates Romanae 4.85.1, 14.26.4, 16.46.3, 20.19.1; Appian, Gallica 6.74; Athenaeus 4.151d; Sallust, Bellum Catilinae 9.1; Cicero, De amicitia 54; Seneca, De clementia 1.9.11; De tranquillitate animi 4.3, 7.3; Josephus, Bellum Iudaicum 1.516; cf. Fronto, Ad Antonium Pium 8.1; Scriptores Historiae Augustae, Hadrian 11. Sallust, Bellum Catilinae 20.2f., 22.2; Plutarch, Pelopidas 8.2. De amicitia 65. Ad Atticum 2.23.3; cf. Seneca, Epistulae 3.2–3. Sallust, Bellum Iugurthinum 31.23–25; Caesar, Bellum Gallicum 1.3; Dionysius of Halicarnassus, Antiquitates Romanae 1.58.4, 14.29.5; Livy 1.38.2, 7.30.6, 24.19.10, 26.14.2, 29.3.3, 32.17.1, 40.49.1, 42.63.10; Silius Italicus 14.79–84; Appian, Bella civilia 1.38. Cf. Dio Chrysostom 73.4 (being regarded as trustworthy leads to an individual’s being entrusted with the rule of a city); Cicero, De officiis 2.21 and Epictetus 2.22.18–20 (in public life one can only fully trust the man who is trustworthy in himself); Dio Chrysostom 73.5 (loss of trust in an individual who has been trusted in public life brings heavy punishment, as people avenge their disappointment).

198

Teresa Morgan

According to Aelian, trust can even exist in relationships between people and animals, and between animals. In On Animals (2.8), he describes how the dolphins of Euboea co-operate with fishermen to catch fish. At the end of the day the fishermen give them their portion πιστῶς καὶ εὐγνωµόνως (‘with good faith and gratitude’). Crows, meanwhile (3.9), are most faithful (pistotatoi) to one another, and when they enter a partnership, love one another intensely (πάνυ σφοδρῶς ἀγαπῶσι). Trust is often characterised as particularly strong – ideally and, it must be said, not always with justification – between husbands and wives. 23 Explaining the proverb ‘Letters of Bellerophon’ (2.87), Zenobius tells us that when the wife of King Proteus of Tiryns tried and failed to seduce Bellerophon, she told Proteus that Bellerophon had tried to rape her. Proteus, who trusted (πιστεύσας) his wife, sent Bellerophon to his death. While waiting for Agamemnon to return from war, Seneca’s Clytemnestra says that once she might have guarded the marital bed coniugis … fide, ‘with a wife’s good faith’, but that the time is past for fides, along with mores, ius, decus, pietas … pudor.24 That she will be able to trick Agamemnon to going to his death is presumably because he still trusts her as his wife. Other family relationships are also among those which – again, ideally – are characterised by the greatest trust. Perhaps the most spectacular example is that of Veturia and Coriolanus in the account of Dionysius of Halicarnassus. Describing how Coriolanus marched on Rome with an army of Volscians and was stopped by his own mother (Roman Antiquities 8.48.1f.), Dionysius makes Veturia say that she puts her trust (πιστεύειν) in the law of nature according to which Coriolanus’ very body and soul are on loan from her, his mother. If he is going to attack his native city, he will have to kill his mother first – but she does not believe that he will, and she is right.25 In Seneca’s Phoenician Women (478–480), Polynices comments that, if even a brother is prepared to break faith (fides) with a brother, one cannot trust the good faith of a mother either.26 Often, of course, it is when expected trust fails that we hear about it. Plutarch’s Alcibiades uses the expected trust between a parent and child to emphasise, paradoxically, how little he trusts the Athenians to give him justice. Asked whether he does not trust (πιστεύεις) the Athenians to decide on his guilt or innocence after the profanation of the Mysteries in 415 BCE, he replies that he would not even trust his mother, lest in a thoughtless moment she cast the wrong ballot.27 For Dio Chrysostom (31.32f.), the essence of wickedness is to care only about whether one’s actions are profitable. Such a person, who will betray even his 23 24 25 26 27

Though Dio Chrysostom 74.9 on distrust claims that one cannot trust any woman (or slave) because they lack judgement. Seneca, Agamemnon 108–124. I take Veturia’s trust in their relationship to imply trust in Coriolanus himself, though Dionysius does not say this explicitly. Cf. Seneca, Medea 430–439, 1002–1004. Alcibiades 22.2; cf. Moralia 186e–f.

Was pistis/fides experienced as an Emotion?

199

friends, is ἄπιστος, not to be trusted. In a pessimistic oration On Distrust, Dio claims (74.1) that one cannot trust anyone because even friends harm each other all the time. He considers famous examples of fast friends: Orestes and Pylades, Theseus and Pirithous, and Achilles and Patroclus, and concludes that, if they could trust each other, they are the only people in history who could (74.28). Epictetus, in his Discourse 2.22 on friendship, is more optimistic: trust is part of good relations between friends, and also sons and fathers (22.18–24). Elsewhere, Epictetus observes (4.13.13) that we feel we can safely trust (πιστεύειν) a man with whom we have at least the beginning of a relationship because he has entrusted knowledge of his affairs to us. The relationship in which pistis/fides is often characterised as most problematic is, unsurprisingly, that of erotic love. On the one hand, authors and their characters are widely agreed that trust is, ideally, part of a loving relationship. On the other, Eros himself is famously inconstant and untrustworthy, and love poetry resounds to cries of apistos! infida! When a girl swears in love, says Tibullus (3.6.47–52), there is no fides in her oath. Jupiter laughs at lovers’ oaths and tells the winds to blow them away unfulfilled. Erotic love and the trust it engenders can even be used disingenuously for political purposes. According to Josephus, this was Cleopatra’s specialty.28 Throughout Graeco-Roman literature, tyranny is the opposite of a relationship of friendship and trust. It is a topos of Greek and Latin literature that tyrants neither love nor trust anyone, nor does anyone love or trust them.29 One of the indications that Numa is a good king, not a tyrant, in Plutarch’s Life is that he refuses to reign over people who do not trust him (7.4). The problems Mithridates has with trust stem in part from the fact that he regards himself as a good king while others see him as a tyrant. We could multiply examples of the relationship between pistis/fides, love and friendship almost indefinitely. It also occurs, however, in conjunction with a number of other emotions. In a complex and interesting passage in his Life of Cato the Younger, Plutarch tells us that: no virtue, by the fame and credit which it gives, creates more envy than justice, because both power and trust (πίστις) follow from it, chiefly among the common people. These do not merely honour the just, as they do the brave, nor admire them merely, as they do the wise, but they actually love the just, and put confidence and trust in them (θαρροῦσιν αὐτοῖς καὶ πιστεύουσιν). As for the brave and wise, however, they fear the one and distrust the other; 30 and besides, they think that these excel by a natural gift rather than their own volition ...

Here, pistis is associated with feelings of love, confidence and envy. Later in the same biography (56.2), the pistis of his soldiers creates feelings of shame and 28 29

30

Josephus, Antiquitates Iudaicae 15.99. E.g., Plutarch, Dion 9.3; Dio Chrysostom 9.15, 33.108; Philostratus, Vita Apollonii 1.38.1; Cicero, De amicitia 52; Nepos 8.6; Seneca, Agamemnon 284–287; cf. Seneca, Medea 221, 248. Plutarch, Cato the Younger 44.7f. (translated by B. Perrin, Loeb).

200

Teresa Morgan

compassion in Cato (αἰδούµενος καὶ οἰκτείρων). Pistis is several times associated in our sources with αἴδως or τιµή, shame or honour. In his essay On Isis and Osiris (359f–360b), Plutarch claims that human beings should not be thought of or treated as gods because this dissipates the timē and pistis towards ‘real’ gods implanted in nearly all human beings at birth, and leads ultimately to religious scepticism and atheism. For Epictetus, there is a strong correlation between pistis and aidōs:31 God has delivered your own self into your keeping, saying, ‘I had no-one pistoteron than you to keep this man for me unchanged from the character with which nature endowed him – reverent (αἰδήµονα), faithful (πιστόν), high-minded, undismayed, unimpassioned, unperturbed.’ After that, do you fail so to keep him?

We noted above the connection between pistis and having confidence (θαρρεῖν) in a leader. In On the Alexandrian War (17), about Caesar’s wars in Egypt, Caesar is described as having confidence in himself, believing (with justification) that he can attack the Pharos and the city of Alexandria at the same time.32 Pistis/fides is also often clearly connected, explicitly or implicitly, with hope. In Plutarch’s Life of Lysander (5.4), Lysander gives the people of the Ionian cities pistis in a future without the Delian League with a series of promises and practical benefits. In his Life of Romulus (7.4f.), Numitor inspires Remus with pistis and elpis by his gentle voice and kind expression. Dio Chrysostom, in his essay On Trust (73.4f.), notes that people have high hopes of those they trust and are regularly disappointed in those in whom they have trusted and hoped too much. In other contexts, pistis/fides finds itself uncomfortably caught between hope and fear. In the Tristia (4.3.11–20), Ovid tries to reassure himself of the loyalty of his wife by believing what he hopes is true and stifling his fears: Why do my hopes falter, blighted by anxious fear? Believe (crede) that which is true and what you wish, and stop being afraid where there is no danger. When fides does not waver, hold it in certain fides and tell yourself that she … loves you still.

In 296/5 BCE, during the Romans’ war with the Gauls, according to Cassius Dio (8.28), a number of portents were seen. The general Manius interpreted them to mean that the Romans would be victorious in war. The multitude, however, was unable either to trust him or not to trust him (πιστεύειν … ἀπιστεῖν). It did not want to see all the points he prophesied fulfilled, but it did want victory, so it found itself torn between fear and hope. Sometimes pistis/fides is described as the opposite of fear, or as driving out fear.33 For Seneca in On Benefits (7.26.4f.) it is a bad thing to be ruled by one’s passions. Fear, for instance, makes one unable to give fidele consilium, ‘faithful’ or ‘trustworthy’ advice. Occasionally, however, pistis/fides is said to go hand in 31 32 33

Dissertationes 2.8.23; cf. 2.4.1, 2.22.18–20, 3.17.1–3, 4.13.20. Cf. Caesar, Bellum Gallicum 7.50, Bellum civile 3.24, 3.49, 3.109 (of the confidence of Caesar and others). E.g., Ovid, Tristia 4.3.1120; [Seneca], Octavia 346–360; Apuleius, Metamorphoses 4.21.

Was pistis/fides experienced as an Emotion?

201

hand with fear to positive effect. According to ps.-Quintilian in a minor declaration (245.4), good faith (fides) cannot be maintained between people in legal contexts unless it is held in check by fear. For Plutarch, the final result of thinking that there are no gods is that we do not fear them; conversely, presumably, the corollary of believing or putting one’s trust in gods is that one does fear them (though not too much, as that would constitute superstition).34 Finally, in a few passages, pistis/fides is said to be the result of the arousal of emotion. Cicero, for instance, in On the Division of Oratory (53), tells his son that from time to time in a speech, and especially in the peroration, an orator may engage in amplificatio, an enlargement of something that has already been said, which is designed to create trust in the speaker by moving the minds/hearts of the listeners (est igitur amplificatio gravior quaedam affirmatio quae motu animorum conciliet in dicendo fidem). That which is aroused by emotion is not necessarily an emotion itself, but by analogy with the action of rational discourse, we may suspect that it has a strong emotional dimension. If rational speech appeals to reason and engenders rational thought, as rhetorical treatises regularly assume, then presumably emotional speech appeals to the emotions and creates an emotion, which in this case is identified as fides. 35 In this section, we have seen passages in both Greek and Latin, from a wide range of authors and genres, and from very diverse contexts within texts, attesting close relationships between pistis/fides and a number of emotions, or qualities that have strong emotional aspects. These relationships are configured in a number of ways, and in individual passages it is rarely beyond doubt how the language of pistis/fides should be interpreted. Cumulatively, however, the association of pistis/fides language with that of emotions is so persistent as to suggest strongly that if pistis/fides is not an emotion pure and simple, it has emotional aspects or can be treated as an emotion in some contexts.36

34

35 36

Plutarch, Moralia 165b (de superstitione). On appropriate fear of the gods, see e.g., Ovid, Metamorphoses 3.650–691, 611–724; cf. Plutarch, Moralia 424a. On De Superstitione, see further below, p. 200. Cf. Quintilian, Institutio Oratoria 6.1.9. Fear creates belief in portents among Romans under threat, according to Cassius Dio 14.7. In principle, an emotion might lead to a purely cognitive condition of pistis/fides or vice versa; moreover, since emotions have cognitive aspects (see below, note 37), in theory the cognitive aspect of an emotion could lead to a cognitive state of pistis/fides or vice versa. However, given that it is so often impossible for us, and may have been no easier for contemporary readers, to distinguish cognitive from emotional aspects of either pistis/fides or emotions in particular contexts, it is safer to assume that emotion is normally involved wherever pistis/fides and emotions interact.

202

Teresa Morgan

4 PISTIS/FIDES AND COGNITION When it is not being described in the same terms as emotions, or associated with emotions, pistis/fides is often described in cognitive terms. This does not rule out its being understood as an emotion: in recent years it has been increasingly recognised by scholars of the emotions that cognition is an intrinsic part of emotion.37 Whether or when emotion is part of cognition, however, is less explored and less clear, and it is to that question that we turn our attention in this section. The cognitive aspect of pistis/fides is clearest when what is under discussion is the foundation of a particular belief or belief in general, and it is common in sources of all kinds for agents to be described as believing or trusting in someone or something on the basis of some kind of evidence or for some reason.38 One frequently cited basis for pistis/fides is personal experience. Dionysius of Halicarnassus, for example, relating the noble behaviour of Lucius Junius Brutus, fears that it will appear incredible to Greeks, since it is natural for all men to judge others by their own experience and determine what is πιστόν or ἄπιστον with reference to themselves.39 The professional expertise of others is also widely acceptable as grounds for pistis/fides. So, for instance, the military strategist Onosander, in his introduction to The General (praef. 7), comments that everyone naturally gives credit (pistis) for truthfulness to those who appear to write with professional expertise.40 The evidence of one’s senses is regularly presented as trusted or trustworthy.41 In Philostratus’ The Hero (7.9), one character begins the story by saying that he is untrusting (ἄπιστος) of myths because he has never met anyone who saw or heard at first hand the things myths tell of. (He is later converted by the testimony of his interlocutor, who claims to have seen and spoken to the hero Protesilaus.)42 The travellers in Lucian’s A True Story (7), at one point on their journey, find a bronze tablet inscribed ‘Heracles and Dionysus came to this place.’ They defer belief, however, until they arrive at a river of wine, when they become much more inclined to trust the testimony of the tablet (πολὺ µᾶλλον πιστεύειν), seeing the evidence of Dionysus’ visit with their own eyes. 37 38 39 40

41

42

E.g., Sorabji 2000, 17–28; Nussbaum 2001, ch. 1; Konstan 2006, 20–23, tracing the development of scholarly thinking about emotion and cognition. Pistis/fides language here is close in meaning to the language of both persuasion and thought. Antiquitates Romanae 5.8.1. Cf. Lucian, Calumniae non temere credendum 31; Philostratus, Heroicus 7.9–12 (on being apistos towards what one has not experienced oneself). Cf. Lucretius 5.104–106; Dionysius of Halicarnassus, Antiquitates Romanae 1.77.1–78.1; Strabo 1.1.21; Pliny, Naturalis historia 29.29; cf. Aelianus, De natura animalium 13.21 (of the authoritative knowledge of a god). Lucretius 4.462–510; Strabo 2.1.11; Cicero, Academicae quaestiones 1.41 (not all sense data are worthy of fides). Occasionally, people willfully disbelieve the evidence of the senses, e.g., Josephus, Bellum Iudaicum 6.288. Persuasion is related to belief both etymologically in Greek and psychologically; rhetoricians’ recognition that persuasion can be both cognitive and emotional is well studied by scholars of rhetoric.

Was pistis/fides experienced as an Emotion?

203

Lucian often has fun with our tendency to believe in the evidence of our senses. In The Lover of Lies or the Doubter, Tychiades reports to his friend Philocles that he has just come from a gathering at which people were telling each other tall stories about gods, spirits, miracles, and journeys to Hades. Their authority for these stories was either alleged personal experience or common knowledge. Tychiades, however, refused to believe any them on the grounds that he had not had comparable experiences himself. Experience therefore failed to give the friends any common basis for belief or debate, so they were stuck at loggerheads (leaving the reader to ponder how far he or she would trust the evidence of the senses, and why). Pistis language is used throughout the story of the characters’ beliefs and mutual (mis)trust. At the end of the dialogue (40), Tychiades and Philocles express the fear that pistis actually bypasses the senses completely and is simply contagious, or that hearing stories may make us believe them in spite of ourselves, in the way that drinking wine makes us drunk. They do not describe pistis as an emotion, but the idea that it might bypass reason to affect us in spite of ourselves is reminiscent of the way orators describe the impact of the emotions on those who listen to speeches, so Lucian perhaps imagines cognitive belief as having some emotional content here.43 In contexts that take the supernatural more seriously than does The Lover of Lies, dreams and visions of the gods, supernatural voices, and signs can all be regarded as comparable with the evidence of the waking senses, and as equally, rationally, credible.44 At the same time, they are clearly not unproblematic for many authors. Plutarch describes how after the death of Romulus at the hands of his senators, Iulius Proculus claims that has seen him alive, and that he has become a god.45 The Roman people believe him (πιστεύειν) and do not, as they might otherwise have done, attack the senate-house. The way Plutarch frames Proculus’ announcement, as a claim rather than a fact, suggests that he does not give Proculus’ story much credence himself.46 Cicero, in On Divination (2.122), dismisses visions of gods in dreams as always untrustworthy: If one cannot have fides in the visions of the insane because they are false, I do not understand why we believe (credatur) the visions people have when they are asleep, which are much more confused.

A cluster of stories told by Valerius Maximus and Frontinus about generals on the eve of battle offers one explanation of why some people seem to find it relatively easy to believe in supernatural phenomena, at least in some situations.47 The Spartan king Archidamus, according to Frontinus, told his troops before a major battle 43 44 45 46 47

E.g., Cicero, De oratore 2.214. Valerius Maximus 1.8.3; Plutarch, Marius 17.4–6; Moralia 589d; Apuleius, Metamorphoses 11.27; Cassius Dio 14.7. Moralia 313d; cf. Romulus 27.8, 28.1–6. Cf. Lucian, Dialogi meretricii 287; Dionysius of Halicarnassus 5.8.1; Seneca, Epistulae 18.12–31. Valerius Maximus 1.2.

204

Teresa Morgan

that Castor and Pollux were watching over them. He was believed; the troops were inspired; the battle was won. The Theban general Epaminondas improved the confidence (fiducia) of his troops in general by fostering their fides towards the gods.48 Perhaps cognitive belief is sometimes easier to achieve when there is some emotional reason for holding it or an emotional advantage to doing so.49 For Lucretius, nothing is more reliable than the senses, and equal fides must be granted to them all.50 Other sources, while agreeing that the evidence of the senses must be taken seriously, disagree on whether all the senses are equally reliable. The eyes are generally considered very reliable.51 Ears can be reliable but are occasionally regarded as less reliable than eyes: both Lucian and Dio Chrysostom record the proverb that ‘eyes are more trustworthy (πιστότερα) than ears’, or ears more untrustworthy than eyes.52 Philostratus, in his Life of Apollonius of Tyana, describes how when Apollonius appears to his followers after miraculously disappearing from a courtroom, they are not convinced that he is alive until they have touched him.53 When the evidence of one’s own senses is not available, one may arrive at pistis/fides based on evidence presented by other people and one’s own powers of reason. This is not always easy, and authors regularly caution readers against accepting other people’s claims too readily, or implicitly or explicitly criticise characters in their narratives who do. In Slander (31), for instance, Lucian advises the reader that when someone tells you something bad about another person, you should not trust (πιστεύειν) their judgement, but investigate the truth for yourself. Necessarily, however, pistis/fides must often be based on other people’s evidence, allied with our own reason. In Longus’ Daphnis and Chloe (2.18), for example, a group of Methymnans are believed/trusted by their fellow-citizens when they say that they have been attacked by the Mytileneans, because they have visible wounds. Such belief-commitments always involve risk. During the Hannibalic wars, Appian describes (Libyka 31) how the Carthaginians sent an embassy to Rome to negotiate for the people. It struggled to convince the Romans that peace was the Carthaginians’ real aim. Some senators pointed out that Carthage had a history of ἀπιστία, untrustworthiness. Others argued that this time, Carthage 48 49

50

51 52 53

Frontinus, Strategemata 1.11.9; 1.11.16; cf. 8, 10f. Cf. Josephus, Antiquitates Iudaicae 5.52f. The difficulty in evaluating the rational trustworthiness of dreams etc. lies partly in the difficulty of categorising forms of sensory perception (e.g., seeing a dog in the street vs. seeing Herakles in a dream vs. seeing a vision of Romulus while waking). How far one should judge such varied experiences by the same criteria when deciding whether or not to place pistis/fides in them is discussed by some philosophers but remains unresolved in most authors (see e.g., Harris 2009, 129–134, 164–184). 4.478–484. Lucretius follows Epicurus here against the Sceptics; on the history of debate on this question, see the commentary of Bailey 1947, vol. 3, ad loc. Lucretius also thinks, however, that the mind can misinterpret even reliable sense data: e.g., 4.462–466. Though not always: e.g., Dionysius of Halicarnassus, Antiquitates Romanae 1.79.6. Dio Chrysostom 12.71; Lucian, Quomodo historia conscribenda sit 29. 8.10.1–2; cf. Luke 24.36–43.

Was pistis/fides experienced as an Emotion?

205

really needed to make peace, so could be trusted. Persuaded by this combination of Carthaginian testimony and their own reason, the Romans eventually took the risk of making peace. There is no explicitly emotional content in these examples of political pistis, but in other political contexts an emotional dimension is more readily identifiable. Describing the power struggles in Rome in the aftermath of the republican revolution, Dionysius of Halicarnassus (6.76.1–86.5) gives Brutus a long speech about the difficulty for plebeians of trusting patricians, in which pistis language features largely (6.78.3): What guarantee shall we put our trust/faith in (πιστεύοντες) when we lay down our weapons and put ourselves in the power of these men again? The decrees which the senate will establish on these matters? … Or the reputation of the envoys who provide their own pledges of good faith (πίστεις)? … Or agreements made with oaths in the names of the gods, gaining our assurances (πιστά) from them?

The repetition of pistis language, which forms the keynote of this highly emotive passage, suggests that the pistis language itself is meant to evoke emotion. Brutus continues (6.78.4) by demanding what kind of friendship (philia) and good faith (pistis) can thrive where agreements are made without trust. Relations are bound to break down into suspicion, mutual accusations, jealousy, hatred, and every other kind of evil. The association of pistis with philia, and their absence with suspicion, jealousy, and hatred, again suggests that effective political trust, which in part is a matter of judgement based on evidence and reason, also has an emotional dimension. The principle that it is desirable to combine evidence with reason to arrive at trust or belief operates in a somewhat specialised form in religious contexts. In Jewish Antiquities (2.168), Josephus tells how when Jacob heard from his other sons about Joseph’s career in Egypt, he did not find it incredible when he took into consideration the greatness of the works of God and God’s kindness towards Jacob in the past. Jacob regards his existing trust in God and experience of God’s goodness as as firm a foundation for belief as any other evidence.54 According to Dionysius of Halicarnassus (2.68.1–3), the Romans’ belief that Vesta manifests herself in defence of virgins who have been falsely accused of unchastity rests on a similar conviction that the gods favour the good and intervene in history to support them. In both cases, attitudes to the divine, which are presented as a matter of cognitive belief, plausibly have an emotional aspect too, which we might identify tentatively as gratitude for or appreciation of a god’s benignity.55 Conversely, lack of trust or belief in the divine may be linked with the absence of an emotional response to them. Plutarch argues that the judgement (κρίσις) that nothing blessed or incorruptible exists (itself presumably based on a combination of experience and a priori reasoning), leads to ἀπιστία τοῦ θείου, lack of trust or belief, and 54 55

Cf. Josephus, Antiquitates Iudaicae 2.60, 7.122. The same principle is invoked in middle- and neo-Platonist philosophy and in Jewish and Christian theology indebted to it: see Edwards (forthcoming).

206

Teresa Morgan

lack of fear of the gods.56 That absence of belief leads to the absence of an emotional response to the gods does not prove that belief has an emotional dimension, but Plutarch’s conclusion that a proper relationship with the gods sits halfway between atheism and superstition, with its excessive fear of the gods, suggests that a modicum of emotion is appropriate in human relations with the divine.57 When the evidence of different kinds of authority comes into conflict, it is never certain which will prove strongest. Seneca cites the proverb that ‘there is no fides except in a wise man’, but in other contexts, the wise man’s authority may need confirmation from another source.58 Philostratus, for instance, in his Life of Apollonius of Tyana (8.25.2–26.1), tells how Apollonius saw the murder of Domitian in Rome in a vision. He told the Ephesians what he had seen, and soon rumours began to emanate from Rome too. Nevertheless, the Ephesians were not completely convinced until messengers came from Rome to make the good news official. One may also, when assessing someone else’s credibility, take into account arguments he does not make himself. So Tacitus tells us in the Germania (39.1) that we can have fides in the antiquity of the Suebi – not based on what they tell Romans but based on the very primitive form of their religion. In these examples, however, there is no obvious reason to identify an emotional dimension in the believer’s thought process. Cognitive pistis/fides based on evidence and/or reason occurs in a wide range of situations and relationships. Not all of them can be shown to involve emotion, but many, I suggest, plausibly can. The examples above also show how often reason and evidence are not enough to create rational cognitive processes, arguments, or conclusions. All too frequently, evidence is doubtful, experience partial, and arguments and conclusions debatable. In these cases, we may suspect that whom or what people trust or believe ultimately depends not only on cognition, but also on their need or desire to believe one person or thing rather than another, or on some non-rational advantage that believing confers.59 Occasionally, our sources acknowledge this possibility explicitly. In his discussion of medicines, Pliny the Elder considers whether words and incantations have an effect on the healing process. This important question, he says, has never satisfactorily been settled. 56

57 58 59

Plutarch, Moralia 165b. Plutarch begins his essay De superstitione (Moralia 164e–f) by dividing ignorance and lack of understanding of the gods into atheism and superstition, and calls superstition an emotion; the implication seems to be that atheism is cognitive but not emotional. Strabo too (1.2.8) connects belief with emotion, observing that ordinary people are induced to have pistis towards the gods through deisidaimonia, fear of gods or superstition created by myths. Plutarch, Moralia 171f. Seneca, De beneficiis 12f. Rhetoricians distinguish between ‘natural proofs’, based on the kind of evidence and reason that everyone accepts, and ‘artificial proofs’, based on more dubitable indications, arguments and examples (e.g. Quintilian, Institutio oratoria 5.8.1–9.37; cf. Lucretius 5.104–106). In modern terms, artificial proofs are based on the coherence of our experience of the world, to which we have an emotional as well as a rational commitment, as opposed to the cognitive correspondence of a claim to evidence (Mitchell 1994, 28–51; Helm 2000, 23–84).

Was pistis/fides experienced as an Emotion?

207

The wisest authorities reject fides in words and incantations, but the mass of people always believes (credit) them. The basis for the masses’ belief might be many things, including experience or testimony, but the contrast Pliny draws between them and well-informed non-believers, suggests that there is an element of sheer determination about their views: they believe, in spite of evidence, because they want to believe or because it serves them in some non-rational way to believe.60 Sextus Empiricus thinks something similar about µυθικὴ πίστις, popular belief in myths.61 Stories such as the legends of Cronus, he says, are unhistorical and completely fictitious; nevertheless, they win pistis from many – apparently because people simply want to believe them or gain something non-rational from believing them. 5 PISTIS/FIDES AS A VIRTUE Finally, we turn to one further way in which pistis/fides is sometimes described in our sources, which may affect whether they, or we, understand it as an emotion. Like philosophical lists of passions, lists of the canonical philosophical virtues do not include pistis or fides. But pistis/fides is quite often described in literary sources as a virtue (virtus or ἀρετή), including by authors with philosophical affiliations.62 Diodorus Siculus describes Pittacus, tyrant of Mitylene, as perfect in virtue: statesmanlike, wise, and κατὰ τὴν πίστιν δίκαιος (‘just in keeping faith’). Listing virtutes of Romans of past generations in The War against Catiline (9.1–2), Sallust includes his belief that they were loyal (fideles) to their friends. Silius Italicus tells us (14.79–84) that one of the virtues of Hiero of Syracuse was that he was slow temerare fidem, ‘to break faith’. At the beginning of The Republic (1.2) Cicero offers a long list of virtues (which, he argues, we must not merely possess but use, ideally in the service of the state): iustitia, fides, aequitas … pudor, continentia, fuga turpitudinis, adpetentia laudis et honestatis … fortitudo.63 An even longer list makes its way into his second Catilinarian oration (2.25), as Cicero compares the virtues of his own 60

61 62

63

Pliny, Naturalis historia 28.10–19. On irrational beliefs of (usually) the masses, women or children, see e.g., Cicero, Topica 78; Pliny, Naturalis historia 28.52; Plutarch, Moralia 503d, 589d; Phocion 26.3; Strabo 1.2.8. Sextus Empiricus, Outlines of Pyrrhonism 1.147. Pistis/fides may also simply be an attribute of a person without being specified as a virtue: e.g., Plutarch, Romulus 7.4f.; Pompeius 1.3; Aemilius Paulus 2.6; Onosander, Strategicus 2.5; Quintilian, Institutio oratoria 5.7.8; Dio Chrysostom 32.96; Lucian, Cataplus 23; Demosthenis Encomium 18; Cassius Dio 14.6b; Marcus Aurelius, Ta eis heauton 1.15.2; 9.42.4. On pistis as part of a man’s nature (physis), see also Epictetus 1.3.7, 2.4.1, 2.8.23; Artemidorus 2.69. At 3.27 he again lists virtues of aequitas, iustitia, fide; cf. De inventione 164f.; De partitione oratoria 78, 90.

208

Teresa Morgan

allies with the vices of Catiline and his. On the good side, the list includes pudor, pudicitia, fides, pietas, constantia, honestas, continentia, aequitas, temperantia, fortitudo, prudentia. Seneca, reviewing virtues in Letter 88.29, describes fides as sanctissimum humani pectoris bonum, ‘the most holy good in the human breast’. In On Anger (2.28.1–2) another list of goods includes pietas, humanitas, liberalitas, iustitia, fides.64 In On Rhetoric (2.343), Cicero lists fides, along with mercy, justice, kindness, and courage, as a virtus which benefits not so much its possessor as the human race in general. Plutarch tells us that it is absurd to regard women as not participating in virtue (aretē). They have prudence, pistis, and justice, and many women have been as courageous as men.65 According to Marcus Aurelius (3.11.2), nothing creates greatness of mind more than the ability to examine everything that happens to us in life, and decide which virtue it requires of us – gentleness, courage, truth, pistis, guilelessness, self-sufficiency, or some other. Sometimes, too, pistis/fides is associated with other qualities that are regularly defined as virtues, even if it is not itself explicitly described as one. Most obviously, justice (δικαιοσύνη/iustitia) is universally regarded as a virtue in the classical world, and pistis and (especially) fides are very commonly associated with it.66 In a letter to Atticus (7.2.7), Cicero complains that Cato, despite giving him an unsought testimonial for integritas, iustitia, clementia, and fides, has refused a request. In On Friendship (19), criticising the Stoics for calling only the wise man good, he insists that anyone should be considered good who demonstrates fides, integritas, aequitas, liberalitas … constantia and who abstains from cupiditas, libido, audacia. Caesar praises Diviciacus, brother of Dumnorix, for his goodwill towards the Roman people, fides, iustitia, and temperantia. It seems clear that pistis/fides can at least sometimes be thought of as a virtue.67 Can something that can be thought of as a virtue be an emotion? This is not the place to discuss ancient theories of virtue, but we may note that in principle, a convergence of virtues and emotions should cause problems for most of the major schools of philosophy. For Platonists, the rational part of a virtuous man’s soul rules the irrational, emotional part. For Peripatetics, the doctrine of the mean suggests that no strong emotion, at least, can be allowed to rule the good man’s

64 65 66

67

Virtus and bonum are not necessarily identical, but since other goods in this list are certainly virtues, they seem likely to mean the same here. Plutarch, Moralia 749b. E.g. (of many possible examples) Cicero, De re publica 2.2; De officiis 2.32; Plutarch, Moralia 275a; Juvenal 13.33–37, 91; Cicero, De officiis 1.23 (fides as foundation of iustitia); Plutarch, Cato Minor 44.7–8 (dikaiosunē as foundation of pistis). The cult of Fides in Rome is accepted as a hypostasis of the cult of Jupiter, the god most strongly associated with justice. Caesar, De bello Gallico 1.19. Cf. Cicero, De re publica 3.27; De officiis 1.121; De oratore 2.343; De partitione oratoria 90; Valerius Maximus 3.3 ext. 2; Pliny, Epistulae 2.9.4; Dio Chrysostom 31.32.

Was pistis/fides experienced as an Emotion?

209

thoughts and actions. The Stoic ideal of apatheia seems to rule emotion absolutely out of the life of the sage. Recent scholarship on all these schools, however, suggests however that such a view would be simplistic. Philosophers from Plato onwards were intensely interested in the emotions, and all schools found ways of accommodating at least some emotions, in some form or in some degree, within the make-up of virtuous human beings.68 Even for philosophers, it seems, emotion and virtue need not be incompatible, and if philosophers could countenance the convergence of virtue and emotion, at least in some contexts and some configurations, other authors and audiences are likely to have worried about it less. Seneca’s On Mercy 1 is a good example of an essay in which a self-identified philosopher, addressing a non-specialist audience (primarily the emperor Nero, but implicitly a wider readership), entwines the language of virtue and emotion to powerful effect. Seneca begins (1.1) by describing the virtue clementia as ‘the greatest pleasure of them all’, and praises Nero (1.5) for desiring for himself (concupisti) innocentia, ‘innocence [of wrongdoing]’. He calls the public interest (4.3) cara, ‘dear’, to rational men, and tells Nero (5.7) that an emperor should look with pleasure (libens) on good citizens and rejoice (gaudeat) to see them living. He even (11.2) characterises clementia as ‘the truest temperance of mind/heart and an all-embracing love of the human race as of oneself’ (verissima animi temperantia et humani generis comprehendens ut sui amor). Among the qualities of a good ruler (13.4) are that he is inclined to gentleness, desires the approval of his people, is happy when the populace shares his good fortune, and bears a loveable countenance (vultu … amabilis). For Seneca, then, in this essay, there is a strong affective aspect to a good ruler’s engagement with his state, and his subjects’ engagement with him. The emperor should desire virtue and enjoy it, and positive emotions between ruler and subjects are a good thing. So far, we have seen pistis and fides and their cognates described in terms indistinguishable from those used of emotions. We have seen them described as closely linked with a range of emotions, and as aspects of cognition, which itself sometimes has an affective dimension. We have seen them described in terms very similar to those in which virtues are described and closely linked with virtues, which themselves may have an affective dimension, at least in some contexts. Can we conclude that pistis/fides is experienced in the early Roman empire as an emotion? It is clear that pistis/fides is not solely an emotion, and in some contexts it may not have much, if any, emotional resonance.69 Nevertheless, we can, I think, conclude with some confidence that in many contexts, pistis/fides has an emotional dimension, and at times a strong one. What is more, pistis/fides language with a likely emotional dimension occurs across a wide range of authors and gen68 69

See notes 9 and 11–15, above. Though it would be possible to argue that even technical concepts such as legal good faith and security carry some emotional resonance.

210

Teresa Morgan

res, and in a wide range of contexts, which suggests that it is not simply a construct deployed by a few authors for limited purposes, but that it is part of the mentalité of our sources, and hence very likely of Greek- and Latin-speaking inhabitants of the late Republic and early Principate in general. On that basis, we can also suggest that the role of pistis/fides in creating, maintaining, and articulating Graeco-Roman socio-cultural relations is in part an emotional one, though how exactly it might operate in that aspect is beyond the scope of this study to consider. 6 PISTIS AND EMOTION IN THE NEW TESTAMENT If pistis has an emotional dimension in the experience of Greeks and Romans of the early Principate, it becomes more likely that early Christians also experienced pistis, at least in some contexts and to some degree, as an emotion. This is not the place to discuss pistis in all its aspects in early Christianity,70 but we can note some ways in which the pistis lexicon is deployed in Christianity’s foundational texts, that would have been recognisable to contemporary non-Christians. In the New Testament, pistis and pisteuein are occasionally located in, or closely connected with the heart (kardia). At Romans 10.10, Paul tells his readers that, ‘One believes with the heart (καρδίᾳ γὰρ πιστεύεται) and so is justified,’ while at Acts 4.32, we hear that, ‘The whole group of those who believed (τῶν πιστευσάντων) were of one heart and soul (καρδίᾳ καὶ ψυχῇ).’71 ‘Do not let your hearts be troubled,’ John’s Jesus tells his disciples (John 14.1). ‘Believe (πιστεύετε) in God; believe also in me.’ The pistis lexicon appears regularly in the New Testament in conjunction with the language of emotion. Faith and love (ἀγάπη), famously connected in 1 Corinthians 13.2 and 13.13, are also linked by Paul at 1 Thessalonians 3.5–7, and several times by the authors of Ephesians and Colossians, as among the most desirable qualities of followers of Christ.72 ‘We have heard of your faith (pistin) in Christ Jesus and of the love (ἀγάπην) you have for all the saints,’ says the author of Colossians encouragingly (1.4). Jesus assures his disciples in John’s gospel (16.27): ‘The father himself loves you (φιλεῖ), because you have loved me and believed that I came from God.’ The author of Revelation praises the Church of Thyatira (2.19): ‘I know your works – your love, faith, service, and patient endurance.’ 70

71 72

See especially Bultmann 1955, ch. 4; Bultmann and Weiser 1961; Ljungman 1964; Furnish 1968, 181–193; O’Rourke 1973; Hays 1981; Hooker 1989; Lührmann 1992; Wallis 1995; Schumacher 2009; Wolter 2011, 310–316, 359–365. Translations are taken from the New Revised Standard Version Bible. Ephesians 6.21, 23; Colossians 1.4, 4.9. Since the authors of Ephesians and Colossians are likely to be followers of Paul, their use of these two terms together suggest that Paul’s use of them is already well-known and thought significant.

Was pistis/fides experienced as an Emotion?

211

Pistis is connected with hope and confidence: repeatedly by Paul, and also by the synoptic gospels.73 ‘Take heart, (θάρσει), daughter; your faith has made you well,’ Jesus tells the woman with the haemorrhage.74 And we should surely hear a challenge to hope or confidence in Jesus’ saying to the disciples:75 Truly, I tell you, if you have faith and do not doubt … even if you say to this mountain, ‘be lifted up and thrown into the sea’, it will be done. Whatever you ask for in prayer with faith you will receive.

Matthew links pistis with pity (ἔλεος), in his criticisms of the Pharisees (23.23), who he claims lack both qualities (along with justice or righteousness). Pistis is contrasted in a handful of passages with fear. ‘Why are you afraid, you of little faith (ὀλιγόπιστοι)?’ Jesus demands of the disciples after he quells a storm at sea.76 ‘Do not fear. Only believe (πίστευε),’ Jesus tells the grieving father of a dead child.77 The cognitive-affective ambiguity of pistis is matched by that of apistia, so when Mark tells us that Jesus could do no deeds of power at Nazareth because of the apistia of the people, he may be referring to their cognitive scepticism or their lack of trust, or perhaps both.78 Pistis occasionally has a more clearly cognitive aspect, particularly where it is juxtaposed with doubt. Paul tells us that doubts among followers of Christ about whether or not they should keep the Jewish food laws are a sign of weakness of faith (Romans 14.22f.), while the author of the Letter of James encourages his community to ask for wisdom, ‘in faith, never doubting, for the one who doubts is like a wave of the sea, driven and tossed by the wind’ (1.5–6). ‘You of little faith (ὀλιγόπιστοι), why did you doubt?’ demands Matthew’s Jesus of the disciples, invoking both emotional and cognitive resonances of pistos after they have been frightened by another storm on the Sea of Galilee.79 Last but not least, pistis language is sometimes found in the New Testament in association with language of goodness or virtue. Concluding his first Letter to the Corinthians, Paul exhorts them, ‘Keep alert, stand firm in your faith, be courageous, be strong’.80 ‘Well done, good and trustworthy (πιστέ) servant,’ says the master to two slaves who, left in charge of some of his talents, have added to them.81 We have already seen that the Pharisees, according to Matthew’s Jesus, lack justice, mercy, or faith (κρίσιν … ἔλεος … πίστιν).82 And pistis appears in 73 74 75 76 77 78 79 80 81

82

Romans 4.18, 15.13; 1 Corinthians 13.13; 2 Corinthians 5.67. Matthew 9.22. Cf. Mark 5.34; Luke 8.48. Matthew 21.21; cf. Mark 11.24; Luke 17.5. Matthew 8.26. Cf. Mark 5.34; Luke 8.48. Mark 5.36. Cf. Luke 8.50. Mark 6.5f. Cf. Matthew 13.58. Matthew 14.31; cf. 21.20. 1 Corinthians 16.13. Matthew 25.21, 23. Cf. Luke 19.17 (in a slightly different story). Even if pistis is thought to bear a distinctive meaning in the New Testament, pistos is usually assumed to bear its ordinary range of Greek meanings. Matthew 23.23.

212

Teresa Morgan

Galatians 5.22 in Paul’s famous list of the gifts of the Spirit, which also includes the emotions love and joy, and the virtues patience, generosity, kindness, and selfcontrol.83 Scholars of the New Testament tend to look for meanings of pistis language in the theology of authors and texts themselves rather than in the mentalité of their surrounding culture. Nevertheless, there are strong reasons for supposing that the way the New Testament uses pistis language is not likely to be radically different from the way in which it is used in wider Graeco-Roman culture at this time (though Christian understandings of pistis/fides evolve away from common usage to some extent over the next few centuries). In the first place, it would be very unusual for a new cult to take a term in common use and immediately assign it a new meaning (as opposed to evolving new meanings gradually, over time). Even more significantly, primitive Christianity was a missionary cult, and missionaries aim to communicate effectively with their audiences. We should therefore assume, unless there is compelling evidence to the contrary, that early Christians used pistis language – and all the other terms that were important to them – in a way that would have made sense to their audiences (both Jewish and gentile), which means: within its normal range of meanings. Alongside theological meanings of pistis language, we should therefore look for meanings of pistis language in the New Testament that are congruent with meanings in wider use at the time. The links we have seen between pistis language and the language of cognition, emotion, and virtue in the New Testament suggest that, as in the culture around it, pistis does indeed have an emotional aspect for very early Christians. BIBLIOGRAPHY Annas, J. (1992) Hellenistic Philosophy of Mind, Berkeley. Asmis, E. (1990) Philodemus’ Epicureanism, in Aufstieg und Niedergang der römischen Welt II.36.4, Berlin, 2228–2252. Aune, D. (2008) The Problem of the Passions in Cynicism, in Fitzgerald (ed.) 2008, 48–66. Badian, E. (1958) Foreign Clientelae, Oxford. Bailey, C. (1947) Commentary on Lucretius, De rerum natura, vol. 3, Oxford. Betz, D. (1979) Commentary on Galatians, Philadelphia. Brennan, T. (2005) The Stoic Life: Emotions, Duties, and Fate, Oxford. Brunschwig, J. and M. Nussbaum (eds.) (1993) Passions and Perceptions: Studies in Hellenistic Philosophy of Mind, Cambridge. Bultmann, R. (1955) Theology of the New Testament, volume 2, London. Bultmann, R. and A. Weiser (1961) Faith, London. Calderone, S. (1968) Pistis-fides: Ricerche di storia di diritto internazionale nell’antichità, Messina. Carcaterra, A. (1984) Dea fides e fides, Studia et Documenta Historiae Iuris 50, 199–234.

83

On lists of virtues in Hellenistic philosophy and their Jewish and Christian relatives, see Betz 1979, 281–283.

Was pistis/fides experienced as an Emotion?

213

Cooper, J. (1999) Pleasure and Desire in Epicurus, in id., Reason and Emotion: Essays on Ancient Moral Psychology and Ethical Theory, Princeton, 485–514. Corrigan, J. (2008) The Oxford Handbook of Religion and Emotion, Oxford. Deissmann-Merten, M. (1965) Fides romana bei Livius, Berlin. Dillon, J. (1977) Middle Platonists, London. ––– (1983) Metriopatheia and apatheia: Some Reflections on a Controversy in Later Greek Ethics, in J. Anton and A. Preus (eds.), Essays in Greek Philosophy 2, New York, 508–517. Donini, P. (1995) Pathos nello stoicismo romano, Elenchos 16, 195–216. Dumézil, G. (1969) Credo et fides, in id., Idées romaines, Paris, 47–59. Edwards, M. (forthcoming) Pistis and Platonism, in Kowalzig and Morgan (eds.) forthcoming. Elliott, M. (2005) Faithful Feelings: Emotion in the New Testament, Leicester. Everson, S. (2005) Epicurean Psychology, in K. Algra, J. Barnes, J. Mansfeld, and M. Schofield (eds.), The Cambridge History of Hellenistic Philosophy, Cambridge, 542–559. Fears, J. (1981) The Cult of Virtues and Roman Imperial Ideology, in Aufstieg und Niedergang der römischen Welt II.17.2, Berlin, 827–848. Fitzgerald, J. (ed.) (2008) Passions and Moral Progress in Graeco-Roman Thought, London. Fortenbaugh, W. (2002) Aristotle on Emotion, London (second edition). ––– (2008) Aristotle and Theophrastus on the Emotions, in Fitzgerald (ed.) 2008, 29–47. Fraenkel, E. (1916) Zur Geschichte des Wortes fides, Rheinisches Museum n.f. 71, 187–199. Freyburger, G. (1986) Fides. Étude sémantique et religieuse depuis les origins jusqu’à l’époque augustéenne, Paris. Freyburger, G. (2002) La fides civique, in S. Ratti (ed.), Antiquité et ψitoyenneté, Paris, 341–347. Furnish, V. (1968) Theology and Ethics in Paul, Nashville. Graver. M. (2002) Cicero on the Emotions: Tusculan Disputations 3 and 4, Chicago. ––– (2007) Stoicism and Emotion, Chicago. Grimaldi, W. (1988) Aristotle Rhetoric II: A Commentary, New York. Gruen, E. (1984) Greek pistis and Roman fides, Athenaeum 60, 50–68. Hankinson, J. (1971) Actions and Passions: Affection, Emotions, and Moral Self Management in Galen’s Philosophical Psychology, in J. Anton and A. Preus (eds.), Essays in Greek Philosophy 2, New York, 184–222. Harris, W. (2009) Dreams and Experience in Classical Antiquity, Cambridge, Mass. Hays, R. (1981) The Faith of Jesus Christ, Grand Rapids. Heinze, R. (1929) Fides, Hermes 64, 140–166. Hellegouarc’h, J. (1972) Le vocabulaire latin des relations et des partis politiques sous la République, Paris. Helm, P. (2000) Faith with Reason, Oxford. Holder, P. A. (1999) Exercitus pius fidelis: The Army of Germania Inferior in AD 89, Zeitschrift für Papyrologie und Epigraphik 128, 237–250. Hooker, M. (1989) Pistis Christou, Journal of New Testament Studies 35(3), 321–342. Kaster, R. (2005) Emotion, Restraint, and Community in Republican Rome, Oxford. King, C. (2003) The Organization of Roman Religious Beliefs, Classical Antiquity 22, 275–312. Kennedy, G. (1963) The Art of Persuasion in Greece, Princeton. ––– (1969) Quintilian, New York. ––– (1972) The Art of Rhetoric in the Roman World, Princeton. Knuuttila, S. (2004) Emotions in Ancient Philosophy, Oxford. Konstan, D. (1997) Friendship in the Classical World, Cambridge. ––– (2006) Emotions of the Ancient Greeks, Toronto. Kowalzig, B. and T. Morgan (eds.) (forthcoming) Belief and Conceptions of the Divine in the Ancient World, Oxford. Ljungman, H. (1964) Pistis. A Study of its Presuppositions and its Meaning in Pauline Use, Lund. Long, A. (2002) Epictetus, Oxford.

214

Teresa Morgan

––– (2005) Stoic Psychology, in K. Algra, J. Barnes, J. Mansfeld, and M. Schofield (eds.), The Cambridge History of Hellenistic Philosophy, Cambridge, 560–584. Long, A. and D. Sedley (1987) The Hellenistic Philosophers, Cambridge. Lührmann, D. (1992) Faith, in Anchor Bible Dictionary, vol. 2, New York, 749–758. Matthews, G. (1980) Ritual and the Religious Feelings, in A. Rorty (ed.), Explaining Emotions, Berkeley, 339–353. Mitchell, B. (1994) Faith and Criticism, Oxford. Nock, A. D. (1925) Studies in the Graeco-Roman Beliefs of the Empire, Journal of Hellenic Studies 45, 84–101. Nörr, D. (1991) Die Fides im römischen Völkerrecht, Heidelberg. Nussbaum, M. (2001) Upheavals of Thought: The Intelligence of the Emotions, Cambridge. O’Rourke, J. (1973) Pistis in Romans, Catholic Biblical Quarterly 35, 58–76. Piccaluga, G. (1981) Fides nella religion romana di età imperiale, in Aufstieg und Niedergang der römischen Welt II.17.2, Berlin, 703–735. Pringsheim, F. (1931) Aequitas und bona fides, Milan. Ramelli, I. (ed.) (2002), Studi su Fides. Premessa alle tradizioni di Eduard Fraenkel, Richard Heinze, Pierre Boyancé, Madrid. Riis, O. and L. Woodhead (2010) A Sociology of Religious Emotion, Oxford. Roberts, R. (1992) Emotions Among the Virtues of the Christian Life, Journal of Religious Ethics 20, 37–68. Russell, D. (1983) Greek Declamation, Cambridge. Sanders, K. (2008) Mens and Emotion: De rerum natura 3.136–46, Classical Quarterly 58, 362– 366. Scheid, J. (2003) An Introduction to Roman Religion, Edinburgh. Scheid, J. (2005) Quand faire, c’est croire, Paris. Schumacher, T. (2009) Der Begriff pistis im paulinischen Sprachgebrauch, in U. Schnelle (ed.), The Letter to the Romans, Leuven, 487–501. Sorabji, R. (2000) Emotion and Peace of Mind, Oxford. Thom, J. (2008) The Passions in Neopythagorean Writings, in Fitzgerald (ed.) 2008, 67–78. Verbeke, G. (1990) Moral Education in Aristotle, Washington, DC. Versnel, H. (2011) Coping with the Gods, Leiden. Wallis, I. (1995) The Faith of Jesus Christ in Early Christian Traditions, Cambridge. Winston, D. (2008) Philo of Alexandria on the Rational and Irrational Emotions, in Fitzgerald (ed.) 2008, 201–220. Winterbottom, M. (1980) Roman Declamation, Bristol. Wisse, J. (1989) Ethos and Pathos from Aristotle to Cicero, Amsterdam. Wolter, M. (2011) Paulus. Ein Grundriss seiner Theologie, Neukirchen-Vluyn. Worthington, I. (1994) Persuasion: Greek Rhetoric in Action, London. Wright, R. (2008) Plutarch on Moral Progress, in Fitzgerald (ed.) 2008, 136–150. Yunis, H. (1988) A New Creed: Fundamental Religious Beliefs in the Athenian Polis and Euripidean Drama, Göttingen.

MODELING ROMAN PRIDE Yelena Baraz 1 THE BASIC MODEL: PHAEDRUS’ PROUD JACKDAW Pride is an important social emotion in the emotional economy of the ancient Romans.1 One striking feature of its perception in Roman discourse, for an observer coming from a different culture, is the fact that pride is, almost exclusively, viewed negatively. Thus, even though we use the English word pride to translate the group of Latin terms that refer to a similar emotion, the mapping is far from perfect.2 In this paper I am interested in modeling the constituent parts of how pride is imagined in Roman discourse, from the inception of the emotion to an account of its source, the behaviors that it underlies, and the reactions that it can elicit. I begin with a text that provides a good basic introduction to how pride is conceived of in Rome. The third poem in the collection of fables written by Phaedrus3 presents the story of the downfall of a bird who wanted to rise above his natural station:4

1 2

3

4

I am grateful to Janet Downie, Robert A. Kaster, and Joshua T. Katz for their comments on this paper. I have discussed the different terms and their development in Baraz 2008. In all of the examples cited in this paper ‘pride’ is used to translate superbia, the most general Roman term for the emotion. On Phaedrus’ identity, see most recently Champlin 2005, who rejects the traditional view that Phaedrus was a freedman of the emperor Augustus and instead argues that the person hiding under the pseudonym was a Roman aristocrat; for a summary of the older view, see Currie 1984. For an extended interpretation of the fables, see Henderson 2001. For commentary on 1.3, see Oberg 2000, 47f. Phaedrus 1.3: ne gloriari libeat alienis bonis, suoque potius habitu vitam degere, Aesopus nobis hoc exemplum prodidit. tumens inani graculus superbia pinnas, pavoni quae deciderant, sustulit, seque exornavit. deinde, contemnens suos immiscet se ut pavonum formoso gregi illi impudenti pinnas eripiunt avi, fugantque rostris. male mulcatus graculus redire maerens coepit ad proprium genus, a quo repulsus tristem sustinuit notam.

216

Yelena Baraz So that no one presume to glory in what is not properly his, but rather leads a life content with his own condition, Aesop has left us the following story to learn from: A jackdaw, swollen with empty pride [superbia], took up the feathers that had fallen off a peacock, and adorned himself with them. Then, looking down his nose at his fellows, he tries to mingle among the beautiful flock of peacocks. The birds tear the feathers off the impudent impostor and chase him away from the speaker’s platform. Having been handled roughly, the jackdaw, full of grief, tried to return to his own kind, but repulsed by them he received a baleful censure. Then a certain jackdaw from among those he had previously disdained: ‘Had you been content with our ancestral seats, and willing to bear what Nature had given you, you would not have the insult from them 5 nor would you in your misfortune be feeling this rejection from us.’

This animal fable provides a useful entry point into a discussion of how the Romans conceived of the social emotion of pride, how it functioned in their society, and how it was used for calibrating the power dynamics in a wide variety of social contexts. Despite the universal appearance lent to this story by its animal characters on the one hand and the clear acknowledgment that the story is borrowed from the loose corpus of the Greek fabulist Aesop on the other,6 this is a tale rooted firmly in the life of an elite Roman male. The beautiful peacocks chase the jackdaw away from the rostra, the traditional speaker’s platform located in the Roman forum, from which only a magistrate or someone invited by a magistrate could speak.7 It appears, then, that it was more than beauty that attracted the jackdaw: he was trying to pass himself off as someone who had the right to speak from the rostra.8 Furthermore, being chased away is not his only punishment: he receives a nota, a mark that diminished the person’s stature, given by the censors in the course of conducting their duties. Thus, our jackdaw emerges as no simple bird, but a Roman man who incurred the official wrath of the establishment in striving to reach beyond his station. What is of interest to us in the context of this paper is the main quality that is seen as responsible for propelling the jackdaw

5 6 7 8

tum quidam ex illis quos prius despexerat ‘contentus nostris si fuisses sedibus et quod natura dederat voluisses pati, nec illam expertus esses contumeliam nec hanc repulsam tua sentiret calamitas.’ All translations are my own. On Aesop and Aesopic tradition, see Kurke 2010. Morstein-Marx 2004, 40: ‘He [the presiding magistrate] ... spoke himself, or brought forward onto the speaker’s platform other speakers at his sole discretion.’ A historical parallel of sorts to this scenario is found in a letter of Cicero (Epistulae ad Atticum 2.24.3), which describes Caesar’s refusal as praetor (62 BCE) to allow Quintus Catulus to speak from the rostra, forcing him to speak from below (ex inferiore loco) in a spatial representation of the power dynamic he wanted to emphasize.

Modeling Roman Pride

217

towards the peacock’s feathers in the first place, namely, his pride. He is identified in the title of the fable as graculus superbus, a proud jackdaw, and is further described (line 3) as ‘swollen with empty pride’. The simple story of the jackdaw’s misadventures in Roman politics, then, will allow us to identify the constituent parts of pride and its functioning in Roman social discourse, providing a blueprint for the rest of this investigation. The essential feature in the operation of pride, as the Romans understood it, is the presence of a status or power differential. Pride involves a self-evaluation; in particular, an evaluation that results in a mismatch between the subject’s sense of self-worth and his assessment by the relevant community or communities. As we shall see, then, there is within the discourse of pride in Rome a strong presence of a discourse of self-deception. This is certainly the case with the jackdaw, who misjudges both himself and the peacocks, misunderstanding the meaning of the feathers and presuming that they are easily transferrable. Yet, there is an additional element that is not made explicit in this story, and that is the factor that underpins the mistaken self-evaluation and leads to pride. Here, in the stripped down world of fable, it seems to be self-delusion pure and simple, but we shall see that understanding the causes of such mistakes forms an important facet of how the Romans think about pride. The next important element is the behavior that results when the author/observer identifies the underlying feeling as pride. In the case of the jackdaw, pride leads to contempt that he directs towards the group to which he belongs by birth, to assumption of improper regalia, and to an attempt to take on responsibility that is in fact beyond his reach (if we assume that he was in fact trying to climb the rostra in order to speak). Proud behavior, in turn, often provokes a reaction, and the fable is useful in showing that this reaction is twofold. There is, on the one hand, the reaction of the group generally considered to be above the subject in the hierarchy. Here, the reaction of the peacocks as a group comes in two stages. First, the peacocks present during the jackdaw’s attempt at self-elevation physically assail him and strip him of the assumed feathers, then chase him away. We could describe this immediate reaction as consisting of violence and exclusion. The jackdaw who addresses the protagonist at the end characterizes this as contumelia, insult. The second stage is the attachment of the censors’ mark, a procedure that in Rome resulted in disgrace, ignominia, and, for the members of the higher orders, loss of standing.9 This implies that in the putative bird census the jackdaw has not only failed to rise but has in fact lost something of his original position. On the other hand, we have the reaction of the members of his group of origin, which in turn rejects the protagonist and, remembering the contempt they suffered from him when he was trying to move up in the world, deny him the chance to reintegrate. Finally, we have the context in which the power dynamic is being played out. Here it is the inter-species hierarchy that stands, in a fairly transparent 9

On the censorial regulation of morals and the nota, see Nicolet 1980, 73–81; cf. Meyer 2004, 93.

218

Yelena Baraz

way, for Roman society. But the hierarchical model in which mistaken selfassessment occurs is easily transferrable, and we shall see that the model can operate in virtually every type of relationship, within hierarchies ranging from military and political to familial, financial, and even erotic. Phaedrus’ fable thus gives us a basic model to guide our exploration of the perception and representation of pride in Rome. 2 PRIDE AS FAULTY SELF-ASSESSMENT As we have seen, at the core of how the Romans see pride is the sense that the emotion arises from the proud person’s mistaken assessment10 of his or her place in a given hierarchy or relationship.11 The point that a behavior is found to be proud through an evaluative process that relies on relative positions of the people involved in a power relationship is well made by a passage from Quintilian’s rhetorical handbook, Institutes of Oratory. In the first chapter of the eleventh book, dedicated to the need for speech to be appropriate to circumstances, Quintilian writes:12 The same words are often an expression, in one man, of free spirit, in another, of fury, in yet another, of pride. The tirade of Thersites against Agamemnon is laughable: give those words to Diomedes, or some other man of equal stature, and they will seem to present to the world a great spirit. Lucius Crassus said to Philippus, ‘Should I treat you as consul even though you do not treat me as senator?’ – words of most honorable freedom, yet you would not abide it if just anyone said them.

Quintilian begins with the earliest example of presumptuous speech in the Western literary tradition: Thersites’ perception of himself as someone who can speak freely and mockingly to the king is violently rejected. In fact the mismatch between his place in the eyes of the Homeric aristocrats and the implicit claim to a higher position made by the very act, as well as the tone, of his speech is so great that the appropriate response is not violence but ridicule.13 It is the very magni10

11

12

13

A useful definition is provided by Seneca’s description of pride as ‘excessive valuation of oneself and swelling inflated beyond all the rest’ (De vita beata 10.2: nimiam aestimationem sui tumoremque elatum super ceteros). It can also come from lack of awareness of a change in one’s circumstances. One such example is Messalina on the eve of her fall in the eleventh book of Tacitus’ Annals: as a result of the extent of her pride she experiences hope and even anger (11.37.1: interim Messalina Lucullianis in hortis prolatare vitam, componere preces, non nulla spe et aliquando ira: tantum inter extrema superbiae gerebat.). Quintilian, Institutio Oratoria 11.1.37: idem dictum saepe in alio liberum, in alio furiosum, in alio superbum est. verba adversus Agamemnonem a Thersite habita ridentur: da illa Diomedi aliive cui pari, magnum animum ferre prae se videbuntur. ‘ego te’ inquit ‘consulem putem’ L. Crassus Philippo ‘cum tu me non putes senatorem?’: vox honestissimae libertatis, non tamen ferres quemcumque dicentem. Quintilian is evoking very briefly what is a highly complex episode in the Iliad. See recently Rosen 2007, 67–116; essential is Nagy 1979, 253–264.

Modeling Roman Pride

219

tude of the gap that makes this prideful speech exemplary. The second example takes us out of the Homeric world into Roman politics, and into a much more nuanced situation. Here Quintilian is referring to an episode described in great detail at the opening of the third book of Cicero’s dialogue On the Ideal Orator. Lucius Crassus is one of the main speakers in that work and in many ways represents the Ciceronian ideal. The line that Quintilian quotes is a culmination in the confrontation between Crassus and the consul for the year 91 BCE, Lucius Marcius Philippus, a confrontation that, Cicero insinuates, led to Crassus’ death. As this episode does much to shed light on the political dimensions of pride in Roman discourse, it is worth a closer look. I have argued elsewhere that behind the predominantly negative conceptualization of pride in Roman discourse lies the intense focus on preserving equality among the members of Rome’s political elite.14 The importance of this quasiegalitarianism to the self-image of the Roman republican order is reflected in the centrality of the foundational story of the republic itself, created in opposition to the abuses of Rome’s last king, Tarquin the Proud. Many confrontations between the senate as a body and an extraordinary individual whose behavior was seen as a threat to senatorial authority can be usefully analyzed with our model as a guide. But this particular incident is especially attractive precisely because both Cicero and Quintilian present it without any background,15 as a paradigmatic example of a confrontation between a senator and a consul. According to Cicero, what spurs Crassus to action is Philippus’ public attack on the senate as a deliberative body:16 ‘... he would have to find himself another council, for he could not conduct public business with the current senate.’ Crassus’ response acknowledges that there is a hierarchical relationship between the consul and the senate as a body (as well as individual senators), but it also makes clear that the distance that constitutes this particular hierarchy is a finely calibrated one. It is possible for both parties to behave in such a way that they arrogate more to their position than is appropriate, and in both cases the behavior would be an index of a certain kind of pride. Both Crassus and Philippus interpret the interaction in this way: Philippus feels that he is not being given his due, Crassus that he and his peers are degraded by the consul’s treatment. The role of the consul, Crassus says, is that of a father or a faithful guardian, but Philippus is acting like a robber, stripping the senate of its inherited dignity.17 The language that Crassus deploys in his attack on Philippus is precisely targeted to 14 15

16 17

See Baraz 2008 for full discussion. At the heart of the conflict are the controversial laws of the tribune Livius Drusus, which Crassus supported and Philippus opposed. For a brief overview of the background, see Fantham 2004, 44f. Cicero, De Oratore 3.2: quem dixisse constabat videndum sibi esse aliud consilium; illo senatu se rem publicam gerere non posse. Cicero, De Oratore 3.3: deploravit enim casum atque orbitatem senatus, cuius ordinis a consule, qui quasi parens bonus aut tutor fidelis esse deberet, tamquam ab aliquo nefario praedone diriperetur patrimonium dignitatis.

220

Yelena Baraz

evoke the image of a tyrannical monarch and the struggle for libertas, freedom, that gave birth to the republic:18 if you wish to coerce Lucius Crassus, you must cut out this tongue; and even if it is torn out my freedom will resist your license with its very breath.

The images of physical violence and repression, as well as the allusion to libido, the same kind of license that led Sextus Tarquinius, the son of the last king, to rape the noble Lucretia, paint Philippus as Tarquin and Crassus as Lucius Junius Brutus, the leader of the uprising against tyranny. The Ciceronian narrative thus focuses on the pride of the consul, and on the senator’s brave stand on behalf of his order in defense of its dignity. But Quintilian, while drawing on Cicero, takes a different approach to the interaction. While Cicero’s emphasis was on the pride of the consul, already in a superior position, who wants to elevate himself even further at the expense of the senate, Quintilian is interested in considering the critical statement from the lower position, of the senator to the consul. His presentation acknowledges that in many circumstances such a statement could be correctly interpreted as a sign of pride on the part of the senator, an attempt to resist the authority of the consul unreasonably and a refusal to accept one’s position in the political hierarchy. Hearing such a statement from some would be hard to take: fero, the verb that he uses, often occurs with words designating pride,19 and he switches here to a generalizing second person, inviting and expecting the reader to identify with the sentiment. Yet, as Quintilian emphasizes at the end of the passage quoted above, in the case of Crassus, it was a sign of the speaker’s honor and freedom. This passage shows that pride and prideful behavior can arise from both parties in a hierarchical relationship when either misapprehends the gap in the hierarchy or the nature of the power dynamic. It also demonstrates that pride is identified when an external observer notes the mismatch between a person’s selfvaluation and an external estimate. Quintilian’s change of focus in his deployment of Cicero’s narrative further underlines the importance of the observer’s perspective in the Roman understanding of pride. A new evaluative community could very well take a different view of the interaction between Crassus and Philippus. Cicero’s more detailed account, in turn, illustrates how successful rhetorical deployment of the Roman discourse surrounding pride is crucial to labeling one’s opponent’s emotion as pride as opposed to what could have been portrayed, in this instance, as righteous indignation.

18 Cicero, De Oratore 3.4, quoting Crassus’ speech (Oratorum Romanorum Fragmenta 41): si L. Crassum vis coercere: haec tibi est incidenda lingua, qua vel evulsa spiritu ipso libidinem tuam libertas mea refutabit. 19 Baraz 2008, 379.

Modeling Roman Pride

221

3 CAUSES OF PRIDE As we saw in the Phaedrus fable, texts that describe pride are not always interested in tracing its origins. Yet seeing what can lead to the self-evaluation, judged to be mistaken, that gives rise to this social emotion is an important part of understanding its makeup. Broadly speaking, the main cause of pride in Roman discourse is good fortune. Good fortune can take a variety of forms – military (victory), financial (wealth), familial (birth or marriage), physical (beauty), or social (support of the powerful) – but is perceived as leading to overconfidence and overestimation of one’s position in the relevant hierarchy. But a faulty selfassessment can also arise in other ways, such as from reliance on praise or a record of impunity. One of the earliest surviving analyses of the causes of pride is the opening of Cato the Elder’s speech On behalf of the Rhodians. An independent Rhodes, informal ally of both Rome and the Macedonian kings, was experiencing much hardship in the course of the Third Macedonian War (171–168 BCE). Having declared their support for Rome, the Rhodians did not offer any practical assistance, and in fact were mainly interested in achieving peace as swiftly as possible. It seems to have been bad luck that brought the Rhodian embassy to Rome on a peace mission soon after the announcement in the city of the outcome of the battle of Pydna (168), Rome’s decisive victory over king Perseus. The speech of the Rhodian ambassadors offended the sensibilities of the senate, and there was a movement to have Rhodes declared an enemy. Cato’s speech during the senatorial debate on how to deal with the Rhodians (167) was directed against this policy, although what precise impact it had is unclear.20 Livy’s account of the events makes clear that the Romans considered the Rhodians’ conduct a sign of their pride: they were judged both to have overestimated their own standing and to have underestimated the power of Rome. In the aftermath of the victory in Pydna, the mistake appeared particularly glaring: Livy’s narrative repeatedly emphasizes this by linking the pride of the Rhodians to their stupidity. Livy first reports that, according to some sources, the only reason the Rhodian ambassadors were summoned to the senate after Pydna was to ridicule their ‘stolid pride’.21 The language of pride and stupidity is also deployed by the Rhodians in their defense speech in 167 BCE. Especially notable is the second time they deploy this rhetoric, found towards the end of the speech:22

20 21

22

Polybius 29.10 and 19.30.4–5, 21, 23, 31; Livy 45.3.3–6, 45.20.4–25.10; Diodorus Siculus 30.24, 31.5.3; Astin 1978, 273–283; Sherwin-White 1984, 30–36; Gruen 1986, 563–566. Livy 45.3.3: tradidere quidam legatos Rhodios nondum dimissos post uictoriam nuntiatam uelut ad ludibrium stolidae superbiae in senatum uocatos esse; cf. Seneca, De beneficiis 2.13.1: o superbia, magnae fortunae stultissimum malum! (‘o pride, the stupidest evil that great fortune brings!’). Livy 45.23.13: sed tamen ea siue superbia, siue stultitia appellanda est, eadem, quae apud uos, et apud Persea fuit; 45.23.18: superbiam, uerborum praesertim, iracundi oderunt, pru-

222

Yelena Baraz … pride, especially when it finds expression in words, is hated by men prone to anger, but laughed at by the prudent, in any case if it is pride of an inferior towards a superior; no one has ever judged it deserving of capital punishment.

The Rhodian argument implies that in the emotional economy of reacting to pride, anger and punishment have the effect of lowering the standing of the superior party; by contrast, calm ridicule, to which the Rhodians declare themselves willing to submit, indicates security. It is in this context that Cato gave his speech, which is not reported by Livy but whose fragments are preserved by Gellius.23 Up to that point everyone seems to have focused on the apparently prideful behavior of the Rhodians. Cato’s speech shifts the emphasis to the emotional foundations of the Romans’ own behavior:24 I know that it is usual for the spirit of most men, when their affairs are fortunate and favorable and prosperous, to rise high and for pride and savagery to increase and to grow. And for this reason I am now much concerned, since this matter [i.e. the war] has gone so successfully, lest something untoward occur in the course of our deliberation to diminish this flourishing state of affairs, or this joy of ours turns out to be excessive. Misfortune tames and teaches what needs to be done; enjoyment of good fortune tends to drive men away from correct counsel and understanding. For this reason I urge and advise you all the more that this matter be postponed for some days, until we return from such great joy into self-possession.

Cato begins with a general diagnosis: pride is a product of good fortune, the importance of this cause emphasized by the polysyndetic tricolon of nearly synonymous adjectives (secundis atque prolixis atque prosperis). He then moves to the specific case before him, analyzing the Romans’ desire to punish the Rhodians as a particularly acute instance of pride in the immediate elation of victory. The passage abounds in the language of expansion, elevation, and growth, all temporary, Cato suggests, and pushing the Romans beyond what is justified based on their place in the balance of foreign relations in the Mediterranean. The dynamic that Cato fears is precisely that of Phaedrus’ jackdaw: trying to take on too much leads to loss of prior position; excess will lead to destruction. Cato presents himself as an objective observer, able to abstract himself from the joy of the moment and

23 24

dentes inrident, utique si inferioris aduersus superiorem est; capitali poena nemo umquam dignam iudicauit. Fragments of the speech with commentary: Courtney 1999, 78–85; Cugusi and Sblendorio Cugusi 2001, I 328–340; Calboli 2003. Oratorum Romanorum Fragmenta 163 = Aulus Gellius 6.3.14: scio solere plerisque hominibus rebus secundis atque prolixis atque prosperis animum excellere atque superbiam atque ferociam augescere atque crescere. quo mihi nunc magnae curae est, quod haec res tam secunde processit, ne quid in consulendo advorsi eveniat, quod nostras secundas res confutet, neve haec laetitia nimis luxuriose eveniat. advorsae res edomant et docent, quid opus siet facto, secundae res laetitia transvorsum trudere solent a recte consulendo atque intellegendo. quo maiore opere dico suadeoque, uti haec res aliquot dies proferatur, dum ex tanto gaudio in potestatem nostram redeamus.

Modeling Roman Pride

223

correctly (recte) evaluate Rome’s new position and the dangers inherent in acting on the basis of overestimating it. The last fragment of the speech quoted by Gellius comes back to the topic of pride:25 They say that the Rhodians are proud, throwing at them a word that I would least want to be used of myself and my children. But grant that they are proud. How does this concern us? Or are we roused to anger if someone is prouder than we are?

Cato’s identification of pride as the least desirable descriptor to a Roman hints at its inescapable connection to the Tarquins and to tyranny. Cato once again tries to expose the chain of emotions resulting from pride, cautioning against anger and against senseless and dangerous competition. Placed in a rhetorical question, superbior, a comparative form of an adjective that is itself comparative in nature and historically full of negative connotations, is meant to reveal how inappropriate Rome’s reaction is: the war has been won, and continuing the competition in the arena of pride would allow the joy of victory to turn into excess and undermine the fruits of that very good fortune. The functioning of good fortune as producing pride and then leading to downfall is represented similarly even when the political overtones associated with pride, explicitly present in the texts discussed so far, do not play a significant role. One example is offered by Ovid’s depiction in the third book of the Metamorphoses of the downfall of Narcissus, a boy made so proud by his exceptional beauty that he perishes from self-love. Here, too, the poet singles out pride as his character’s defining characteristic when he first introduces him:26 Many youths, many girls desired him but such a harsh pride resided in his tender frame, that no youths, no girls touched him.

The poet does not deny Narcissus the superiority of beauty, and yet the narrative reveals that the emotion it inspires in the boy, his pride, is excessive. In Ovid’s narrative his self-regard leads to his disproportionately violent rejection of Echo, as he says that he would rather die than be with her, and then to the rejection of others, characterized by the poet as mockery.27 His pride results in the inability to envision an appropriate partner and finds fitting punishment in his death from the desperate love for his own reflection.28 25

Oratorum Romanorum Fragmenta 169 = Aulus Gellius 6.3.50: Rhodiensis superbos esse aiunt id obiectantes, quod mihi et liberis meis minime dici velim. sint sane superbi. quid id ad nos attinet? idne irascimini, si quis superbior est quam nos? 26 Ovid, Metamorphoses 3.353–355: multi illum iuvenes, multae cupiere puellae; sed fuit in tenera tam dura superbia forma, nulli illum iuvenes, nullae tetigere puellae. 27 Rejection of Echo: id. 3.391: ‘ante’ ait ‘emoriar, quam sit tibi copia nostri’; mockery: luserat, 403. 28 On Narcissus, see the commentary of Barchiesi and Rosati 2005, with further bibliography.

224

Yelena Baraz

4 PROUD BEHAVIORS We have already seen several examples of the kinds of behavior inspired by pride. The jackdaw attempted to join a superior group and climb the rostra, Thersites tried to mock the king, the consul Philippus treated the senate with contempt, and the Rhodians were seen as disdaining Roman power by daring to bring a proposal for peace on the eve of a victory, while Cato portrayed Rome’s plans for excessive punishment as an expression of its own pride; finally, Narcissus’ rejection of any partner but himself showed his arrogance. A pattern of behavior has emerged from these examples. Specific behaviors are often connected to pride and function almost as symptoms, external manifestations that allow the underlying emotion to be identified. How pride manifests itself is dependent in part on the relative position of the subject in a given power hierarchy: the inferior’s pride translates into presumption, the superior’s into contempt. Such contempt is also often seen as finding expression in either mocking speech or cruel action. In other words, an emotion that results from an excessive self-valuation results in a variety of excessive behaviors. An extended example of cruelty explicitly colored by pride is found in Seneca’s treatment of the emperor Caligula in the third book of his treatise On Anger. While anger is the subject of the work, Caligula’s function within it is to provide an extreme example, an instance of cruelty taken to an extraordinary level, and it is precisely the emotional baggage that comes with pride that makes Caligula’s behavior exceptional.29 The particular incident in Caligula’s impressive record of cruelty that is singled out for being designated as proud involves the unusual circumstances of the execution of three senators:30 Just recently Gaius Caesar whipped and tortured, in one day, Sextus Papinius, whose father was a consular; Betilienus Bassus, his quaestor, a son of his procurator; and others, both senators and Roman knights, not in the course of an inquiry but to sate his spirit. Then, he was so unable to endure deferring his pleasure, as his cruelty demanded much of it and without delay, that, while walking in the colonnade of his mother’s gardens (which separates the portico from the riverbank), he beheaded some of these in the presence of matrons and other senators, by lamplight. What was the rush? What danger, either public or private, was threatening in the course of one night? How much trouble was it, finally, to wait for daylight, so as not to slaughter senators of the Roman people while wearing sandals? It is essential to know how 29 30

On Caligula as exemplum in Seneca’s corpus, see Wilcox 2008. Seneca, De ira 3.18.3–19.1: modo C. Caesar Sex. Papinium, cui pater erat consularis, Betilienum Bassum quaestorem suum, procuratoris sui filium, aliosque et senatores et equites Romanos uno die flagellis cecidit, torsit, non quaestionis sed animi causa; deinde adeo inpatiens fuit differendae uoluptatis, quam ingentem crudelitas eius sine dilatione poscebat, ut in xysto maternorum hortorum (qui porticum a ripa separat) inambulans quosdam ex illis cum matronis atque aliis senatoribus ad lucernam decollaret. quid instabat? quod periculum aut priuatum aut publicum una nox minabatur? quantulum fuit lucem expectare denique, ne senatores populi Romani soleatus occideret! quam superba fuerit crudelitas eius ad rem pertinet scire, quamquam aberrare alicui possimus uideri et in deuium exire; sed hoc ipsum pars erit irae super solita saeuientis.

Modeling Roman Pride

225

proud his cruelty was, although we could appear in someone’s eyes to be wandering off topic and moving into the outlandish; but this very thing is a feature of anger when it rages beyond what is usual.

In what follows a second-person interlocutor is imagined, in a technique common in Seneca’s treatises, to object to the idea that this incident constitutes a particularly extraordinary example of Caligula’s behavior, given just how many stories of his cruelty there are, many of them more famous and memorable. Seneca employs this objection as a foil to move his argument along, but it is also useful in the context of pride as an invitation to go back to the description of the nocturnal execution and understand what marks it as an expression not just of Caligula’s anger and cruelty in general but of cruelty as a derivation of his pride. It is no surprise that pride as a desire to elevate oneself beyond one’s actual position should be assigned to this emperor. In fact, ascription of pride understood in this way provides a useful model for describing the patterns of behavior preserved in our sources, both senatorial and hostile. Most of the actions reported as scandalous or insane fit into a pattern of Caligula’s trying to widen the gap between himself and everyone else by both moving further upwards (thus, gestures towards deification, the only place to go from the highest position available to a man) and demeaning his closest inferiors, the senators (the impression that Caligula planned to install his horse Incitatus as consul is the best and most famous instance of this).31 Given this general background we can then ask what leads Seneca to choose this particular episode. As the objector makes clear, it is not the fact of the execution itself – simply cruel, and one of very many – but the circumstances surrounding it. In how he chooses to conduct the execution, Caligula shows an excessive indulgence of his own pleasure (there is no possible reason why he would postpone what he desires to do now until the next morning) and demeans both the victims and the unwilling spectators, women and the victims’ senatorial peers. What seems crucial to Seneca is a point that could seem minor to an outside observer: Caligula’s footwear. The fact that the emperor is wearing sandals indicates his unwillingness to treat the occasion as deserving the dignity of a public event.32 The casualness of the killing elevates the executioner and degrades the victims and the witnesses, and, by implication, the senatorial order as a whole. This incident makes a good case that it was Caligula’s pride, not his anger, which made him a bête noire in Seneca’s book. Cruelty and violence are common expressions of pride. But just as common is the translation of pride into a verbal assertion of superiority, which can take the 31 32

Suetonius Caligula 55.3; Dio Cassius 59.14.7; see most recently Winterling 2011, 103f. The first great Ciceronian villain, Verres, was similarly attacked by the orator for disgracing his public position and showing his contempt for his audience by wearing soleae as part of a generally inappropriate ensemble. When governor, he allegedly appeared thus attired to greet the arriving fleet in sight of Roman citizens and Sicilian provincials: stetit soleatus praetor populi Romani cum pallio purpureo tunicaque talari muliercula nixus in litore (Cicero, In Verrem 2.5.86).

226

Yelena Baraz

form of self-elevation, in boasting, or of attack on the standing of others, in mockery. Naturally, the two often go together. A familiar example of such behavior is found in the ninth book of Vergil’s Aeneid, when, with Aeneas away from the camp, the Trojans are under pressure and the position of young Ascanius, who has never before taken part in battle, is in question. Numanus Remulus, the son-in-law of Aeneas’ chief opponent, Turnus, elated by the capture and punishment of Nisus and Euryalus and by the success of Turnus’ rampage through the Trojan ranks earlier in the book, assails the Trojans in an extended speech:33 That man, in front of the first battle line, was pronouncing some things that deserve to be reported, some that do not, and was strutting about puffed up in his heart because of his new royal connection, making himself great with his shout: ‘Aren’t you ashamed, to be hemmed in once again by siege works, twice-captured Phrygians, and hold out walls to protect you from death? Look who is demanding marriage by war! What god, what madness drove you to Italy? The sons of Atreus are not here, nor the weaver of lies Ulysses: a hard race from birth, we carry our sons to the river right away and harden them with the savage cold of the waves; our boys stay awake for the hunt and tire themselves out in the forests, their game is to tame horses and to shoot arrows from a bow. 33

Vergil, Aeneid 9.595–620: is primam ante aciem digna atque indigna relatu uociferans tumidusque nouo praecordia regno ibat et ingentem sese clamore ferebat: ‘non pudet obsidione iterum ualloque teneri, bis capti Phryges, et morti praetendere muros? en qui nostra sibi bello conubia poscunt! quis deus Italiam, quae uos dementia adegit? non hic Atridae nec fandi fictor Vlixes: durum a stirpe genus natos ad flumina primum deferimus saeuoque gelu duramus et undis; uenatu inuigilant pueri siluasque fatigant, flectere ludus equos et spicula tendere cornu. at patiens operum paruoque adsueta iuuentus aut rastris terram domat aut quatit oppida bello. omne aeuum ferro teritur, uersaque iuuencum terga fatigamus hasta, nec tarda senectus debilitat uiris animi mutatque uigorem: canitiem galea premimus, semperque recentis comportare iuuat praedas et uiuere rapto. uobis picta croco et fulgenti murice uestis, desidiae cordi, iuuat indulgere choreis, et tunicae manicas et habent redimicula mitrae. o uere Phrygiae, neque enim Phryges, ite per alta Dindyma, ubi adsuetis biforem dat tibia cantum. tympana uos buxusque uocat Berecyntia Matris Idaeae; sinite arma uiris et cedite ferro.’

Modeling Roman Pride

227

But the youth, who know how to bear work and are used to do with little, either subdue the land with mattocks or shake towns with war. In every time of life we are worn out with sword, and we exhaust the backs of our oxen with an upturned spear, nor does tired old age diminish the strength of the spirit and change its vigor: we press down the helmet on white hair and ever delight in gathering fresh booty and living by plunder. Your clothes are dyed saffron and shining purple, sloth is your heart’s desire, and you delight in the indulgent dances, your tunics’ sleeves are long and your turbans covered with fillets. Women of Phrygia – yes, not men – go traverse the heights of Dindyma, where the pipe grants its double song to those who know it well. The tambourines and the Berecynthian boxwood of the Idaean Mother call you; leave arms to men and yield to the sword.

While I cannot address the complex dynamics of Numanus’ sketches of Trojan vs. Italian identity in the framework of this paper,34 the two parts of his attack align clearly with the two types of verbal expression of pride. The poet shows that Numanus’ account of the Italian lifestyle is a boast immediately, when the novice Ascanius, following a prayer to Jupiter for help, kills him with the very first arrow he ever shoots. Furthermore, while there is much that is true, within the world of the poem, in Numanus’ picture, the final outcome of the conflict between the Trojans and Italians does mark the extent and the grandness of his claims as excessive. Numanus is driven by the recent success of his side, and Vergil describes him as swollen, tumidus, and huge, ingens (596f.). The image of pride as tumor is frequently associated with proud behavior in Latin texts.35 What pride drives Numanus to do here is exaggerate, and the poet gets this point across by having him repeat key words and synonyms: durum ... genus and duramus (603f.), fatigant and fatigamus (605 and 610), praedas and rapto (613). Further, as the speech goes on, Numanus’ self-identification with his people as a whole grows, another sign that he is getting carried away: his presentation of the ages of Italian men begins in the third person, but in line 610 he switches and concludes his description of how Italians spend their old age in the first person plural. It is also clear in the structure and language of the speech that the praise of Italian hardiness is designed as it is in order to better serve Numanus’ second goal, mockery of the Trojans. Vergil uses word repetition to highlight this aspect of the speech as well: the Italian youth are accustomed to doing with little, the Trojans are used to the pipes of Cybele (607: adsueta; 618: adsuetis); the Italians delight in plunder, the Trojans in ecstatic dances (613 and 615: iuuat). The emphasis on the warlike quality of the Italians and on the labeling of the Trojans as effeminate is particularly problematic here since Numanus is exhilarated entirely by the deeds of others: we do not see him performing any feats of valor himself; his attack is exclusively verbal. Ascanius’ joyful exclamation, as he sees his opponent 34 35

For a discussion of the episode, see Hardie 1994, 15–18, 188–205; bibliography: 188. Cf. tumens in the description of the jackdaw in Phaedrus above. See Baraz 2008, 379.

228

Yelena Baraz

fall, makes this exact point:36 ‘go, mock excellence with proud words: twicecaptured Phrygians send this answer to the Rutulians.’ Ascanius, having accomplished the goal of his prayer, identifies Numanus’ behavior as mockery born of pride, appropriates his designation of the Trojans as ‘twice-captured Phrygians’, and transforms the now lifeless body of his enemy into a message (responsa) that refutes the claims of his speech. What follows Numanus’ death further illustrates the potential for pride to follow in the immediate aftermath of success, especially military victory, and the risk that the emotion this success generates will find expression in overestimating the meaning of the episode and lead to dangerous behavior. The immediate reaction on the part of the Trojans is joy and elation:37 ‘the Teucrians follow Ascanius’ words with a shout, and roar with joy, and raise their spirits to the stars.’ Joy (laetitia), though it does often occur together with pride, does not necessarily indicate the emotion. On the other hand, the image of rising to the stars, strongly associated with deification, does imply that the possible next step based on this emotion could be disastrous, as both Ascanius and the Trojans around him might misinterpret the meaning of his first successful killing. This is prevented by the intervention of Apollo. The Trojans, both in recognizing that the presence among them is divine and in immediately obeying Apollo’s words and keeping Ascanius out of further fighting, are able to step back from the emotion that they collectively experience before it finds expression in action. 5 REACTING TO PRIDE Many factors determine what reaction pride will provoke. It depends in part, as we have seen, on whether the proud behavior in question is directed upward or downward. It also depends on the magnitude of the overestimation that gives rise to pride in the first place, from the point of the view of those who react to it. Thus, if the overestimation is so dramatic that it is expected to be obvious to a majority of observers,38 it is more likely treated as a subject for ridicule.39 If the overestimation is comparatively minor and can either sow self-doubt or be seen by the ‘victim’ as likely to give rise to a re-evaluation by other observers of his standing 36 37 38

39

Vergil, Aeneid 9.634f.: ‘i, uerbis uirtutem inlude superbis capti Phryges haec Rutulis responsa remittunt.’ Vergil, Aeneid 9.636f.: Teucri clamore sequuntur laetitiaque fremunt animosque ad sidera tollunt. In the Roman version of Stoic philosophical discourse found in the works of Seneca, the wise man, who is unassailably located at the top of any meaningful hierarchy, reacts to all proud behavior with tranquility, seeing it as unfounded and having no effect on him; see, e.g., On the Constancy of the Wise Man, a treatise devoted to proving the paradox ‘the wise man can receive no injury or insult.’ According to Livy, some of his sources claimed that the Rhodians were initially summoned to the senate ad ridiculum (45.3.3).

Modeling Roman Pride

229

or qualities, it can provoke a variety of more severe reactions. In the emotional domain, this can take the form of anger, grief, hatred, or even fear. The discourse of pride is full of language that describes it as impossible to bear, intolerable.40 In terms of action, it gives rise to violence and other types of revenge. A good general account of what makes pride ridiculous is provided by Cicero in a defense speech for Aulus Cluentius, who in 66 BCE was accused of poisoning his stepfather. One of the characters that Cicero brings into this lively speech is Lucius Quinctius, a man who, when tribune of the plebs in the year 74 BCE, defended the stepfather against Cluentius in another case, lost, and argued that judicial bribery was responsible for the defeat. Cicero wants to impugn Quinctius’ credibility and does so in a number of ways: his lack of judicial experience, his dress, and his arrogant facial expression are all singled out as excessive and ludicrous. He is clearly portrayed as trying to act in a way that is not justified by an external assessment of his status and qualifications. The most relevant passage for our purposes comes at the end of this opening character attack:41 So Quinctius ... if he had been a member of the nobility, who would have been able to endure him, with his pride and intolerance? But because he did come from that low place, they did put up with him in such a way that they thought that, if he had something naturally good in him, that quality ought to be able to be of use to him, and, on the other hand, judged that his pride and arrogance were more to be mocked, on account of his lowly birth, than feared.

The new man Cicero channels the nobility’s contempt for people like himself to score a rhetorical point.42 But the assessment he offers here is very useful for clearly presenting how the size of the gap in the hierarchy affects the evaluation of proud behavior by those in a superior position. Had Quinctius been an equal, or close, his demeanor would have caused offense, and even fear, but given his low starting point, to take offense would mean taking him more seriously as a threat than he deserved. Thus, the appropriate reaction to a display of pride from a man like Quinctius is ridicule. Moreover, ridicule is not simply what Quinctius can expect from the nobles, but Cicero, even as he describes such mockery as apt, also performs it in his attack on Quinctius, using his position as an advocate to assimilate himself rhetorically to the viewpoint of the nobles.43 His preceding description of Quinctius dresses him down so that the conclusion of the passage – that we, the audience, are to see Quinctius’ behavior as proud, dismiss and mock him – comes as no surprise.

40 41

42 43

Baraz 2008, 379. Cicero, Pro Cluentio 112: ut Quinctius ... si fuisset homo nobilis, quis eum cum illa superbia atque intolerantia ferre potuisset? quod eo loco fuit, ita tulerunt ut, si quid haberet a natura boni, prodesse ei putarent oportere, superbiam autem atque arrogantiam eius deridendam magis arbitrarentur propter humilitatem hominis quam pertimescendam. On Cicero’s novitas, see Dugan 2005 and Blom 2010. On Cicero’s rhetoric in the speech, see Kirby 1990, 47f. (on Quinctius).

230

Yelena Baraz

At the other end of the emotional spectrum, Cicero’s letters contain an example of a detailed description of reaction to pride as injury to one’s dignity:44 I hate the queen. Ammonius, the guarantor of her promises, knows that I am justified in this since they were learned and so worthy of my dignity that I would dare mention them even in a public meeting. Furthermore, Sara, besides the fact that he is a dreadful fellow, I now know is insulting towards me. I saw him at my house only once. When I asked him, in a friendly manner, what he needed, he said he was looking for Atticus. But the pride of the queen herself, when she was living in the gardens across the Tiber, I cannot recall without great grief. So, nothing more to do with them; they think that I lack not so much spirit as an ability to get irritated.

Cicero is finding his treatment at the hands of Cleopatra herself, and her representative Sara, painful and difficult to bear. In the case of the queen, we do not know enough about the promises to which Cicero is referring to give a precise account of what finally became unendurable for Cicero, but the general tenor of the letter is clear: her behavior has caused him dolor, a painful feeling of grief resulting from injured dignity. The conduct of her subordinate, who insults Cicero in his own home by refusing to accord him and the members of his household any importance, becomes part of the overall picture of the queen’s attitude towards him (the incident with Sara is framed by references to the queen and the two are grouped together at the end).45 Cicero, a senior consular, is ignored by the queen, who is not fulfilling her obligations (whatever they may have been, they are contractual, as they involve a sponsor), and treated as a doorman by Sara. Dolor is the main feeling that the whole interaction evokes, but the attendant emotions are hatred and irritation or frustration. Despite his stature, Cicero is in no position to reassert his dignity vis-à-vis the queen and is therefore left permanently slighted by her pride, reliving the injury each time he remembers these events. It is precisely the inability to respond that makes the effects of the Egyptians’ pride fester. A very different emotional approach to what constitutes an appropriate reaction to pride is found when revenge is a possibility. An interesting example is presented by the pseudo-Ovidian elegy addressed to the empress Livia on the occasion of the death of her younger son, Drusus, who died on campaign in Germany

44

45

Cicero, Epistulae ad Atticum 15.15.3 = SB 393 (June 44 BCE): reginam odi. id me iure facere scit sponsor promissorum eius Ammonius, quae quidem erant φιλόλογα et dignitatis meae ut vel in contione dicere auderem. Saran autem, praeterquam quod nefarium hominem, cognovi praeterea in me contumacem. semel eum omnino domi meae vidi. cum φιλοφρόνως ex eo quaererem quid opus esset, Atticum se dixit quaerere. superbiam autem ipsius reginae, cum esset trans Tiberim in hortis, commemorare sine magno dolore non possum. nihil igitur cum istis; nec tam animum me quam stomachum habere arbitrantur. Cicero, who spent his entire life trying to establish and defend his standing, is generally quite sensitive to proud behavior, even from undoubted inferiors. Witness his correspondence with Atticus on the subject of the latter’s freedman Dionysius, a teacher of Cicero’s children, who declined to follow him abroad in 49 BCE (Epistulae ad Atticum 8.4.1–2 = SB 156 and 8.10 = SB 159; 8.4.1: superbum se praebuit).

Modeling Roman Pride

231

after falling from a horse.46 Despite the accidental nature of his death, the unknown poet fits it into the greater pattern of the conflict between Rome and the German tribes against whom Drusus successfully campaigned for a number of years.47 The death of a successful and popular Roman general is portrayed as giving rise to pride in his German adversaries, and the poet responds by imagining the shape that revenge, a proper response to this pride, will take:48 But to you, Germany, there remains no just claim on forgiveness. Soon enough will you, barbarian, pay the penalty of death. I will see the necks of your kings black and blue from chains, and harsh fetters woven around their savage hands, and, finally, their faces full of terror, and tears streaming down the unwilling cheeks onto those bestial faces. That threatening spirit, made proud by Drusus’ death, will have to be handed to the executioner in a gloomy dungeon.

The Germans have to be punished because Drusus’s death could be seen as a defeat and could arouse in them – as the Roman poet’s channeling of the loss suggests to him – a feeling of pride. Such an emotion would be unjustified, from the Roman point of view, not just because Drusus’ death did not occur in a military context but because of the Romans’ investment in seeing the Germans as inferior, barbarians who would soon come begging for mercy. The lines cited also show that such an attitude towards the Germans is not easy to achieve: they are ferocious and hard, normally fearless, and, ultimately, threatening. It is this ambivalence that fuels the desire for revenge in this text and produces the gleeful forecast of the triumphal procession after Germany’s defeat. Death has to be repaid with death (272 and 277: morte), but that is not enough: Rome has to reassert its superiority. Thus the emphasis on finally instilling fear in the kings who will be humiliated by being paraded in a triumph, the desire for tears inflicted by breaking their will, and the delight in their physical suffering and humiliation resulting from the surroundings in which they are to die. The rhetorical pitch of the poem and the indignation and desire for revenge sparked by the very possibility that the enemy might feel pride is particularly striking since the event to which it responds cannot compare with the humiliating military defeats such as Crassus’ loss to the 46 47 48

The text is known under several titles: Ad Liviam de Morte Drusi, Epicedium Drusi, and Consolatio ad Liviam. See Schoonhoven’s 1992 edition with commentary. Cf. line 170 with Schoonhoven 1992 ad loc., 129. Epicedium Drusi 271–278: at tibi ius veniae superest, Germania, nullum. postmodo tu poenas, barbare, morte dabis. adspiciam regum liventia colla catenis, duraque per saevas vincula nexa manus: et tandem trepidos vultus: inque illa ferocum invitis lacrymas decidere ora genis. spiritus ille minax et Drusi morte superbus, carnifici in maesto carcere dandus erit.

232

Yelena Baraz

Parthians at Carrhae in 53 BCE and the Varian disaster in the Teutoburg Forest in 9 CE.49 Having begun this investigation with a fable, I will end with a myth that brings an additional dimension into the economy of pride by introducing the relationship between men and gods. Perhaps the best-known example of pride severely suppressed is the story of Niobe, alluded to by Ovid in the Heroides and by Propertius.50 Although as a human Niobe is obviously inferior to the gods, her pride, caused by her ancestry and her fertility, finds in Latona a vulnerable member of the divine cohort. The two are analogous in that they derive their sense of self-worth in large part from their progeny, and this fact allows Niobe to establish an axis on which she is superior to the goddess. The revenge of Latona’s divine offspring specifically targets and destroys the source of her pride by killing all of her children. The violent reaction not only deprives Niobe of her status as a mother but also, in leading to her transformation into a stone, robs her of her humanity, thus lowering even further her position relative to Latona. The mythical status of this tale and the excessive nature of the punishment make Niobe a useful exemplum of extreme pride met with extreme punishment. The one extended narrative of the myth, that in the sixth book of Ovid’s Metamorphoses, fills in and amplifies the basic features of the story to show the excess of Niobe’s pride.51 Ovid’s Niobe first tries to reverse the basic contrast between the human and the divine in her favor, by suggesting that those present (seen) should be worshiped above the absent (heard).52 She then boasts at length about her genealogy and proceeds to attack Latona on the basis of her history, using her exclusion from all land when about to give birth to her twins to label her an exile.53 Her final boasts, as presented by Ovid, do more than compare herself to the goddess in terms of number of children, even using that number to elevate herself to quasi-divine position:54 I am lucky (for who will deny this?) and I will remain lucky (this too who will doubt?): my fertility has made me safe. I am too great for Fortune to be able to do me harm: though she may take away many things, she will leave me many more. The good things I possess have already advanced beyond fear.

49 50 51 52 53 54

On the battle of Carrhae, see most recently Traina 2010; on the Varian disaster, Wolters 2008. Ovid, Heroides 20.105; see also Propertius 2.20.7. On the episode, see most recently Feldherr 2010, 295–313; for an overview, Rosati 2004, 271–296. Ovid, Metamorphoses 6.171f.: ‘quis furor auditos’ inquit ‘praeponere visis/caelestes?’ Ovid, Metamorphoses 6.189: exul mundi. Ovid, Metamorphoses 6.193–197: sum felix (quis enim neget hoc?) felixque manebo (hoc quoque quis dubitet?): tutam me copia fecit. maior sum quam cui possit Fortuna nocere, multaque ut eripiat, multo mihi plura relinquet. excessere metum mea iam bona.

Modeling Roman Pride

233

In assuming that her future is secure and Fortune has no power over her, Niobe lays claim to a stability greater even than that of the gods, who are immortal but still subject to changes of fortune. If, in a sense, Fortuna stands above the gods, Niobe has created a competing axis on which her superiority to Fortune has elevated her high above their level. Her pride also has the effect of banishing an emotion essential to human beings when contemplating the divine: fear. The end of the speech brings this process to its logical conclusion when Niobe stops the sacrifices to Latona, her pride no longer limited to just boasting but finding expression in attacking the goddess’ prerogatives. The goddess’ reaction to this assault is indignation. In her complaint to her children, the emotion that she describes experiencing is the same that Cicero felt at being slighted by Cleopatra: dolor, the pain of injured dignity.55 She also makes it clear to Apollo and Artemis that their dignity is damaged along with hers. The twin gods’ killing of Niobe’s sons has the effect of directing a dolor of a different sort back to Niobe and her family (267: populi dolor; 280: nostro dolore, as Niobe prays to Latona and accuses her of cruelty). But Niobe still sees the encounter as a competition, first yielding to her rival in triumphal language, then immediately reappropriating that language and claiming victory based on the number of children who still remain.56 The continuation of the gods’ revenge, the killing of her daughters, has the effect of finally reversing Niobe’s initial boast: having claimed to be greater than Fortune (195: maior), she now begs for her youngest child’s life (299 and 300: minimam), and having mocked the goddess as being virtually childless, she is left childless herself (200 and 301: orba). With all of Niobe’s claims to superiority destroyed, and following her transformation into a weeping rock, Latona’s revenge is complete. 6 CONCLUSION Examples drawn from a variety of texts, different in genre and time period, have demonstrated that there is a stable core to the Romans’ conception of how the social emotion of pride comes into being, what it consists in, what behaviors it can result in on the part of the person or group experiencing the emotion, and what reactions those who become the object of such behaviors are likely to have. The key feature of pride in Roman discourse is a self-evaluation that is found to be inaccurate, to a lesser or greater extent, when measured against some external criterion. Any type of good fortune can lead to the rise of the emotion. Once the 55 56

The connections between representations of Niobe and Cleopatra in Augustan art are discussed by Schmitzer 1990, 244–249 and Feldherr 2010, 298f. Ovid, Metamorphoses 6.283–285: exsulta victrixque inimica triumpha! cur autem victrix? miserae mihi plura supersunt, quam tibi felici; post tot quoque funera vinco!

234

Yelena Baraz

emotion is felt, it is believed to lead to a variety of excessive behaviors from mockery to violence, as the subject attempts to assert his new estimate of his standing, position, character, etc. The reactions to pride are the most varied, as they depend on the relative position of the person reacting, his ability to exact revenge of some kind, and the magnitude of the particular mistaken self-estimation that led to the proud behavior. Despite the variety, the concept of pride is sufficiently unified in Roman discourse to be briefly evoked and for its constituent features to be easily recognized. BIBLIOGRAPHY Astin, A. E. (1978) Cato the Censor, Oxford. Baraz, Y. (2008) From Vice to Virtue: the Denigration and Rehabilitation of Superbia in Ancient Rome, in R. M. Rosen and I. Sluiter (eds.), KAKOS: Badness in Classical Antiquity, Leiden, 365–397. Barchiesi, A. and G. Rosati (2005) Ovidio: Metamorfosi. Volume II (Libri III–IV), Milan. Blom, H. van der (2010) Cicero’s Role Models: The Political Strategy of a Newcomer, Oxford. Calboli, G. (2003) Oratio pro Rodiensibus: Catone, l’oriente greco e gli imprenditori romani, Bologna. Champlin, E. J. (2005) Phaedrus the Fabulous, Journal of Roman Studies 95, 97–123. Courtney, E. (1999) Archaic Latin Prose, Atlanta. Cugusi, P. and M. T. Sblendorio Cugusi (2001) Opere di Marcio Porcio Catone Censore. Volumes 1–2, Turin. Currie, H. MacL. (1984) Phaedrus the Fabulist, in Aufstieg und Niedergang der römischen Welt II.32.1, Berlin, 497–513. Dugan, J. (2005) Making a New Man: Ciceronian Self-Fashioning in the Rhetorical Works, Oxford. Fantham, E. (2004) The Roman World of Cicero’s De Oratore, Oxford. Feldherr, A. (2010) Playing Gods: Ovid’s Metamorphoses and the Politics of Fiction, Princeton. Gruen, E. S. (1986) The Hellenistic World and the Coming of Rome, Berkeley. Hardie, P. (1994) Virgil: Aeneid, Book IX, Cambridge. Henderson, J. (2001) Telling Tales on Caesar: Roman Stories from Phaedrus, Oxford. Kirby, J. T. (1990) The Rhetoric of Cicero’s Pro Cluentio, Amsterdam. Kurke, L. (2010) Aesopic Conversations: Popular Tradition, Cultural Dialogue, and the Invention of Greek Prose, Princeton. Meyer, E. (2004) Legitimacy and Law in the Roman World: Tabulae in Roman Belief and Practice, Cambridge. Morstein-Marx, R. (2004) Mass Oratory and Political Power in Late Roman Republic, Cambridge. Nagy, G. (1979) The Best of the Achaeans: Concepts of the Hero in Archaic Greek Poetry, Baltimore. Nicolet, C. (1980) The World of the Citizen in Republican Rome, Berkeley. Oberg, E. (2000) Phaedrus-Kommentar, Stuttgart. Rosati, G. (2004) Ovidio: Metamorfosi, Volume III (Libri V–VI), Milan. Rosen, R. M. (2007) Making Mockery: The Poetics of Ancient Satire, Oxford. Schmitzer, U. (1990) Zeitgeschichte in Ovids Metamorphosen: Mythologische Dichtung unter politischem Anspruch, Stuttgart. Schoonhoven, H. (1992) The Pseudo-Ovidian Ad Liviam de Morte Drusi (Consolatio ad Liviam, Epicedium Drusi), Groningen. Sherwin-White, A. N. (1984) Roman Foreign Policy in the East, 168 BC to AD 1, London.

Modeling Roman Pride

235

Traina, G. (2010) La resa di Roma: 9 giugno 53 a.C., battaglia a Carre, Rome. Wilcox, A. (2008) Caligula as exemplum in Seneca’s Dialogues, in R. M. Rosen and I. Sluiter (eds.), KAKOS: Badness in Classical Antiquity, Leiden, 451–473. Winterling, A. (2011) Caligula: A Biography, Berkeley. Wolters, R. (2008) Die Schlacht im Teutoburger Wald: Arminius, Varus und das römische Germanien, Munich.

GRIEF AND MOURNING IN THE ROMAN CONTEXT The Changing Sphere of Female Lamentation Katariina Mustakallio 1 GRIEF IN HISTORICAL AND SOCIAL SETTINGS Emotions are a difficult subject in historical research. They are elusive and ambiguous. Private emotions stay in the private sphere but emotions shown in public may have an impact on historical developments. Cultural forms or canons, or, as Barbara Koziak has put it, paradigm scenarios for individual emotions,1 direct the way we feel or show our emotions. Feeling happy or sad might appear instinctive, but the way we show or represent our feelings in public depends on the culture in which we are living; on the traditions of our family, gender, social class, ethnic origins etc.; and on our own personality. Occasions when we should show joy or sorrow are culturally determined. It is not surprising that emotional control became a matter of focal interest even early in Classical philosophy and a public matter in ancient city states. The classical theory of emotions often connected uncontrolled feelings such as anger and fear with femininity, and these were seen as dangerous to the controlled civic life of the polis. Stoic philosophers were especially interested in developing self-control and fortitude as a means of overcoming destructive emotions; for the Stoics, passions and emotions were connected to the irrational.2 In Rome the discourse was presented in a different manner, as Francesca Prescendi has emphasised. According to Pliny, mourning (luctus) is one of the qualities that separate human beings and animals: animals do not feel sorrow.3 On the other hand, for Cicero mourning belongs to the emotions of confusion or disturbance (perturbationes).4 According to modern anthropologists, ritual mourning consists of a set of conventional behavioural responses to crisis or death that are both required and controlled by society. Ritual mourning is not a spontaneous outburst of feelings

1 2 3 4

See, e.g., Koziak 2000, preface. For emotions and Stoic philosophy in general, see Sihvola and Engberg-Pedersen 1998, and in particular Brennan 1998, 21–70; for attitudes towards women, see Harris 2002, 264–282. Prescendi 2008, 297–313. Pliny, Naturalis historia 7.4–5: uni animantium luctus est datus. On the social capacity of human beings and animals, see Sihvola 2010, 14f. Cicero, Tusculanae disputationes 4.16.

238

Katariina Mustakallio

but often a well-organized social practice.5 In many societies and cultures, ritual lamentation is considered to belong to the feminine sphere. Women have been considered more suitable to take care of the corpse and mourning, as Margaret Alexiou pointed out in her classic study.6 Owing to their ability to give birth, women have generally been considered to be more familiar with birth and death. In Rome the goddess of death, Libitina, was also the ruler of birth.7 We find the same kind of double character in the cult of Ceres, the goddess of fertility, to whom certain sacrifices were offered by a family after the death of one of its members.8 The connection of women with the world of death has been interpreted as a sign of women’s low and marginal status in patriarchal Mediterranean societies.9 Was this the case in Rome as well? Public mourning was a strictly controlled act in traditional pre-modern societies. Nevertheless, at funerals it was a social obligation for family members to cry in public. Ann Suter has argued that: ritual lament was a prominent activity for women in ancient societies. It seems to have been the one medium in which they might have expressed themselves and their concerns publicly 10 and thereby have influenced a community’s affairs.

On the other hand women in general did not form a single interest group since there was no common female interest. In many cases we also have other groups acting collectively at funerals, and very often women and children together, so lamentation was not exclusively a female affair. Even if the discussion concerning the classical theory of emotions has strongly influenced the ways feelings are represented in the ancient literary tradition, the main focus of this article is not on the philosophical theories of emotion. The main concerns here are how public sorrow and ritual mourning were represented, and especially lamentation in public. These subjects are approached from their social and public side, paying special attention to status, gender, and agency. The essay is divided into four sections: 1) women in Roman funerary ceremonies; 2) social control over funerals and mourning; 3) female groups in historical memory; and 4) conclusions.

5 6 7 8 9 10

See the discussion in Mustakallio 2003, 86–99. On emotional communities in the context of Greek religion, see Chaniotis 2011. Alexiou 1974. Plutarch, Quaestiones Romanae 23. The sacrifice of Porca praecidanea: see Cicero, De legibus 2.22.57; Varro, De vita populi Romani 3. fr. 104, ed. Riposati; Le Bonniec 1958, 27. There is plenty of discussion of this topic, e.g., Thomas 1992, 85–137; Zaidman 1992, 368– 372; Scheid 1992, 377–408. See Suter 2008, introduction.

Grief and Mourning in the Roman Context

239

2 WOMEN IN ROMAN FUNERARY CEREMONIES According to the Law of the Twelve Tables from c. 450 BCE, the Urbs was the space in which it was prohibited to cremate or bury the dead.11 Two of the main issues in funeral rites were the fear of defilement caused by the death itself and the horror of the dreadful forces of the underworld. By celebrating the funeral rituals properly the door between everyday life and the dangerous world of Death was closed, and the possible defilement of the whole city was prevented. Under these circumstances the interdiction of funerals held inside the city-walls was taken extremely seriously.12 When death was imminent, family and friends gathered together around the dying person to hear his or her last words.13 After the last kiss the family called the deceased person’s name three times (conclamatio).14 At this moment the women’s ritual mourning began, and it was repeated until the last purification rites after the burial were completed. All who came into contact with the dead, and especially the deceased’s family members , were considered to be defiled by death (familia funesta), and before resuming normal life and contact with others they needed to be purified of that pollution after the funeral.15 Some Roman writers analysed the degrees of the sorrow more profoundly and described how, after the initial shock, the process of grief developed from the dolor sine voce, pain without a voice, to the second phase, the conclamatio of the deceased’s name, and then to the planctus, the acts of the mother or female members of the family first embracing the corpse and then crying with their hair loosened.16 Traditionally, women of the deceased’s family played a central role in all the activities related to the corpse at home. They washed the body and dressed him or her according to his or her social status. We know there were also professional undertakers, libitinarii and pollinctores, who took care of the arrangements for the funerals.17 There is an interesting marble relief from the tomb of the family of the Haterii which shows a scene of mourning at the house of the deceased with the lectus funebris in the middle, and mourning women and flute-players around it (Figure

11 12

13 14

15 16 17

Lex XII Tabularum X fr. 1: hominem mortuum ... in urbe ne sepelito neue urito. See Maurin 1984, 207; according to Ovid, neglected bodies could haunt and harm future generations: see Ovid, Fasti 5.419. See Audibert 1885, 13f. and Cumont 1945, 352–354. On defilement, see e.g. Parker 1983, 32. See e.g. Suetonius, Augustus 98f.; Cicero, De divinatione 1.30.64. See Toynbee 1971, 44. Servius, Commentarius in Vergilii Aeneidos 6.218; Lucanus 2.21–23. Cf. Pliny, Naturalis historia 7.52. The last kiss was of central importance in Roman death culture; see e.g. Seneca, De consolatione ad Marciam 3.2; Cicero, Orationes Verrinae 5.118; Statius, Silvae 5.1.207. On familia funesta and familia pura, see Varro, De Lingua Latina 5.4; Varro, De vita populi Romani 3. fr. 104, ed. Riposati. See e.g. Lucan, Pharsalia 2.20–28; Prescendi 2008, 297–313. For female mourners, see e.g. Mustakallio 2003, 86–99. For undertakers, libitinari (and pollinctores) named after the goddess of death, Libitina, see Seneca, De beneficiis 6.38; Martial, Epigrammata 10.97; Plutarch, Quaestiones Romanae 23; Numa 12.1f.

240

Katariina Mustakallio

1).18 The relief represents the lectus funebris on which lies the dressed corpse of a woman. Two women mourners (praeficae) stand behind and by their side a man, perhaps in the act of putting a garland on the head of the corpse. On each side of the lectus funebris is a torch. On the left side is a woman playing the flute and above is another with folded hands; on the right side sit three women wearing the pilleus (probably manumitted slaves); below is the family of the deceased.

Figure 1. The lectus funebris on a relief from the tomb of the Haterii in Rome (late first century CE). Vatican Museum.

At public funerals the funeral procession was of central importance; it was a performance with several actors. In the centre was the deceased represented by his or her effigy. Around the dead individual there were carriers, relatives, and friends.

18

See, e.g., Toynbee 1971, fig. 9 from the Tomb of the Haterii depicting a lying-in at a house, with mourning women around the deceased, late Flavian or early Trajanic in date according to Toynbee 1971, 44; and fig. 10, a marble relief found in Paris, showing the body of a girl lying on a couch and mourners around her.

Grief and Mourning in the Roman Context

241

There were people lamenting who acted like a chorus in classical drama. 19 Important evidence of the public side of funerals is provided by another Roman marble relief of Amiternum from the Aquila Museum (Figure 2), from the LateRepublican or Augustan period, depicting the funeral procession of a rich but not aristocratic family.20 Here the deceased is shown almost in the centre of the panel. There are eight bearers carrying the funeral bier, with the effigy of the deceased lying on his couch. During the funerals of wealthy Romans the funerary bier was carried on the shoulders of the deceased’s male relatives or hired bearers. Later in imperial funerals it became a sign of honour to carry the bier. According to Tacitus, in the funeral procession of Augustus the bearers were senators.21

Figure 2. Representation of a funerary procession on a grave relief from Amiternum (late first century BCE/early first century CE). National Museum of Abruzzo, L’Aquila.

There were special professionals who took care of orchestrating the whole pompa, called dissignatores. Harriet Flower has identified as a dissignator the figure facing the bier as he directs the other bearers. In this panel the procession is represented on two main levels. The musicians and the professional mourners, praeficae, walk ahead of the bier on the upper level, hands raised and with dishevelled hair, in a typical gesture of mourners.22 The figures identified by Flower as the family, perhaps the widow with two daughters, follow on the other side of the bier. According to Flower, a funeral’s grandeur could be measured in terms of how many walked before the body. 23 Funeral music in Rome had a strong emotional effect on the audience. Musical performance consisted of flutes and horns played by male musicians and funeral songs presented usually by women and children. Interestingly enough, Plutarch discusses the psychological effect of funeral music and says that the main 19 20 21 22 23

Bianchi Bandinelli 1981, 59, fig. 60, with a group of mourning women with their hands raised and folded together and their hair dishevelled, two in front of the bier and eight behind it. Toynbee 1971, fig. 11 and p. 46. See, e.g., Tacitus, Annales 1.8. Prescendi 2008, 313. Flower 1996.

242

Katariina Mustakallio

purpose was to create a suitable atmosphere for the occasion and to help the mourning process to start. It was a common belief that the malicious powers of death were frightened away by funeral music.24 According to M. Terentius Varro, at funerals female mourners made up the chorus, lamenting according to the conducting of the praefica, who was an older woman. They performed with dishevelled hair and hands raised to beat their breasts. The praefica had to have a remarkable voice to sing the nenia in honour of the dead.25 in that phase he [the deceased] would be praised by the woman who had the best voice; thereafter the usual dirge would be chanted, accompanied by the pipes/flutes and stringed instruments of those who were usually responsible for accompanying the buffoonery at the public games. In the old days, this woman was called a praefica, a ‘praise-leader’, right up to the Punic War.

Citing Ennius, a Roman poet from the second century BCE, Varro points out that the older woman was put in charge of the maidens to show them how they should perform their lamentations.26 They were trained to act in public in a proper way. The act of mourning was a formal practice with its own regulations, not an improvised show of sentiment.27 Public mourning showed a Roman citizen’s social status and worth after his or her death. The function of a praefica included not only starting the mourning process and conducting the procession; Dorota Dutch sees in their acts even more dimensions and emphasises their role in performing the nenia, the commemorative song sung in front of the deceased’s house. Praeficae were also needed to close the whole funeral procession.28 As Servius says in his comment on the Aeneid:29 Our ancestors called the women concerned with the funeral, such as a mother or sister, ‘funerea’: for praeficae ... stand out in mourning but not in pain.

So we have two different groups of women in funeral processions, professional mourners and family women, both showing grief in different ways. According to Roman historical tradition, the luctus matronarum, the public crying of the Ro-

24 25

26 27 28 29

31

Plutarch, Moralia 657A [Quaestiones conviviales]; for mourning in a literary context, see e.g. Marcus 2004, 105–135. Varro, De vita populi Romani 3, fr. 110 ed. Riposati: ibi a muliere, quae optuma uoce esset, perquam laudari; dein neniam cantari solitam ad tibias et fides eorum qui ludis tricas curitassent. haec mulier uocitata olim praefica usque ad Poenicum bellum. For the role of the praefica in the earlier tradition, see also Varro, De Lingua Latina 7.70; for mourners, see Horace, Ars poetica 431–433; Lucilius 27.18; Adrados 1999, Appendix I; see especially fables transmitted by the Anonymous collections, H221, p. 283. Varro, De lingua Latina 7.70. For the mourning rituals in different cultures, see Huntington and Metcalf 1979, 27f. Dutch 2008, 260f. Servius, Commentarius in Vergilii Aeneidos 9.484: nec tua funera mater id est ‘funerea’: nam apud maiores funeras dicebant eas ad quas funus pertinet, ut sororem, matrem: nam praeficae, ut et supra diximus, sunt planctus principes, non doloris. See, e.g., Mustakallio 2003, 83–99.

Grief and Mourning in the Roman Context

243

man matrons with the deceased’s female family members, was an important phenomenon in every public funeral during the Republican period.31 The proper time for luctus in Rome depended on the relationship between the mourner and the deceased person, as well as on the age of the deceased. According to Plutarch, the normal period for mourning was ten months.32 From the literary evidence we learn that, in the case of the death of an important statesman, the matrons showed their special respect by mourning him in public for a year. This happened, according to Livy, especially when defenders of the Republic and female dignity, like Iunius Brutus and Publius Valerius Publicola, were concerned.33 This kind of extended mourning differed from the mourning at funerals; it was characterised by distinctive external appearances in dress and hair-style, and it also limited the ability of matrons to participate in public religious festivals. Furthermore, the matrons could also show their opinion and display their disrespect by not mourning someone. In some cases, the denial of a funeral and mourning was even added to certain death sentences by the Senate.34 This shows that the social value of mourning was strongly associated with the status of the deceased, of showing the worth of his or her life. The ritual of mourning and crying was also believed to placate the souls of the dead and the underworld.35 The denial of mourning was part of the prohibition of burial rites that left the deceased without proper rituals. According to Ovid, spirits of those buried without proper burial rites haunted and harmed people, especially future generations of their family.36 3 SOCIAL CONTROL OVER FUNERALS AND MOURNING Fragments of the Law of the Twelve Tables from c. 450 BCE offer good examples of the early Roman state’s efforts to control funeral practices. We know that in fifth-century Greece there were comparable restrictions in several city-states as well.37 The laws consisted in detailed restrictions on behaviour, the proper way of dressing, and the number of flute-players in the funerals. The luxury allowed in the funeral banquets was severely restricted.38 Special restrictions concerned the appropriate way of showing grief during the funerals too. Once again, the behaviour of women was mentioned in detail: women were not allowed to tear at their

32 33 34 35 36 37 38

Plutarch, Numa 12.1f. Livy 2.7.4; 2.16.8. Mustakallio 1994, especially 11; for the later period see Vittinghoff 1936, 65; for special cases, see Flower 1998, 155–185. For the placatory offerings given to the displeased souls of dead, see Cicero, De legibus 2.22f.; Festus 83L; Ovid, Fasti 2.615 and 5.419. Ovid, Fasti 5.419. See, e.g., Engels 1998; Frisone 2000. Lex XII Tabularum 10.3.

244

Katariina Mustakallio

cheeks or even wail loudly during the funerals.39 According to Dorota Dutch, selfmutilation by mourning women was carried out in the context of the idea of nourishing the dead with their blood and milk.40 Restrictions concerning female lamentation are mentioned even later.41 The governments of the city-states attempted to control chaotic events and anarchic behaviour in general, not only during funerals. Uncurbed public mourning would have threatened the harmonious life of the city and encouraged shows of emotions in general.42 Furthermore, during the Punic Wars new laws related to funeral practices were issued by the Senate. The reason was obvious: certain important religious festivities considered essential for the well-being of the Roman people were in danger of being neglected for the sake of the long mourning period. Livy points out how dangerous it was that women were mourning in public. They disturbed everyday life and caused trouble, he argues. Some men in the Senate, Q. Fabius Maximus for example, demanded that families should keep women in their houses by force.43 It seems that the demands of the Senate were not heard. After the disastrous battle of Cannae in 216 BCE the laments of women were so overwhelming that even religious rites were neglected. According to Livy, ‘the entire city was so filled with lamentation that the annual rite of Ceres was allowed to lapse’.44 The reason was that mourning women could not participate in any public religious rites because they were ritually polluted by the death.45 Livy tells us that in the same year the Senate decreed that the mourning period should be limited to thirty days.46 This decree has been seen by modern scholars as evidence of the repression of Roman women by the Senate.47 In any case it shows the significance of women for Roman public religion and society. The rites of Ceres were closely connected to women, and they were pivotal for the well-being of the whole city: even the traditional mourning of women was interrupted to celebrate the cult. During the late Republic Roman funeral practices, especially those of upper class families, were strongly influenced by Hellenistic funeral festivals, as Harriet 39

40 41 42

43 44 45 46 47

Lex XII Tabularum 4: Mulieres genas ne radunto, neve lessum funeris ergo habento; Lex XII Tabularum 10 (= Cicero, De legibus 2.22f.); Toynbee 1982, 43. For comparable restrictions in Greece, see Parker 1983, 36–40. Dutch 2008, 258. Livy 22.56.4f.; 26.79.7f. For control of funerals, see e.g. Loraux 1991, 20–26; Humphreys 1981, 267; Stears 1998, 113–127. In this context women who were tearing their cheeks and breasts and crying aloud in public were not considered simply as producers of pity but sometimes even as terrifying; see Foley 1993, 143. Livy 22.7.7, 12–14; 22.55.3 (Q. Fabius Maximus’ speech in 216 BCE). Livy 22.56.4. For the pollution caused by death (funus, miasma), see e.g., Parker 1983, 32. Livy 22.56.4f: Itaque ne ob eandem causam alia quoque sacra publica aut privata desererentur, senatus consulto diebus triginta luctus est finitus; cf. 26.79.7f. It has been seen as an act against plebeian women because the cult of Ceres was especially a plebeian cult; see, e.g., Bauman 1992, 24.

Grief and Mourning in the Roman Context

245

Flower has shown in her study.48 Roman families of noble origin competed with each other in the luxury and splendour of their funeral festivals and games. One of the great changes in ‘Female Funeral Culture’ was the change in the nature of mourning. Traditionally, an important part of public funeral rituals was played by respected matrons showing grief and mourning (luctus matronarum). In Imperial Rome we do not find mention of the mourning of matrons in public. There were other changes as well. During the Imperial period hymns or neniae were performed mostly by youths and virgins at public funerals.49 This was not a feature peculiar only to funerals. In other great ceremonies of the state, for instance, in the Ludi Saeculares, the younger generation, boys and girls, also had an important role in public rituals. In the purification process for the whole society the presence of youth was highly necessary.50 Funeral processions were a long-lasting phenomenon in Roman funeral ceremonies. In depicting the funeral of the emperor Vespasian, Suetonius mentioned that the participants could even make jokes during the procession. In this case, the actor who carried the image of the late Vespasian joked about the stinginess of the former emperor. Tragedy and comedy, mourning and laughter, were present in imperial funerals, as Maurizio Bettini has pointed out. In the multicultural world of the Roman Empire new funeral habits and traditions were created, and new ethnic groups became visible.51 4 FEMALE GROUPS IN HISTORICAL MEMORY In Roman historical memory relating to Early and Republican history, women are often described as taking part in public matters by demonstrations of their emotions. In many instances we are told of groups of women in the forum or other public places. In these stories women formed groups especially during times of crisis and periods of pestilence and war.52 In these situations women gathered together and appealed to men and the gods by lamenting and crying.53 In Roman historical memory the first of these stories is that of the Sabine women.54 According to the tradition, after the rape of the Sabine women, king Romulus reassured the women that they would be properly married and honoura48 49 50 51 52

53 54

Flower 1996, 122–127. Neniae were connected in the Roman tradition with hired praeficae; later the habit seems to have changed, see note 25. Cf. Suetonius, Augustus 99.2; Dutch 2008. See Horace, Carmen saeculare 6.36, where boys and girls are mentioned as symbolic representatives of the future of Rome, and Hänninen 2000, 106. Tacitus, Annales 1.84.5. See Bettini 2005, 196. Women acting in public during internal crises: Livy 2.40.2; 2.40.9; 3.47.4; 3.48.8. During war, both Roman women and foreign: Livy 1.29.5; 2.33.8; 4.40.3; 5.21.11; 5.40.3; 5.42.4; 6.3.4; 22.7.7, 12–14; 22.55.3; 22.55.6f.; 22.56.4f.; 26.9.7f.; 38.22.8. See also Gagé 1963, 104f. See Hänninen 2000, 71–93; Mustakallio 2003. For the story of the Sabine women, see Livy 1.11–13; Dionysius of Halicarnassus 2.38–45; Plutarch, Romulus 14–19.

246

Katariina Mustakallio

bly treated as matrons. After that the women decided to do their best to stop the war between their Sabine fathers and their Roman husbands and, according to Livy, they rushed between the two armies appealing to them. They are described as carrying their children in their arms and crying, their hair dishevelled55 – in the same manner as women during funerals. In the accounts of Dionysius and other authors, this dramatic occasion is enhanced by the appearance of the powerful figure of Hersilia.56 As an older woman she is the leader of the group and the spokeswoman of the Sabine wives. Livy does not mention Hersilia, but emphasises the importance of the Sabine women by reporting that Romulus was very pleased with their actions.57 Another story with a group of lamenting women acting in public is the story of Coriolanus, a famous Roman general who went into voluntary exile, made a pact with the Volscians, and finally attacked Rome, the town where he was born. Roman senators and priests tried to make peace with him; then a committee of matrons went to his camp carrying their children, lamenting and crying aloud – once again with dishevelled hair as in the funerals. At the head of the procession walked three important women – Veturia, Coriolanus’ mother, Volumnia, his wife, and Valeria, the sister of Valerius Publicola – carrying their children, and also lamenting and crying aloud. Livy speaks about an ingens mulierum agmen (‘huge crowd of women’) in this connection.58 In Dionysius’ version the relationship between the women is clear. Valeria is here the promoter of the women’s mission. The story clearly reveals the relationship and ties between the matrons: their mutual loyalty was based on their participation in the same cult.59 Even if the women on this occasion were lamenting and crying, the old mother of Coriolanus, Veturia, was not a miserable woman weeping over her destiny, but a powerful figure demanding obedience from her son.60 Consequently Coriolanus withdrew his forces from before the city.61 The story of Veturia’s women reminds us once again of the Sabine women.62 In these stories there emerges an interesting picture of the public role of mothers and wives in the Roman historical memory. 55

56

57 58 59 60 61 62

Livy 1.13.1–3: tum Sabinae mulieres, quarum ex iniuria bellum ortum erat, crinibus passis scissaque veste victo malis muliebri pavore, ausae se inter tela volantia inferred. (‘then the Sabine women, the wronging of whom had caused the war, with loosened hair and torn garments, their woman’s timidity lost in the face of their misfortune, dared to go amongst the flying missiles’). Dionysius of Halicarnassus, Antiquitates Romanae 2.45.2. Hersilia was the mother of one of the Sabine virgins and the leader of the Sabine women. cf. Cassius Dio 1.5.5–7 and Plutarch, Romulus 19, especially 5f. Livy 1.13.6. Livy 2.40.3. Dionysius of Halicarnassus, Antiquitates Romanae 8.40.2. Cf. Gagé 1963, 100; Sirago 1983, 83–102. Livy 2.40.5–9; for an analysis of this story, see, e.g., Bonjour 1975, 157; Mustakallio 1990, 130 and 1999, 56f. Livy 2.40.10f. Livy 1.10.1; 1.13.1.

Grief and Mourning in the Roman Context

247

There were also other occasions where women protected one of their own by lamenting and crying. During the internal crisis concerning the despotic power of the decemviri, the comitatus muliebris tried to protect the young virgin, Verginia, from the desires of the arrogant Appius Claudius.63 Verginia’s father was sent out of town to a war-camp. There was only a group of women, a comitatus muliebris, without any specific leader, who tried to protect her by crying and lamenting.64 When her father finally came onto the scene to protect her, the game was already lost and, despite all, her father ‘sacrificed’ her. In this situation the pitiful comitatus muliebris cried their protest against the ruling despot: ‘Was it on these terms that children were brought into the world? Were these the rewards of chastity?’65 The agency of women around Verginia in this case was based on lamentations and cries; the girl herself remained silent. In this connection women were no longer described by Livy as pitiful mulieres but as respectable matrones! This event finally affected the whole of society and resulted in Appius Claudius’ exile.66 We may therefore suggest that ritual lamentation was an essential component of female public demonstrations, at least in the literary tradition of the Roman historiography. It seems that, according to the Romans’ historical memory of Early and Republican Rome, when women took part in public matters they did it in the same ritualized manner as they acted during funerals. 5 CONCLUSION We have already discussed hired mourners who took care of the mourning rituals during funerals, according to the fragments of the Law of the Twelve Tables and Roman historiography. They performed the neniae and sometimes even put on extreme shows of grief by self-mutilation. As Servius says, in Roman funerals there were in fact two different female groups taking part, professional mourners with their leader (praefica), who was in general an older woman with a good voice, and the female family members whose mourning was directed by the mother of the deceased.67 The topos of groups of matrons lamenting in public during times of war and crisis was quite common in Livy’s and Dionysius of Halicarnassus’ descriptions of Republican Rome. It is also interesting that the leaders of the mourning groups in these stories were usually older women of higher status. It is even more significant that the writers pictured them as delivering fearless speeches in front of great

63 64 65 66

67

Livy 3.44–58 and Dionysius of Halicarnassus, Antiquitates Romanae 11.28–46. Livy 3.47.4 and 8: comitatus muliebris, globo mulierum. Livy 3.48.8: sequentes clamitant matronae: eamne liberorum procreandorum condicionem, ea pudicitiae praemia esse? Translation by B. O. Foster. Livy 3.58.6. For discussion of the terms mulier and femina, see Santoro L’Hoir 1992, 81f., and Mustakallio 1999, 54. On the silence of Verginia, see Cantarella 1996, 56. For the stories in their Augustan context, see Mustakallio 2012, 165–174. Servius, Commentarius in Vergilii Aeneidos 9.484.

248

Katariina Mustakallio

generals. In a hierarchical society like Rome’s it indicates the important role and influence of matrons in society as controllers of public morals and values.68 As we have already seen, the mourning rituals and lamentations performed at Roman funerals altered in many ways from the Republican Era to the Principate. During the late Republic, Roman funeral practices, especially those of upper class families, were strongly influenced by Hellenistic funerals, as Harriet Flower has already shown.69 Roman families of noble origin competed with each other in the luxury and splendour of their funeral festivals, games, and so on. One of the developments here was the change in the nature of the luctus matronarum.70 When the matrons at funerals were totally replaced by hired mourning women, the literary topos of luctus matronarum also disappears from Roman historiography. During the first centuries BCE and CE the duty for people to show their respect after a death of a statesman and mourn in public was extended. Not only the family members of the Emperor but also the magistrates were included. 71 Nevertheless, we find hardly any mention in Roman historiography of groups of women lamenting in public. In the Roman Empire new funeral habits and ways to show respect after death were created. At the same time, descriptions of mourning in public are not standardised but reveal individual features. One example is Cleopatra, whose mourning and wailing in front of Mark Antony’s dead body is reported in a highly exotic tone.72 In the case of Caesar’s murder and the improvised funeral, as well as in the context of the death of Drusus, crowds are said to have shown their strong emotions in public.73 Concerning the death of Germanicus and his funeral, Tacitus describes a universal grief that was so overwhelming that ‘you could not distinguish kinsfolk from strangers, or the laments of men from those of women’, though the imperial family did not show their grief at all.74 In Imperial Rome funeral and commemorative celebrations were arranged for dead members of noble families, both male and female, including laudationes funebres as well as funeral games – which became very popular during the period of the late Republic.75 One of the main changes in funerary practices concerns the place of gender and status. In public, funerals hymns and neniae were performed by youths and virgins. Matrons were no longer active in luctus. In the Roman historical tradition public mourning by matrons became the exception. At the same time literary descriptions of mourning became individualised.76 68 69 70 71 72 73 74 75

76

Mustakallio 1999, 53–64; on the status of older women, see Mustakallio 2011, 45–53. Flower 1996, 122–127. Horace, Ars Poetica 431–433; Lucilius 27.18. See, e.g., Šterbenc Erker 2009, 156f. Plutarch, Antonius 82. Appian, Bellum civile 2.146–8; Seneca, De consolatione ad Marciam 3.2. Tacitus, Annales 3.1. In this context the picture of common grief stresses the untraditionally cold behaviour of the imperial family itself, see Tacitus, Annales 3.5. Munera, usually understood as the gladiatorial games, were sometimes organised to honour the memory of noble ladies such as the daughter of Iulius Caesar; see Suetonius, Divus Iulius 26.2: Munus populo epulumque pronuntiavit in filiae memoriam; and Plutarch, Caesar 55.3. See Hopkins 1983, 14–25. E.g., Plutarch, Antonius 88f.

Grief and Mourning in the Roman Context

249

BIBLIOGRAPHY Adrados, F. R. (1999) History of Graeco-Roman Fables I, Brill. Alexiou, M. (1974) The Ritual Lament in Greek Tradition, Maryland. Audibert, R. (1885) Funérailles et sépultures dans la Rome païenne. Des sépultures et de la liberté des funérailles en droit civil, Paris. Bauman, R. (1992) Women and Politics in Ancient Rome, London. Bettini, M. (2005) Death and its Double. Imagines, Ridiculum, and Honos in the Roman Aristocratic Funeral, in K. Mustakallio et al. (eds.), Hoping for Continuity. Childhood, Education, and Death in Antiquity and the Middle Ages, Rome, 191–202. Bianchi Bandinelli, R. (1981) Roma. Arte romana nel centro del potere, Milano. Bodel, J. (1994) Graveyards and Groves. A Study of Lex Lucerina, Cambridge. Bonjour, M. (1975) Les personnages féminins et la terre natale dans l’épisode de Coriolan (Liv.2.40), Revue des Études Latines 53, 157–181. Brennan, T. (1998) The Old Stoic Theory of Emotion, in Sihvola and Engberg-Pedersen (eds.) 1998, 21–70. Cantarella, E. (1996) Passato prossimo. Donne romane da Tacita a Sulpicia, Milano. Chaniotis, A. (2011) Emotional Community Through Ritual. Initiates, Citizens, and Pilgrims as Emotional Communities in the Greek World, in A. Chaniotis (ed.) Ritual Dynamics in the Ancient Mediterranean. Agency, Emotion, Gender, Representation, Stuttgart, 263–291. Cumont, F. (1944) Recherches sur le symbolisme funéraire des Romains, Paris. Dutch, D. (2008) Neniae: Gender, Games, and Lamentation in Ancient Rome, in A. Suter (ed.) Lament: Studies in Ancient Mediterranean and Beyond, Oxford, 258–279. Engels, J. (1998) Funerum sepulcrorumque magnificentia. Bergräbnis- und Grabluxusgesetze in der griechisch-römischen Welt mit einigen Ausblicken auf Eischränkungen des funeralen und sepulcralen Luxus im Mittelalter und in der Neuzeit, Stuttgart. Foley, H. P. (1993) The Politics of Tragic Lamentation, in A. Sommerstein et al. (eds.), Tragedy, Comedy, and the Polis, Bari, 101–143. Flower, H. I. (1996) Ancestor Masks and Aristocratic Power in Roman Culture, Oxford. ––– (1998) Rethinking ‘Damnatio memoriae’: The Case of Cn. Caplurnius Piso Pater in AD 20, Classical Antiquity 17, 155–185. Frisone, F. (2000) Leggi e regolamenti funerari nel mondo greco. I. Le fonti epigrafiche, Galatina. Gagé, J. (1963) Matronalia. Essai sur les devotions et les organisations cultuelles des femmes dans l’ancienne Rome, Bruxelles. Hänninen, M. (2000) Women as Worshippers of Juno from the Mid-Republican to the Augustan Era, Helsinki. Harris, W. V. (2001), Restraining Rage. The Ideology of Anger Control in Classical Antiquity, Cambridge, Mass./London. Hopkins, K. (1983) Death and Renewal, Cambridge. Humphreys, S. C. and H. King (eds.) (1981) Mortality and Immortality: The Anthropology and Archaeology of Death, London. Huntington, R. and P. Metcalf (1979) Celebrations of Death. The Anthropology of Mortuary Ritual, Cambridge. Koziak, B. (2000) Retrieving Political Emotion: Thumos, Gender, and Aristotle, Pennsylvania. Le Bonniec, H. (1958) Le Culte de Cérès à Rome, des origines à la fin de la République (Études et commentaires 27), Paris. Loraux, N. (1991) Le madri in lutto, Roma, Bari. Marcus, D. D. (2004) Grim Pleasures: Statius’s poetic consolations, Arethusa 37, 105–135. Maurin, J. (1984) Funus et rites de separation, in A. Fraschetti (ed.), Aspetti dell’ideologia funeraria nel mondo Romano (A.I.O.N., sez. Archeologia e Storia Antica VI), Napoli, 193–208.

250

Katariina Mustakallio

Mustakallio, K. (1990) Some Aspects of the Story of Coriolanus and the Women Behind the Cult of Fortuna Muliebris, in H. Solin and M. Kajava (eds.), in Roman Eastern Policy and Other Studies in Roman History, Helsinki, 125–131. ––– (1994) Death and Disgrace. Capital Penalties with ‘post mortem’ Sanctions in Early Roman Historiography, Helsinki. ––– (1999) Legendary Groups and Female Networks in Livy, in P. Setälä and L. Savunen (eds.), Female Networks and the Public Sphere in Roman Society, Rome. ––– (2003) Women and Mourning in Ancient Rome, in L. Larsson Lovén and A. Strömberg (eds.), Gender, Cult, and Culture in the Ancient World from Mycenae to Byzantium, Sävedalen, 86– 99. ––– (2011) Representing Older Women: Hersilia, Veturia, Virgo Vestalis Maxima, in C. Krötzl and K. Mustakallio (eds.), On Old Age. Approaching Death in Antiquity and the Middle Ages, Turnhout, 41–56. ––– (2012) Women Outside Their Homes. The Female Voice in the Early Republican Memory. Reconsidering Cloelia and Veturia, Index Quaderni camerti di studi romanistici. International Survey of Roman Law 40, 165–174. Parker, R. (1983) Miasma. Pollution and Purification in Early Greek Religion, Oxford. Prescendi, F. (2008) Le deuil à Rome: mise en scéne d’une émotion, Revue de l’histoire des religions 225/2, 297–313. Santoro L’Hoir, F. (1992) The Rhetoric of Gender Terms, ‘Man’, ‘Woman’, and the Portrayal of Character in Latin Prose, Leiden. Scheid, J. (1992) The Religious Roles of Roman Women, in P. Scmitt Pantel, (ed.) A History of Women From Ancient Goddesses to Christian Saints, Cambridge, 377–408. Schmitt Pantel, P. (1992) (ed.) A History of Women From Ancient Goddesses to Christian Saints, Cambridge. Sihvola, J. (2010) Friendship and Sociability in Aristotle, in K. Mustakallio and C. Krötzl (eds.), De Amicitia. Friendship and Social Networks in Antiquity and the Middle Ages, Rome. Sihvola, J. and T. Engberg-Pedersen (eds.) (1998) The Emotions in Hellenistic Philosophy, Dordrecht. Sirago, V. A. (1983) Femminismo a Roma nel primo impero, Catanzaro. Stears, K. (1998) Death Becomes Her. Gender and Athenian Death Ritual, in S. Blum and M. Williamson (eds.), The Sacred and the Feminine in Ancient Greece, London/New York, 113–127. Šterbenc Erker, D. (2009) Women’s Tears in Ancient Roman Ritual, in T. Fögen (ed.), Tears in Graeco-Roman World, Berlin, 135–160. Suter, A. (ed.) (2008) Lament: Studies in the Ancient Mediterranean and Beyond, Oxford. Thomas, Y. (1992) The Division of Sexes in Roman Law, in Schmitt Pantel (ed.) 1992, Cambridge, 85–137. Toynbee, J. M. C. (1971) Death and Burial in the Roman World, London. Vittinghoff, F. (1936) Der Staatsfeind in der römischen Kaiserzeit. Untersuchungen zur damnatio memoriae, Berlin. Zaidman, L. B. (1992) Pandora’s Daughters, in Schmitt Pantel (ed.) 1992, 368–372.

PICTURE CREDITS Figure 1: Figure 2:

Marble relief from the tomb of the Haterii with representation of a mourning at scene. Rome. Late first century CE. Photo: Archivio Fotografico, Musei Vaticani. Marble grave relief with representation of a funerary procession. Amiternum. Late first century BCE/early first century CE. Photo: Archivio Fotografico, Museo Nazionale d’Abruzzo, L’Aquila. Su concessione del Ministero per i Beni e le Attività Culturali – Soprintendenza per i Beni Storici, Artistici e Etnoantropologici dell’Abruzzo di L’Aquila.

GALEN AND GRIEF The Construction of Grief in Galen’s Clinical Work Daniel King 1 INTRODUCTION You already know that Philippides, the grammarian, when he lost his books in a fire, he with1 ered as a result of depression and grief, and died.

Galen’s reference to the grammarian Philippides announces, with remarkable irony for one so bookish as Galen, the danger of being too attached to one’s library. Within the context of Galen’s treatise On Freedom from Grief (henceforth: Freedom), such comments might be justifiably read as a strategic move, alerting the reader to exactly what is at stake in his instructions for living free from grief: take heed, reader, lest you follow Philippides down a dangerous path! However, Galen’s anecdote also raises more historical questions. Does Galen imagine the experience he describes as one to which his readers might well succumb? How does one die from grief? And, if one can die from grief, what implications does this have for recognising and treating the experience in the late second century CE? In short, how, and why, does the example of Philippides bite? This chapter considers Galen’s construction of lupē (grief) in the light of the questions raised by Freedom. My discussion proceeds through three different, but related stages. In Section Two, I show how Galen subtly positions his approach to grief within the context of a broader philosophical debate about the nature and experience of the emotion. I show how, especially in On the Doctrines of Plato and Hippocrates (henceforth: On Plato and Hippocrates), Galen continually approaches the experience through a scientific and medical framework, which privileges division, definition, and categorisation, and which locates lupē within a family of other pain experiences. In Section Three, I investigate Galen’s treatment of grief in a number of practical texts, notably On the Therapeutic Method and On Affected Places. I investigate how Galen conceptualises grief alongside other con1

Galen, On Freedom from Grief 7: Πεπύσθαι δὲ ὡς καὶ Φιλίδης µὲν ὁ γραµµατικὸς ἀπολλυµένων αὐτῷ τῶν βιβλίων κατὰ πυρκαϊὰν ἀπὸ δυσθυµίας καὶ λύπης διεφθάρη συντακείς. Text is taken from the edition of Boudon-Millot 2007. All other Greek is taken from the Kühn edition of Galen, where available, or the latest Corpus Medicorum Graecorum (CMG) volume. References to Galenic texts, for the sake of clarity, use the standard English title first, followed by the abbreviations used in the Cambridge Companion to Galen (cf. Hankinson 2008). All translations are my own, unless otherwise stated.

252

Daniel King

ditions such as fever, and those who suffer gastric and cardiac sunkopē. I show how grief is presented as a cause of other diseases, and how this impacts on the way in which Galen approaches, treats, and organises the historical experiences of his patients. Finally, in Section Four, I will attempt to contextualise Galen’s treatment of grief by investigating how his approach to the emotional experience is presented in a number of case-histories from across the Galenic corpus. Galen’s treatment of grief was situated within questions about the way in which he views the soul, its pathē, and their relationship to the body.2 There are two themes in the scholarly engagement with Galen’s broad view of the emotions, and their therapy, which I want to outline before starting my analysis. Firstly, Richard Sorabji argues that Galen’s approach to grief must be understood within the context of Galen’s generally physicalist approach to emotional pathē. Sorabji’s reading of Galen’s views is primarily taken from an analysis of The Faculties of the Soul Follow the Mixtures of the Body (henceforth: Faculties of the Soul), where Galen develops a model of emotional experience and treatment, which emphasises the interdependency of the body and the soul.3 On Sorabji’s reading, the corollary of the view advanced in Faculties of the Soul involves managing emotional experience through controlling the mixtures of the body with dietetic practices, including food and exercise, music and harmony, and their combination with philosophical instruction.4 Christopher Gill, in Naturalistic Psychology in Galen and Stoicism, suggests, in contrast, that the view presented in Faculties of the Soul is highly contextual, and, in fact, in some ways contradicts the models of psychē, soma, and their relationship that Galen supposes at other points in his corpus, notably in The Passions of the Soul and in Freedom. Gill holds, further, that Galen maintains a division between medical therapy, which approaches the individual as a body, and philosophical therapy of the emotions, which treats the individual as a psychological agent.5

2

3

4 5

Gill 2010, 316–322; Tieleman 1996. There is, of course, a large bibliography on these various issues. I give only an outline. For Galen’s approach to the soul and psychology in general, see Hankinson 1991, 197–233; von Staden 2000, 79–116; Tieleman 2003b, 131–161; Gill 2007, 97–120; Donini 2008, 184–209. For emotions and their management, see Hankinson 1993, 184–222. For the mind-body relationship, see Garcia Ballester 1988, 117–152; Sorabji 2000; Tieleman 2003b; Hankinson 2006, 232–258. Sorabji 2000, 255f.: ‘Galen applies the thesis that psychological capacities follow blends not only to the intellect, but to emotional traits and emotions [...]. He takes the message that [blood seed, humours, climate, food and drink] all affect our emotional states via their effect on the blend of hot, cold, fluid and dry in our bodies’; Galen, Faculties of the Soul iv.676K, 779Kf. Food and drink combined with harmony: Sorabji 2000, 257f. On the importance of philosophical instruction: 258–260. Gill 2010, 316–322. For Faculties of the Soul as invalidating psychological theories: 320; the division between medical and psychological therapy, 320: ‘... Galen’s medical therapy, like that of ancient medicine generally, was directed primarily at the body. ... Galen does not develop a separate areas of medical therapeutics directed at the person as psyche.’

Galen and Grief: the Construction of Grief in Galen’s Clinical Work

253

In this chapter, I take an approach which has been influenced heavily by anthropological approaches to medical culture and illness.6 I am interested, primarily, in investigating the ‘experience’7 of grief as part of a more general experience of being ill, or suffering. In the context of this investigation, what interests me is less the question ‘what is grief qua pathos?’, and more ‘how is this affliction located or understood as part of the process of being ill?’. How is it explained to patients? Why might it be considered an important element of discussion in the encounters between patients and doctors? How and where does Galen locate grief in relation to other conditions in such conversations? How are patients (and doctors) invited to negotiate illness alongside a notion of grief? Where, in short, is grief to be placed in the broad landscape of experiences that constitute this culture’s notion (or notions) of being ill? This entails questioning passages or texts in which Galen defines his understanding of grief, but also ones in which Galen explains to his patients (and to his readers) how to understand, diagnose, or recognise various illnesses which he sees as related in some way to grief. This focus will attempt to achieve two things. Firstly, it will move the discussion from the philosophical (con)texts in which Galen discusses grief to ones in which he discusses grief and other conditions, such as the medical handbooks and technical treatises. Secondly, I hope it will provide a basis for relocating Galen’s understanding of the experience of grief within a broader social context. 2 DEFINING GRIEF AND PAIN The definition and classification of different diseases, conditions, and symptoms was a central tenet of both Galen’s understanding of medical knowledge and his clinical practice.8 In this classificatory process, pain was given particular empha-

6 7

8

On medical anthropology generally, see Kleinman 1989. See Frank 1995, 4–6 on the role of narrative in turning illness or disease into a personal and unique experience. Medical anthropology, in its focus on the narrative aspects of clinical encounters and casehistories, has seen, in the last three decades, an explosion of interest in illness qua experience. For the purposes of this paper, I start from the following assumption about experience, taken from Arthur Kleinman (Kleinman 1988, 5): ‘local cultural orientations ... organise our conventional common sense about how to understand and treat illness; thus we can say of illness experience that it is always culturally shaped. Paradoxical as it sounds, then, there are normal ways of being ill (ways that our society regards as appropriate).’ There are obvious caveats to Kleinman’s emphasis on cultural influence (as he himself acknowledges). What interests me, primarily, are the ways in which Galen organises for his patients and his readers (other doctors?) the phenomenon of grief in relation to other diseases, other symptoms and conditions, and other feelings of distress. For further discussion of the notion of illness experience, see Frank 1995; Kleinman 1988, 3–30. The loci classici for Galen’s interest in definition and classification of bodily conditions and symptoms are: Differences of Diseases, Causes of Diseases, Differences of Symptoms, Causes of Symptoms. On these texts, see Johnston 2006.

254

Daniel King

sis as it constitutes a major diagnostic tool.9 In what follows here, I examine how his concern to classify and define pain inflects his approach to the emotional experience of grief. I turn to a number of contexts in which Galen confronts the issues of definition and classification head-on. I begin by, briefly, outlining one definition of grief in a number of philosophical texts, notably in Passions of the Soul. I then attempt to show how the definition developed in that context is eschewed in other clinical and philosophical settings such as Causes of Symptoms and On Plato and Hippocrates. These two works demonstrate an intense concern with the terminology with which we designate the various forms of pain. Galen’s treatment of the language of grief, in these contexts, is underpinned by an Aristotelian process of division and definition which incorporates the experience of the emotion into a scientific discussion about the various pain symptoms the individual is heir to: Galen is determined to engage with grief through the Aristotelian framework of his clinical work.10 By using that classificatory framework to construct the way in which grief, as an emotional condition, might be understood in relation to other painful experiences Galen repositions himself against certain philosophical thinkers, especially in On Plato and Hippocrates, and also emphasises the connection between lupē and other conditions. In order to create a foil for thinking about the way in which the languages of pain and grief are treated in the two main texts of this discussion, I want to turn very briefly to a number of passages in Passions of the Soul. Here, Galen is primarily concerned to criticise Antonius’ work On Guarding Against the Passions, which he sees as confusing the categories of what constitutes an error (hamartēma) and what are pathē.11 One way this polemic objective is met is through Galen’s attempts to define the emotions, which he does in a number of ways: firstly, by emphasising a division between the pathē of the soul, and the ponoi of the body; and secondly, between by stressing the relationship between different pathē. 12 At chapter v.35K, in valorising, once again, his long-held method for dealing with the emotions, he proceeds to outline what he sees as the relationship between various emotional states:13 For obstinacy, love of glory, and love of power are diseases of the soul. Greediness is less harmful than these, but it, too, is, nevertheless, a disease. And what must I say of envy? It is the worst of evils. I call it envy when someone is grieved over the success of others. All grief

9 10 11 12 13

Pain is a common theme throughout Galen’s works. The most explicit treatment of its diagnostic importance is Galen, On Affected Places viii.70Kf. On Aristotle’s influence on Galen in terms of scientific method and categorisation, see Tieleman 2008, 49–65; Hankinson 1992, 3511–3518. Galen, Passions of the Soul v.1K. Galen, Passions of the Soul v.37K. Galen, Passions of the Soul v.35K: ... γὰρ ἡ φιλονεικία καὶ φιλοδοξία καὶ ἡ φιλαρχία πάθη τῆς ψυχῆς εἰσι. τούτων ἔλαττον µὲν , ἀλλ’ ὅµως καὶ αὐτὴ πάθος. περὶ δὲ τοῦ φθόνου τί δεῖ καὶ λέγειν; ἔσχατον τῶν κακῶν ἐστιν· ὀνοµάζω δὲ φθόνον, ὅταν τις ἐπ’ ἀλλοτρίοις ἀγαθοῖς λυπῆται. Πάθος µέν ἐστι καὶ λύπη πᾶσα, χειρίστη δὲ ὁ φθόνος ἐστίν, εἴτε ἓν τῶν παθῶν εἴτε λύπης ἐστὶν εἶδος πλησιάζον δέ πως αὐτῇ. Translation from Harkins 1963, 53.

Galen and Grief: the Construction of Grief in Galen’s Clinical Work

255

is a disease, and envy is the worst grief, whether we call it a passion or a kind of pain which borders on grief.

Galen’s strategy is obvious at this point: what matters is not only determining what counts as a pathos (although this is by no means a complete list), but what is the relationship between these particular experiences. In this context, grief appears to be given a particular nuance. All the various forms of grief, so Galen argues, are to be classified among the pathē, but a role is also attributed to grief in other emotional states. Grief, on this reading, is central to how we experience the emotion of envy. Galen rarely returns to explain this complex of inter-relations in the remainder of the treatise. The Passions of the Soul adopts a range of models, then, for thinking about both emotions in broad terms, and grief in particular. It is, perhaps, not surprising in this context that Galen spends little time outlining how we experience grief, or what the implications of grief are for the broader condition of the individual. What is alluded to, but never elaborated, in this passage is precisely what the experience of grief is. Moreover, the type of approach which inheres in Causes of Symptoms is eschewed in Galen’s clinical work. This work takes a slightly different approach by locating lupē amongst other experiences of distress. According to vii.115K, pain results when a violent or intense affection contrary to nature occurs in perceiving bodies (τὸ µὲν παρὰ φύσιν καὶ βιαίως γιγνόµενον ἀθρόως ἐν ἡµῖν πάθος, ἀλγεινόν).14 Galen’s definition of the pathos draws on a long philosophical tradition which can be traced to Plato, at least, and, in this particular case, is explicitly presented as a pastiche of both Hippocratic and Platonic views.15 This explanation of what is common to all pain experiences initiates a detailed discussion of various pain symptoms in the different sense organs (and the conditions which cause them) and the sub-categories into which they can be divided. Galen’s approach divides the experience of pain into that which constitutes either a destruction of, or damage to, the continuity of a solid body or anomalous humoural states (duskrasia).16 There are two points to note about Galen’s treatment of the first category, the destruction of continuity. Firstly, by outlining how different sensory bodies are distressed by overwhelming experiences which cause a disruption to their continuous nature, Galen argues that pain experiences which occur in vastly different sensorial contexts – taste, sight, and hearing – and arise from different phenomena – hot and cold, cutting, light, and the induction of vapours – should be collapsed into a broad category of a ‘contrary to nature experience’; they can be seen as fundamentally the same type of experience. Secondly, Galen’s approach to the experience of pain underpins a particular focus on the 14 15 16

Galen, Causes of Symptoms vii.115K. Both Hippocrates and Plato are listed in Galen’s passage. Plato’s Timaeus is explicitly referred to, but there appears to be no exact reference to Hippocrates. Continuity of solid body: Causes of Symptoms vii.116Kf; duskrasia: vii.176K; cf. Galen, On Affected Places viii.79Kf for further references to this twofold division between the two types of pain experience. I assume Galen’s two types are so far removed in this passage because he wants to delay the discussion of anomalous duskrasia until the section at which he is explicitly concerned with the humours.

256

Daniel King

way in which the language of distress might be mobilised. In speaking of what is meant by an ‘overwhelming experience of separation and discontinuity’, Galen evinces a particular approach to the nomenclature used for pain or distress:17 I shall call ‘overwhelmingly’ what is at the same time great and sudden. This, I imagine, is also what Plato meant when he said that disturbed sensation is brought about from a violent and at the same time overwhelming affection (pathema) coming upon perceiving bodies. For it will make no difference whether one speaks of a distressed, vexed, pained, disturbed or burdened perception, just as with respect to the affection (pathema) itself, [whether one speaks of] distress, vexation, pain, suffering or grief.

Read at one level, Galen takes aim at the process of differentiating pain experiences linguistically: that which is ‘distressing’ (ἀνιαράν) and that which is ‘grievous’ (λυπηράν) is no different from that which is ‘painful’ (ὀδυνηρὰν ἢ ἐπίπονον). According to Galen, the experience of the pain transcends terminological form: distress = grief = pain = suffering. Galen’s categorical system, ironically, seems directed towards collapsing the distinction between what might be seen as fundamentally different pain experiences. To be sure, Causes of Symptoms vii.118K is primarily concerned with what he sees as unnecessary linguistic or terminological overload. When he turns more explicitly to the analysis of psychic conditions, and their relationship with other experiences such as fever and rigors, he continues to emphasise the interconnection between different types of pain. According to Causes of Symptoms vii.176K, when an anomalous duskrasia occurs in any one of the sensory bodies, the experience of pain is equal to that of the duskrasia. Galen’s interest, here, is particularly focussed on the genesis of fevers and the rigors which, in certain circumstances, follow the contraction of fever. Both these conditions have implications for the experience of pain and are linked to the pathē. Fevers are painful: they distress the solid parts of an animal because they are changed and altered by a heat that is ‘contrary to nature’.18 In some psychical conditions, fear and shame especially, Galen suggests that there is a movement of innate heat both inward (towards the archē) and outward towards the superficial layers of the body.19 When this occurs, fevers and/or rigors ensue: the pathē are seen as part of the experience of distress because they cause, and their experience is bound up with, those things which are painful. While this indrawing or expulsion of heat is linked primarily to fear and shame, it is also presented as part of the experience of grief, during which the ex-

17

18 19

Galen, Causes of Symptoms vii.117–118K: καλῶ δὲ ἀθρόως τὸ κατὰ τὰ µεγάλα ἅµα καὶ ταχέως. τοῦτο δὲ δήπου καὶ ὁ Πλάτων ἐβούλετο, σύνθετον ἐκ βιαίου τε ἅµα καὶ ἀθρόου παθήµατος ἐγγινοµένου τοῖς αἰσθητικοῖς σώµασι τὴν ἀνιαρὰν αἴσθησιν ἀποτελεῖσθαι φάσκων. οὐδὲν γὰρ διοίσει λέγειν ἀνιαρὰν ἢ λυπηρὰν ἢ ἀλγεινὴν ἢ ὀδυνηρὰν ἢ ἐπίπονον αἴσθησιν̣· ὥσπερ οὐδὲ αὐτὸ τὸ πάθος, ἀνίαν ἢ λύπην ἢ ὀδύνην ἢ πόνον ἢ ἀλγηδόνα. Translation from Johnston 2006. Galen, Causes of Symptoms vii.176K. Galen, Causes of Symptoms vii.191-195K.

Galen and Grief: the Construction of Grief in Galen’s Clinical Work

257

perience is ‘shorter’ (κατὰ βραχύ) but not different ‘in class’.20 Grief, then, is incorporated into Galen’s categorical system on a number of levels: in linguistic terms, in which it is simply presented as one of many pain experiences; and in causal terms, in which it is seen as a central part of other experiences which are, themselves, painful. Causes of Symptoms, then, attempts to organise the experience of pain into an Aristotelian framework (at least in its adherence to the notion of categorical division) that divides that which is common into two further sub-categories. This allows individual and specific experiences to be seen as part of a broad family of pain. In such a system the experience of grief is assimilated not only to other psychical conditions, but also to other forms of pain and distress. If what I have been suggesting for Causes of Symptoms is true, how do these themes map onto Galen’s definitional and classificatory aims in his On Plato and Hippocrates? As is well known, On Plato and Hippocrates is one of Galen’s major works in which he advances his views on the tri-partition and immortality of the soul. Throughout, Galen uses a combination of anatomical demonstration and philosophical thought to prove that both Hippocrates and Plato were correct (and in agreement!) in their belief in the tri-partite nature of the human soul.21 Fundamental to this objective is an attempt to prove that Chrysippus’ reading of, along with other Stoic approaches to, the soul was incorrect. This polemical engagement with earlier Stoic thinkers involves quoting extensively (although, arguably, not always fairly) from Chrysippus’ lost On the Soul.22 What interests me in this context is the moments when Galen engages with Chrysippus’, and others’, views on how and where individuals perceive (αἰσθάνεσθαι) or experience (πάσχειν) the πάθη διανοίας or the symptoms which occur ἐν ταῖς λύπαις. Galen’s engagement with Chrysippus’ views ultimately provides a number of clues as to how Galen approaches the experience of grief, and which reiterate the view of grief and its relationship to other pain experiences outlined in Causes of Symptoms. Galen’s self-positioning in relation to Chrysippus’ views about the experience of grief is hard to miss. At On Plato and Hippocrates v.335Kf, in the course of recounting an argument for locating and associating the hegemonic part of the soul with the sensory nerves of the brain, Galen attributes an argument about the soul’s experience of grief to Chrysippus. According to Galen, Chrysippus held the 20

21

22

Galen, Causes of Symptoms vii.193K: ὅπερ δ’ ἐν φόβοις ἀθρόως, τοῦτ’ ἐν λύπαις κατὰ βραχὺ πάσχουσι, µεγέθει καὶ σφοδρότητι τῶν παθῶν, οὐχ ὅλῳ τῷ γένει διαφερόντων; cf. ibid. vii.194K. For a general introduction to On Plato and Hippocrates, see De Lacy 1981, 12–58. For its capacity to inform a reconstruction of earlier Stoic thought, see Tieleman 1996. Sorabji 2000, 29–54 is another example of the reconstructive process which relies heavily on Galen’s On Plato and Hippocrates; cf. SVF II 879–911. For an analysis of Galen’s views of the soul using On Plato and Hippocrates the best are Hankinson 1991, 197–233; Donini 2008, 184– 209; Manuli 1988, 185–214; De Lacy 1988b, 43–63. On Galen’s relationship with Chrysippus’ On the Soul and the use of On Plato and Hippocrates as a source for that lost work, see Tieleman 1996, 134–146, especially 135f.; cf. Tieleman 2003a; Stoics and Galen in general, see especially Gill 2007, 88–120.

258

Daniel King

view that individuals feel the pains associated with grief, distress, and anxiety in the chest, and that these pains must be felt in the governing part of the soul.23 Chrysippus’ construction of the experience of grief goes on, assimilating the experience to other common pain symptoms such as the experience of hurting in our foot or head.24 Chrysippus’ point, at least as Galen presents it, emphasises the similarity in terms of location between different experiences of distress.25 Galen’s response is telling, oscillating between agreement with some points, and refutation of the critical contention:26 For whenever a person says this while reasoning with us, as Chrysippus did, or as an assertion, we shall praise his discourse and say that anxiety and distress and suffering belong to the genus pain, or rather, if we must use words in accordance with Greek usage, that there is no difference between suffering and pain, just as there is none between a column and a pillar, or between eyes and organs of sight; but pain, as it were, is the genus of anxiety and distress. But that the pains arise in the place where the governing part is, this we shall no longer concede to Chrysippus.

Galen’s polemic, here, rephrases Chrysippean language on his own terms. In paraphrasing Chrysippus, Galen adds further definitional complexity: he connects lupē with a broader range of pain terminology by inserting ὀδύνη into the list of symptoms – Galen appears, perhaps, to be thinking more broadly of the sum of painful experiences than Chrysippus.27 More significantly, he refigures the process in terms of scientific distinction and division. Galen’s repeated reference to τῷ γένει or τὸ γένος flags his Aristotelian intention. The movement towards a more explicitly categorical system is, arguably, added to by Galen’s suggestion that we should understand that the various pains are experienced (διαφέρειν) no differently from each other.

23 24

25 26

27

Galen, On Plato and Hippocrates v.335K. Galen, On Plato and Hippocrates v.336K: ... τρόπον γὰρ ὅταν τὸν πόδα πονῶµεν ἢ τὴν κεφαλήν, περὶ τούτους τοὺς τόπους ὁ πόνος γίνεται, οὕτως συναισθανόµεθα καὶ τῆς κατὰ τὴν λύπην ἀλγηδόνος περὶ τὸν θώρακα γινοµένης. On the reconstruction of Chrysippus’ argument in this passage, see Tieleman 1996, 249–255, especially 254. On Chrysippus’ view of pain, see ibid., 254 n. 13. Galen, On Plato and Hippocrates v.303K: εἴτε γὰρ ἐρωτῶν ἡµᾶς οὕτως εἴποι τις, ὡς Χρύσιππος εἴρηκεν, εἴτε ἀποφαινόµενος, ἐπαινέσοµεν αὐτοῦ τὸν λόγον καὶ φήσοµεν ἀγωνίαν καὶ λύπην καὶ ὀδύνην ἀλγηδόνας εἶναι τῷ γένει, µᾶλλον δ’ εἰ χρὴ τοῖς ὀνόµασι κατὰ τὴν τῶν Ἑλλήνων χρήσασθαι συνήθειαν, ὀδύνας µὲν καὶ ἀλγηδόνας οὐδέτερον ἀλλήλων διαφέρειν, ὥσπερ οὐδὲ κίονα καὶ στῦλον οὐδ’ ὦπας καὶ ὀφθαλµούς· ἀγωνίας µέντοι καὶ λύπης οἷον γένος εἶναί τι τὴν ἀλγηδόνα. τὸ δ’ ἐν τῷ ἡγεµονικῷ τόπῳ τὰς ἀλγηδόνας γίνεσθαι, τοῦτ’ οὐκέτι τῷ Χρυσίππῳ συγχωρήσοµεν [αὐτῷ τὴν ἀλγηδόνα κατά τε τὸ ἡγεµονικὸν συνίστασθαι]. Translation from De Lacy 1981. Galen’s use of the indefinite optatives here suggests his reading of Chrysippus to be slightly disingenuous. The reference to ὀδύνη, further, in the passage attributed to Chrysippus in the lines above the passage presented here, is questionable: in De Lacy’s text καὶ τὴν ὀδύνην at III.7.2 is an interpolation; and one possibly based on the appearance of the term in Galen’s own statement about what Chrysippus has said: cf. CMG v.4.1.2, III.7f and critical apparatus to text. On Chrysippus’ view of pain in this passage, see Tieleman 1996, 253–255 and n. 13; cf. SVF III 412f; see also Sorabji 2000, 36–40.

Galen and Grief: the Construction of Grief in Galen’s Clinical Work

259

Galen’s concern with the way in which we experience grief in the region of the stomach or heart is one that he returns to throughout his tome. His interest in the algē associated with lupē is developed in slightly more explicit terms, for instance, at On Plato and Hippocrates v.272–275K. According to this passage, he turns to an assumption (which he attributes to Chrysippus) that we feel ‘bites in grief’ (τῆς ἐν ταῖς λύπαις δήξεως) in the region of the heart.28 Galen proceeds to correct Chrysippus’ suggestion in a number of provocative ways. He begins by emphasising that it is not the heart (that is, the kardia) that feels such things – ‘for we never perceive the heart being bitten during lupē nor during any other pathē of the soul’29 – but rather the sensation is perceived ‘in the mouth of the stomach’ (τῷ στόµατι τῆς γαστρός).30 Galen goes on to explain precisely why we experience this biting sensation. The mouth of the stomach doesn’t feel ‘pain’ (ἄλγηµα), he explains, during every experience of heartburn (καρδιαλγία), but only when it is irritated or bitten by acrid fluids (... ἀλλὰ µόνον ἐπειδὰν ὑπὸ δριµέων ὑγρῶν ἐρεθίζηταί τε καὶ ἀναδάκνηται).31 Galen’s remoulding of this passage continues in more dramatic terms in the ensuing lines:32 This is what happens to it in distress of mind, and this is why distressed persons vomit bile; but in some the bile goes downward, and their stomach passes unmixed bilious excrement. And not only the distressed, but many who have fasted excessively experience an irritation at the mouth of the stomach, and even more if they do not eat after violent exercise. For yellow bile collects in the stomachs of persons who are distressed or have exercised too strenuously; being irritated by it, they suffer heartburn.

Apart from the suggestion that those who grieve vomit or evacuate bile, Galen’s repetition of the metaphor of the bite is particularly evocative. Richard Sorabji holds that dēxis is used as a specific Stoic term for the shrinking in the heart or soul which occurs during certain emotions. For Sorabji, Chrysippus holds that the soul, which shrinks inward during grief, fear, or shame, is understood as being perceived as a ‘bite’.33 If Sorabji is right, then Galen’s use of the term appears to take on a greater poignancy. Galen employs the term throughout his corpus as a metaphor for everyday pain experiences. It also has particular emphasis as a term Galen applies to a certain form of humour: he refers to ‘the biting humour’ at 28 29 30 31 32

33

Galen, On Plato and Hippocrates v.274K. Galen, On Plato and Hippocrates v.274K: καὶ οὐδεὶς ᾔσθετο πώποτε τῆς καρδίας αὐτῆς δακνοµένης οὔτ’ ἐν λύπαις οὔτ’ ἐν ἄλλῳ πάθει ψυχῆς ... Galen, On Plato and Hippocrates v.274K. Galen, On Plato and Hippocrates v.276K. Galen, On Plato and Hippocrates v.276K: τοῦτο δ’ αὐτῷ συµβαίνει ἐν ταῖς λύπαις· διὸ καὶ χολὴν ἐµοῦσιν οἱ λυπηθέντες, ἐν οἷς δ’ ὑπέρχεται κάτω καὶ ἡ γαστὴρ αὐτοῖς ἄκρατα χολώδη διαχωρεῖ. συµβαίνει δὲ οὐ µόνον τοῖς λυπηθεῖσιν ἀλλὰ καὶ τῶν ἐπὶ πλέον ἀσιτησάντων οὐκ ὀλίγοις δάκνεσθαι τὸ τῆς γαστρὸς στόµα καὶ µᾶλλον εἰ σφοδρῶς προγεγυµνασµένοι µὴ προσενέγκοιντο σιτία. τοῖς τε γὰρ λυπηθεῖσι καὶ τοῖς γυµνασαµένοις εὐτονώτερον εἰς τὴν γαστέρα συρρεῖ χολὴ ξανθή· πρὸς ταύτης οὖν δακνόµενοι καρδιαλγοῦσιν. Translation from De Lacy 1981. Sorabji 2000, 34, 36–40, especially 38. Although the experience appears to be attributed to a range of thinkers, not just Chrysippus, or Stoics: cf. Sorabji 2000, 36–38, with accompanying notes.

260

Daniel King

various points in On Affected Places.34 Galen’s repetition of language flags up, arguably, the movement from a philosophical approach to the experience of grief, to one in which it is associated with the humoural condition of the body. Galen’s use of the term appears to look towards earlier Stoic uses of the term to refer to the emotional experience, and simultaneously flag his divergence from that usage. Galen, here, through this process of rephrasing, is able to connect the experience of grief with his humoural understanding of the body. It is easy to see from this passage the influence that Galen’s adherence to the theory of the humours might have on his construction of grief. The connection between the humoural condition of the body and the pain of grief is brought home repeatedly – grievers vomit yellow bile, and even if they do not bile collects in the stomach. The passage also underscores much of the experience of grief: it is likened to the experience of heartburn; it is given attributes that connect it with those who eat sparsely, and with those who exercise too much. As a way of thinking about how individuals experience lupē, Galen’s approach walks a delicate line between adoption and rejection of Stoic terminology, adapting, remoulding, and in the process placing it in a more scientific, medicalised context. By thinking about how he locates the experience of grief alongside other pain symptoms, it is possible to create a framework for understanding how Galen approaches the experience of grief. Defining what constitutes, and what underpins, a painful or distressing situation ensures that the distinctions between lupē and other forms of pain are continually collapsed: grief is integrated into a model of pain which fails to draw distinctions (linguistic or otherwise) between the pain we feel in our foot, and the distress we associate with the pathos. 3 MEDICAL MANUALS, GRIEF AND OTHER DISEASES In this section, I build on the conclusions of the previous discussion by investigating further the way in which Galen incorporates grief into his clinical writings. My principal focus in this section is on two practical handbooks – Galen’s On the Therapeutic Method and his On Affected Places – in which the author attempts to instruct other doctors on both clinical practice and diagnostic strategy. I use these texts to show how Galen interprets and explains the experience of various conditions through reference to grief and other emotional pathē. The manuals draw a connection between grief and the development of other conditions, such as fever, gastric sunkopē, and (in extreme cases) death. In doing so, these explanations offer a complex combination of medical thought about the structure of the body and various earlier philosophical approaches to the psychē, repackaging for a medical and clinical context a number of philosophical ideas about the soul. I turn to these two texts because they are especially valuable sources for examining the construction and experience of grief in the late second century CE 34

Cf. references to the δακνώδη χυµόν at Galen, On Affected Places viii.40K, ἀλγήµατα δακνώδη, at viii.41; cf. for more general usages, ibid. viii.388K.

Galen and Grief: the Construction of Grief in Galen’s Clinical Work

261

from a historical perspective. Both are explicitly concerned with instruction of other doctors: On the Therapeutic Method is dedicated to, at least, two other medical professionals, Hiero and Eugenianus.35 They are also explicitly written to provide advice about how doctors should understand the body, its conditions, and how to conduct medical treatment and care.36 Finally, the theoretical arguments for the nature of the body and Galen’s therapeutic method are repeatedly drawn, so he claims, from historical experience: at the very least Galen presents himself as having seen, talked to, diagnosed, and treated patients who suffered the various conditions he touches on.37 There are, of course, important caveats to these last two points: we know neither to what extent these accounts were read by other doctors, nor to what extent their models of grief were incorporated into their own medical practice; it is also difficult to judge Galen’s claims to historical autopsy.38 Nevertheless, Galen’s claims at least provide evidence of what could be presented as patient experience within the medical debates of the late second century. Galen’s treatment of these issues allows us to see how he and, possibly, a number of other doctors might perceive the patients who came to them, and understand, categorise, and treat their experiences; they also provide a glimpse of how patients might expect to have their experiences explained to them, and how they might negotiate such experiences themselves; where both medical professionals and lay sufferers might place the experience of grief on the complex landscape of illnesses, diseases, and conditions the individual suffered. In so doing, they provide a context for examining further why Galen’s anecdote about Philippides, with which we began this paper, might bite. Let us begin by considering the relationship between grief and the experience of fever. In a particularly telling discussion in Book 10 of On the Therapeutic Method, Galen turns to the question of the relationship between fever and various humoural mixtures within the body. In the course of his discussion, Galen claims that he has shown people by his actions (τοῖς µὲν ἔργοις ἀνθρώπους ἐπιδεικνύντες αὐτοῖς) that if they go without food for two days they will develop a fever (οὕς ἐὰν ἀσιτήσωσιν ἐπὶ δυοῖν ἡµέραιν, ἀνάγκη πυρέξαι), just like those individuals who have become angry, or suffer from grief, or who are sleepless (καθάπερ ἐνίους ὀργισθέντας ἢ λυπηθέντας ἢ ἀγρυπνήσαντας).39 Galen’s primary concern here is with the development of febrile conditions, and grief (along with the other pathē) only appears because of its connection with fever. This connection is elaborated further moments later: the ‘biting vapours and humours’ (τοὺς τοιούτους οὖν ἀτµοὺς καὶ χυµοὺς ...), Galen claims, ‘when they move through sensing bodies generate rigors and shivering’ (διὰ τῶν αἰσθητικῶν σωµάτων φερόµενοι φρίκας καὶ ῥίγη γεννῶσιν), and, when the pores of the body are blocked, they ‘ignite a fever’ (πυρέτον ἐξάπτει). Galen goes on to claim that 35 36 37 38 39

On dedications, see Galen, On the Therapeutic Method x.1K: ὦ Ἱέρων φίλτατε ... ἄλλοι τινές ... ἑταῖροι; ὦ Εὐγενιανὲ φίλτατε, at x.456K. Cf., e.g., Galen, On the Therapeutic Method x.810K, 872K. Cf., e.g., Galen, On the Therapeutic Method x.857—8K. On the question of Galen’s autopsy, see Mattern 2008, 38f., 217 n. 123. Galen, On the Therapeutic Method. x.679K.

262

Daniel King

too much exercise, insomnia, grief, and other pathē, generate many of these humours.40 Galen’s explanation, here, recalls references to the experience of pain discussed in the last section, where he emphasised that those who fast and grieve are bitten by acrid humours. In this context, the passage points more definitely towards substantial theory for the body’s structure and the role of grief in the development of various forms of both pain and fever. At this point, the manual broadens out from a number of case-histories and historical examples of grief to explore themes developed throughout the Galenic oeuvre. The treatment of fever taps into his discussion of anomalous duskrasia and its relationship to rigors, shivering, and fever at Causes of Symptoms vii.192– 3K. There, Galen’s analysis had been more in-depth in terms of showing how emotional pathē operate as antecedent causes of rigors, shivering, and fevers. Certain psychical affections, he points out, cause either an inward and/or an outward movement of the individual’s innate heat, their pneuma and their blood.41 Fear, for instance, draws together both the pneuma and the blood and leads them towards the archē and consequently cools the exterior and superficial parts of the body; the blood and heat may return to the exterior parts of the individual, but is often extinguished, and thus people who are afraid suffer rigors.42 While Galen’s main concern here is with fear – which he sees as having particular effects on the individual’s innate heat – he does connect it with other forms of emotional experience: 43 That which they suffer in a concentrated way in fears, they suffer slightly in distresses, the affections (πάθη) differing in magnitude and strength but not in the whole class.

Galen’s system emphasises the way in which grief has the capacity to have a series of serious physical and health implications, whose symptoms can be recognised on the body. What intrigues me about this explanatory model is the way it seems to provide a corporeal, humoural explanation for concepts with which we are familiar from Stoic explanations of the experience of various emotional states. This is particularly true of concepts Galen himself attributes to Stoic thinkers in On Plato and Hippocrates. The ‘drawing in’ of pneuma and blood recalls Chrysippus’ assumption that grief should be seen as a shrinking or a contraction of the irrational soul.44 Indeed, the idea that certain emotional pathē are caused through this ‘in40

41 42 43

44

Galen, On the Therapeutic Method. x.679Κ: τοὺς τοιούτους οὖν ἀτµοὺς καὶ χυµοὺς αἵ τε κινήσεις αἱ πολλαὶ καὶ σφοδραὶ καὶ ἀγρυπνίαι καὶ οἱ θυµοὶ καὶ αἱ λῦπαι καὶ αἱ φροντίδες ἐργάζονται πλέονας. Galen, Causes of Symptoms vii.193K. Galen, Causes of Symptoms vii.194K. Galen, Causes of Symptoms vii.193K: ὅπερ δ’ ἐν φόβοις ἀθρόως, τοῦτ’ ἐν λύπαις κατὰ βραχὺ πάσχουσι, µεγέθει καὶ σφοδρότητι τῶν παθῶν, οὐχ ὅλῳ τῷ γένει διαφερόντων. Translation from Johnston 2006. Cf. ibid. vii.194–5K. The question of Chrysippus’ definition of grief is difficult. Galen begins his quotation of Chrysippus by asserting that he (Chrysippus) defines grief as a ‘fresh notion of close-at-hand evil’ (On Plato and Hippocrates v.366K: τὴν λύπην ὁριζόµενος δόξαν πρόσφατον κακοῦ παρουσίας ...), but then attributes to him definitions that Galen claims he has taken from

Galen and Grief: the Construction of Grief in Galen’s Clinical Work

263

drawing’ or ‘shrinking’ has a long philosophical tradition.45 Galen’s explanation of fever appears, then, to offer a way of explaining or reframing the philosophical concept in decidedly medical terms for patients and other doctors. The causal status of grief is reiterated immediately after, when Galen turns to a case-history to support his point. In this context, the humoural explanation is complicated by the addition of various other explanatory elements:46 Most of all, as I have often stated, the dry and hot kraseis are seized by the hectic fevers right from the start, as happened in the case of the woman who, during autumn, began to be febrile due to sleeplessness and grief, extending for most of the winter.

There are a number of significant themes in this passage. Galen’s approach to fever builds on the notion of the mixtures (flagged by κράσεις) which inhere in individuals. In this context, the causal role of grief exacerbates an already present humoural proclivity towards fever present in those whose mixture is hot and dry. The reference to the humours and grief also intersects with ideas about the climatic conditions which prevail in certain seasons (and certain locations?) and, in due course, impact on the patient. The passage taps into a theme Galen develops particularly in his work Faculties of the Soul. There Galen argued that the mental conditions, especially grief and other pathē, followed not only the mixtures of the body, but were actively influenced by the various climatic, dietary, and life-style practices.47 Indeed, their capacity to shape the individual’s emotional life was such that Galen argues that one method of treatment for pathē is to attend to the dietetic elements of the individual patient’s life: food, lifestyle, and climatic conditions are key considerations in trying to shape the kraseis of the body and, in due course, the individual’s experience of certain emotional conditions.48 In On Therapeutic Method, the causal line is less explicit, but Galen builds, I suggest, on the same interpretative and explanatory framework: the woman, already inclined to fever, is driven to it because of her sleeplessness and her grief, and this occurs at the beginning of those seasons which are naturally suited to the development of fever. The connections that Galen continually seeks to draw for his readers and his patients make this significant. Faculties of the Soul, as Christopher Gill points

45 46

47

48

Epicurus and Zeno, namely that τὴν ... λύπην ... µείωσιν εἶναι φησιν ἐπὶ φευκτῷ ... (On Plato and Hippocrates v.367K). However this may be, it seems clear that Galen sees the concept of ‘shrinking’ or ‘contraction’ to be central to, at least, early Stoic thought: cf. the reference to αἱ µειώσεις and αἱ συστολαὶ at On Plato and Hippocrates v.367K, and the reference to Zeno’s definition, ibid. v.416K: ἡ τοῦ κακοῦ δόξα πρόσφατος µὲν οὖσα συστέλλει τε τὴν ψυχὴν καὶ ἐργάζονται λύπην. Galen, On Plato and Hippocrates v.416K. Galen, On the Therapeutic Method x.686K: µάλιστα δ’, ὡς εἴρηται πολλάκις, αἱ ξηραὶ καὶ θερµαὶ κράσεις ἁλίσκονται τοῖς ἑκτικοῖς πυρετοῖς εὐθὺς ἐξ ἀρχῆς, ὥσπερ καὶ ἐπὶ τῆς ἀνθρώπου, τῆς φθινοπώρου µὲν ἀρξαµένης πυρέττειν δι’ ἀγρυπνίαν καὶ λύπην, ἐπὶ πλεῖστον δὲ παρατεινάσης τοῦ χειµῶνος. Translation from Johnston and Horsley 2011. See above, p. 252 and notes 3–4. Galen, Faculties of the Soul iv.802—3K, 805K, 807K. For the concept as a leitmotif of other discussions of emotions, see also On Plato and Hippocrates v.464–5K. See above, p. 252 and notes 3–4.

264

Daniel King

out, might sit oddly within Galen’s theories about the psychē and its relationship to the body of the sufferer.49 What this example seems to suggest is that the explanatory model we find in Faculties of the Soul occupies an important place in Galen’s strategies for the explanation of sufferer’s conditions to patients and other doctors. The emotional life of the patient simply cannot be separated, in this example, from her humoural proclivities, and her consequent reaction to climatic conditions. Galen’s explanation of fever highlighted the capacity of emotional pathos to effect physical changes through the influence on the humoural condition of the body. Related to this model is the ability of grief to influence the condition of the individual more substantially through sunkopē. Galen defines sunkopē as the acute collapse of the ‘capacities’ (dunameis) which are maintained by either the ‘substance’ (ousia) or ‘mixtures’ (kraseis) of either the ‘pneuma or of solid bodies’.50 There are two important discussions of the experience in On Therapeutic Method and On Affected Places on which I want to focus in what follows. Galen’s interest in these discussions is in how psychical affections continue to influence the underlying qualities and harm the health of the individual. This continued development of grief’s status as a type of causal experience underpins a more concerted effort to locate the experience of grief amongst various patient experiences. According to On the Therapeutic Method x.842K, psychical affections have extreme effects on people’s health. Galen knows people who have died from both extreme pleasures and their opposite, extreme and sudden fears. This is due, Galen emphasises, to the way in which the psychic affections (τὰ πάθη τὰ ψυχικά) ‘render the pneuma weak and dissipated’ (... ἔκλυτον ἐργάζονται καὶ ἄτονον τὸ πνεῦµα); grief and other conditions damage and destroy the capacity:51 Grief, anguish, outbursts of strong anger and anxiety, and in a like manner frequent episodes of insomnia also cause harm by dissipating the capacity.

In this context, the soul moves itself, and, in due course, moves the body. Galen’s broad argument, here, is that doctors and patients need to attend to either maintaining or restoring the condition of the pneuma which has been undermined by various causes. Nevertheless, his adherence to the causal status of grief and other emotions continues to underpin his approach to medical conditions. The discussion of sunkopē at On Therapeutic Method x.842K draws on broader theories about how the structure of the soul and the body are combined. Arguably, the suggestion that psychical affectations release (ἔκλυτον) and render the pneuma weak (ἄτονον) recalls philosophical theories about the way in which the soul and the various elements of the individual are kept in tune with each 49 50

51

Gill 2010, 316–322. See above, p. 252 and note 5. For this particular definition, Galen, On the Therapeutic Method x.838K. For further discussion of this pathē, see Causes of Symptoms vii.252; cf. Siegel 1968, 347–352; Durling 1993, s.v. συνκοπή, ‘a sudden loss of strength’. Galen, On the Therapeutic Method x.842K: καὶ λῦπαι δὲ καὶ ἀγωνίαι καὶ θυµοὶ καὶ φροντίδες, ἐν οἵῳ τρόπῳ καὶ αἱ πλείους ἀγρυπνίαι, βλάπτουσι καταλύουσαι τὴν δύναµιν. Translation from Johnston and Horsely 2011.

Galen and Grief: the Construction of Grief in Galen’s Clinical Work

265

other. The reference to rendering the soul literally ‘without tonos’ is one case in point. It recalls Stoic theories about the way in which the different constituent elements of either different bodies or of pneuma are maintained in a delicate, tensile relationship that defines complex living beings.52 The reference to pneuma in this context also looks towards Galen’s broader theories of the various pneumata which inhere in the human body. The term picks up Galen’s explanation of the various pneumata that he presented only chapters earlier at On Therapeutic Method x.838Kf. The doctor’s theory of pneuma is built around a division between a psychic pneuma, which originates in the brain, a vital pneuma, which originates in heart, and physical pneuma, which is produced by the liver.53 In order to function properly, the individual needs to maintain the particular balance of all these three elements. As Heinrich von Staden has pointed out, parts of Galen’s conception of the varieties of pneuma are shared with a range of earlier medical and philosophical schools. Erasistratus, particularly, emphasised that the individual was maintained by both a vital and a psychic pneuma.54 Here, Galen suggests that psychical affections have the capacity to undermine the psychic pneuma and its delicate relationship with other forms of pneuma that inhere in the individual. Galen is combining, arguably, philosophical notions of the soul drawn from various points on the conceptual landscape to help elucidate the condition of sunkopē, and the need for maintaining one’s condition through dietetic practices. In this process, grief is attributed a causal role – serving to undermine the physical condition of the individual in a way that will require therapeutic action. Galen’s engagement with sunkopē is replicated in other discussions in On Affected Places. At viii.301Kf, he reminds us that other people have died from gastric sunkopē:55 Some people die in another manner from gastric syncope and under very severe pain as a result of terrible fright or extreme joy. For in those persons who have a low vital energy but suffer from strong emotional passions because of a lack of education, the essential mental faculties are easily impaired. Some of these patients die of grief, but not as fast as those mentioned above. However, a strong personality [literally: a man of great soul] will not succumb

52

53 54 55

Von Staden 2000, 100f. For Galen’s awareness of the Stoic notions of tonos, see Galen, On Plato and Hippocrates v.438; cf. Plen.vii.525K. For a brief explanation of tonos in Stoic thought, see von Staden 2000, 100f.; Algra 2009, 361–371. Galen, On the Therapeutic Method x.839K. For a discussion of the notion of pneuma in Galen’s theories of the soul and its relationship to the body, see Hankinson 2006, 234–238. Von Staden 2000, 111–113. On the connection with Erasistratus’ notion of the pneuma cf. 93–96. Galen, On Affected Places viii.301—2K: καὶ µὴν καὶ καθ’ ἕτερόν τινα τρόπον, ἐπί τε στοµαχικαῖς συγκοπαῖς, ἀλγήµασίν τε σφοδροῖς, καὶ φόβοις ἰσχυροῖς, ἡδοναῖς τε µεγίσταις, ἀποθνήσκουσί τινες· ὅσοις γὰρ ἀσθενής ἐστιν ὁ ζωτικὸς τόνος, ἰσχυρά τε πάθη ψυχικὰ πάσχουσιν ἐξ ἀπαιδευσίας, εὐδιάλυτος τούτοις ἐστὶν ἡ τῆς ψυχῆς οὐσία· τῶν τοιούτων ἔνιοι καὶ διὰ λύπην ἀπέθανον, οὐ µὴν εὐθέως ὥσπερ ἐν τοῖς προειρηµένοις· ἀνὴρ δ’ οὐδεὶς µεγαλόψυχος οὔτ’ ἐπὶ λύπαις οὔτ’ ἐπὶ τοῖς ἄλλοις ὅσα λύπης ἰσχυρότερα θανάτῳ περιέπεσον· ὅ τε γὰρ τόνος τῆς ψυχῆς αὐτοῖς ἰσχυρός ἐστι τά τε παθήµατα σµικρά. Translation from Siegel 1976.

266

Daniel King to grief or other passions stronger than grief since his mental strength great and his passions are small.

In contrast to his discussion in On Therapeutic Method, the causal role attributed to the pathē is, here, more nuanced, providing a specific focus on the way in which grief operates as a cause of sunkopē. There are a number of points to be drawn from Galen’s explanation for why the individual might suffer dissipation and ultimately death from grief. The first issue raised by Galen’s passage concerns the way in which grief is once again located within a hierarchy of emotional experiences: ... ὅσα λύπης ἰσχυρότερα ... returns us to the discussion of Causes of Symptoms in the last section, where we saw that grief was presented as a less dangerous form of emotion than fear or shame.56 The passage also elaborates further on the notion of the tonos and its connection with how individuals experience grief. In the first instance the reference to ὁ ζωτικὸς τόνος ups the stakes when compared to the passage we encountered in On Therapeutic Method, implying that Galen is more determined to point his readers to Stoic theories about the tonos that maintains the tensile relationship between the vital aspects of the individual.57 More importantly, in this context the treatment of tonos appears to connect the relationship between different elements of the soul, their ability to manage psychic pathē, and the individual’s moral qualities (flagged in this context by the contrasting qualities of ἀπαιδευσία and µεγαλοψυχία): those who lack education or who are ἀπαιδευσία, suffer more dramatically from emotional experiences such as grief. Galen’s use of these terms, arguably, looks towards a contemporary culture in which references to an individual’s paideia and ‘great-souledness’ were used as markers of elite, masculine, cultured identity. What are we to make of Galen’s willingness to write a moral layer onto the experience of the pathos? Read at one level, it constructs differing experiences of grief (and other emotional conditions) for those who could be, or wished to be, classified among the educated or the great-souled and those who were excluded from those moral and ethical categories. Simultaneously, the passage nuances the physicalist approach that I have been tracing throughout this paper: the experience of grief is not simply reduced to the level of the body, or of the humours, but is rather modified, shaped, and ameliorated by fundamentally moral qualities. Over the course of this section, I have tried to investigate how grief is presented in connection with other physical conditions and experiences. The intention was to try and understand how and where Galen (and, potentially, his patients and other doctors) mapped the experience of grief onto the complex and variegated landscape of illness in the late second century CE. We have seen that it has primarily been associated with fever and the experience of sunkopē as an initial cause of these conditions. We have drawn out a number of conclusions. Firstly, that throughout Galen’s medical or clinical work he combines, and repackages, a range of medical and philosophical ideas about the body and the soul. This is 56 57

See above p. 257 and notes 20–21. For ὁ ζωτικὸς τόνος as a Stoic term, see SVF II, 235.

Galen and Grief: the Construction of Grief in Galen’s Clinical Work

267

hardly surprising for a figure such as Galen who claimed that the best doctor is also a philosopher.58 But it does raise questions about how we might approach the dissemination of philosophical and medical thought in this period. Secondly, we have also developed a context through which we can understand Galen’s anecdote about the death of Philippides in Freedom:59 Galen’s anecdote bites because according to his medical system it was a possibility that individuals could die from the pathos. 4 CASE-HISTORIES AND GRIEF Over the last two sections, I have emphasised how Galen’s medical concerns inflect his approach to grief. In both the Causes of Symptoms and On Plato and Hippocrates, as well as in the medical manuals On the Therapeutic Method and On Affected Places, grief is presented through a framework that connects it with either other forms of pain or other forms of illness (such as fever or the experience of sunkopē). As we have discussed, this has significant implications for how grief is used to help explain and understand the experiences of Galen’s patients; and, as a consequence, how they located (and we map) grief onto the landscape of emotional and illness experience. In this section, I look further at some of Galen’s case-histories drawn from across his corpus in order to build a broader understanding of how Galen approaches the pathos. In these instances, the model of grief experience that I have been developing is nuanced by Galen’s presentation of grief as something that originates from a mental idea. However, rather than simply emphasising the mental aspects of grief, Galen continues to imagine the emotional experience in terms of the interplay between psychic and somatic condition, pain, and illness. A considerable number of Galen’s references to grief emerge from more explicitly philosophical works: we have already touched briefly on Galen’s quotations of different Stoic definitions of grief as the ‘fresh appearance of evil’60 throughout On Plato and Hippocrates. In The Passions of the Soul also, Galen’s treatment of grief emphasises the connection between the pathos and one’s (false?) judgement of various situations. Galen insists there, for instance, that there is ‘one cause of grief for all the Greeks, namely insatiable desire’ (ἀπληστία).61 The discussion is reinforced by Galen’s assertions that he has seen ‘many people’ (ὁρῶ τοὺς πολλούς) ‘grieving’ (λυπουµένους) when they want to be honoured by someone, or when they have suffered losses to their property

58 59 60

61

Galen wrote a treatise entitled The Best Doctor is also a Philosopher (Opt. Med. i.53–63K). Galen’s combination of philosophical and medical thought is well-known. See above, p. 251 and note 1. Cf. Galen, On Plato and Hippocrates v.366Kf for Chrysippus’ definition. There are a number of different quotations: cf. Zeno: δόξαν γὰρ εἶναι πρόσφατον τοῦ κακὸν αὐτῷ παρεῖναι φησι τὴν λύπην. Galen, Passions of the Soul v.45K, 49K.

268

Daniel King

(χρηµάτων ἀπωλείᾳ).62 The focus that inheres in these contexts is repeated throughout a number of case-histories. This is particularly true in Galen’s autobiographical work On Prognosis. Here, Galen refers to a number of case-histories in which he treats individuals’ experiences of grief. One such example, the case of a nameless slave, is particularly telling in its focus on the connection between grief, the slave’s mental ideas or assumptions, and his physical condition. Galen begins by telling us that he found the man ‘wasting away through grieving’ (λυπούµενος ἐτήκετο); he was ‘distressed’ (λυπούµενος) because he was about to give an account of his master’s money and knew that some would be found missing, and he was ‘sleepless because of this thought’ (ὑπὸ τῆς φροντίδος ἠγρύπνει). 63 The focus on the thoughts of the slave is brought home repeatedly throughout this passage. Galen tells us moments later that he informed the slave’s master that he did not have a ‘physical condition’ (σωµατικὸν πάθηµα), but simply that he was ‘afraid’ (φοβεῖτο). Galen’s account, once again, reiterates the connection between the experiences of sleeplessness, fear, and grief which we have seen throughout this paper. What matters in this context, however, is that Galen’s approach is primarily directed at the mental assumptions of the sufferer. He advises the master to tell the slave he will not be worried if any money might not be accounted for; in due course, the patient ‘becomes free from grief’ (γινόµενος ἄλυπος) and three days later he recovers his normal physical condition (τὴν κατὰ φύσιν ἕξιν τοῦ σώµατος ἀνεκτήσατο).64 This particular case-history is directed towards proving the inter-connection between the individual’s mental state and their pulses – it is part of a long disquisition on Galen’s ability to diagnose, following Erasistratus, the patient’s mental conditions on the basis of their physical appearance or state.65 The reference to the experience of wasting away recalls Galen’s broader understanding of, and interest in, the physical experiences associated with the pathos. That said, however, the focus on grief downplays the broader physical aspects of the experience. The implications that are subtly developed in On Prognosis receive greater treatment in other Galenic case-histories, in which both the mental and physical aspects of the experience are combined. This is particularly true in some of the cases treated in Galen’s On Hippocrates Epidemics VI, which presents a slightly different approach. This work is a problematic source for Galen’s medical views: it survives primarily in Arabic and it is, of course, difficult to separate the Arabic commentator or translator from Galen.66 Nevertheless, the case-histories contained in it do provide some valuable historical examples of individuals who suf62

63 64 65 66

Galen, Passions of the Soul v.45K; for the themes of aplēstia and the inability to deal with apōleia, see a common theme across Galen’s consolatory work: cf. Galen Freedom, especially 46–50. Galen, On Prognosis xiv.633K. Galen, On Prognosis xiv.634K. Galen, On Prognosis xiv.634—5K. The original Greek of much of On Hippocrates Epidemics VI has been lost. I have used here Pfaff’s German translation of Wenkebach’s edition of the extant Arabic treatise.

Galen and Grief: the Construction of Grief in Galen’s Clinical Work

269

fer from grief. In the context of discussing Hippocrates views on ‘mental activity’, Galen suggests that he knows a number of people who have suffered from grief.67 Galen’s primary concern in this context is to illuminate how ‘mental activity’ has played a role in the emergence of grief and led, eventually, to the patient passing away. One case in point is that of the augur Maiandros. The history of Maiandros is recounted at some length, in which the details of his illness, its cause, and course are narrated: Maiandros, every year at the time of his birthday, attempts to receive from the auguries an indication of his own fate for the coming year.68 On the particular year in question, he receives a sign that he will die within the next twelve months. The consequences of this revelation are profound. According to Galen, after returning to the city, he became depressed, and was afflicted with insomnia, and he began to waste away.69 At length, the condition worsened: he began to suffer fevers and eventually was so confused he became mad and died. My paraphrase of Galen’s account (which I have taken from the German translation of the Arabic) of Maiandros’ fate presents a number of critical connections with the On Prognosis cases. Maiandros’ condition clearly begins from an activity that can be seen as mental – Galen ascribes it to a fear of death. Such fear, along with its capacity to produce insomnia and grief, certainly recalls the treatment of the pathos in other contexts. Indeed, it is ‘mental activity’ which leads to the condition and he goes on to emphasise that the art of therapy, in this situation, consists in discovering and removing the cause of grief or suffering which lies in the initial thought or in the mind of the individual. Cure, here, cannot be achieved by food or physical remedies, through dietetics or clinical practice, but only by removing the ‘fixed idea’ that the individual has.70 Unlike the On Prognosis examples, however, Galen also emphasises the connection between these pathē and their consequent physical experiences: fear, insomnia, and grief intersect with the experience of fever; the ‘wasting’ which was referred to in the slave’s casehistory is, here, mentioned with a higher degree of specificity about the physical consequences that grief sufferers experience: the mental activity with which Galen is no doubt concerned is embedded in a broader approach to understanding the emotional experience. Although it is impossible to investigate how Galen imagines this connection, the collocation of these different symptoms is suggestive of the model discussed in Section Three, above. In these contexts, grief, itself, might well be understood as a type of illconceived ‘fixed idea’. Yet, the experience of that pathos is conceptualised as part of a complex matrix of symptoms, further conditions, and experiences, of which grief is both a cause and a consequence. These two aspects of Galen’s approach to grief are not necessarily contradictory, especially when considered in the context 67

68 69 70

On Hippocrates Epidemics VI = CMG v.10.2.2, 485.5–487.17 for a number of case-histories. See especially, On Hippocrates Epidemics VI = CMG v.10.2.2, 486.19–26 for the case of Kallistos, who died from grief over the loss of his books. On Hippocrates Epidemics VI = CMG v.10.2.2, 485.35–486.10. On Hippocrates Epidemics VI = CMG v.10.2.2, 486.10–486.12. On Hippocrates Epidemics VI = CMG v.10.2.2, 487.18–487.22; cf. Pfaff’s use of ‘fixe Idee’ at 487.22.

270

Daniel King

of Galen’s multi-faceted corpus. However, they do suggest that patients who experienced grief in the late second century CE might negotiate and understand that experience in ways that go beyond the approach presented in explicitly philosophical contexts, depending on their outlook, the type of treatment they sought, and the narratives they encountered in a clinical or philosophical context. Put more broadly, grief (and its attendant emotional and physical experiences) in Galen’s medical work at least – and, as much as his corpus gives an indication of this, in certain historical contexts – was explained, conceptualised, and understood as part of a complex landscape of experiences. 5 CONCLUSIONS Over the course of this paper, I have tried to achieve a number of things. Scholarship on Galen’s approach to grief and other pathē has emphasised either his physicalist approach, or the view that he treats psychological affections as mental problems. I have tried to elucidate a middle path between these two views by investigating how Galen understands, and categorises, the experience of grief alongside his patients. In Section Two, I argued that Galen positioned himself against various philosophical approaches to grief by locating the emotion amongst a family of pain experiences, rather than differentiating sharply between lupē and other forms of distress. In Section Three, I showed that Galen’s practical medical texts advocated an approach that, as well as drawing heavily on philosophical ideas about grief, the soul, and its relationship to the body, emphasised the connection between grief and other conditions such as fever or sunkopē. Grief, here, was not simply a mental pathos but an active cause and effect of more serious illnesses and physical conditions. Finally, in Section Four, I tried to contextualise some of these conclusions by turning to a number of Galen’s case-histories, which appear to present contradictory approaches to the pathos. I showed how some of Galen’s case-histories combined the notion of grief stemming from a fixed idea, with the need to contextualise the experience within a range of broader physical and emotional consequences. Galen’s anecdote about Philippides, was taken from an overtly philosophical (and consolatory) text, has meaning – it bites – well beyond that context. It does so because of Galen’s (and arguably his culture’s) broader understanding of the physical and emotional consequences of grief: according to Galen’s medical system, being grieved over losing your books could kill you because of the way he and others constructed the experience of grief.

Galen and Grief: the Construction of Grief in Galen’s Clinical Work

271

BIBLIOGRAPHY Algra, K. (2009) Stoics on Souls and Demons: Reconstructing Stoic Demonology, in D. Frede and B. Reis (eds.), Body and Soul in Ancient Philosophy, Berlin/New York, 358–389. Amneris, R. (1989) Epidemics and Aphorisms: Notes on the History of Early Transmission of Epidemics, in G. Baader and R. Winau (eds.), Die Hippokratischen Epidemien: Theorie – Praxis – Tradition (Sudhoffs Archiv, 27), Stuttgart, 182–190. Barnes, J., J. Jouanna, and V. Barras (eds.) (2003) Galien et la philosophie, Geneva/Vandœuvres. Boudon-Millot, V. (2007) Une traité perdu de Galien miraculeusement retrouvé, le Sur l’inutilité de se chagriner: texte grec et traduction française, in V. Boudon-Millot, A. Guardasole, and C. Magdelaine (eds.), La science médicale antique: nouveaux regards, Paris, 67–188. Brunschwig, J. and M. Nussbaum (eds.) (1993) Passions & Perceptions: Studies in Hellenistic Philosophy of Mind, Cambridge. De Lacy, P. (1981) On the Doctrines of Hippocrates and Plato (Corpus Medicorum Graecorum v.4.1.2), Berlin. ––– (1988) The Third Part of the Soul, in P. Manuli and M. Vegetti (eds.), Le opere psicologiche di Galeno, Naples, 43–63. Donini, P. (2008) Psychology, in R. Hankinson (ed.), Cambridge Companion to Galen, Cambridge, 84–119. Durling, R. (1993) A Dictionary of Medical Terms in Galen (Studies in Ancient Medicine, 5), Leiden. Frank, A. (1995) The Wounded Storyteller: Body, Illness, and Ethics, Chicago / London. Fitzgerald, J. (2008) Passions and Moral Progress in Greco-Roman Thought, London / New York. Frede, D. and B. Reis (eds.) (2009) Body and Soul in Ancient Philosophy, Berlin / New York. Garcia Ballester, L. (1988) Soul and Body, Disease of the Soul and Disease of the Body in Galen’s Medical Thought, in P. Manuli and M. Vegetti (eds.), Le opere psicologiche di Galeno, Naples, 117–152 Gill, C. (2007) Galen and the Stoics: Mortal Enemies or Blood Brothers?, Phronesis 52.1, 88–120. ––– (2010) Naturalistic Psychology in Galen and Stoicism, Oxford. Hankinson, R. (1991) Galen’s Anatomy of the Soul, Phronesis 36, 197–233. ––– (1992) Galen’s Philosophical Eclecticism, in Aufstieg und Niedergang der römischen Welt II.36.5, Berlin, 3505–3522. ––– (1993) Actions and Passions: Affection, Emotion, and Moral Self-Management in Galen’s Philosophical Psychology, Brunschwig and Nussbaum (eds.) 1993, 184–222. ––– (1994) Galen’s Theory of Causation, in Aufstieg und Niedergang der römischen Welt II.37.2, Berlin, 1757–1774. ––– (2003) Causation in Galen, in Barnes, Jouanna, and Barras (eds.) 2003, 31–66. ––– (2006) Body and Soul in Galen, in R. King (ed.), Common to Body and Soul, Berlin, 232– 258. ––– (ed.) (2008) The Cambridge Companion to Galen, Cambridge. Harkins, P. (1963) Galen on the Passions and Errors of the Soul, Ohio. Johnston, I. (2006) Galen on Diseases and Symptoms, Cambridge. Johnston, I. and G.Horsley (2011) Galen’s Method of Medicine, Vols. I-III, Cambridge, Mass./ London. Kleinman, A. (1988) The Illness Narratives: Suffering, Healing and the Human Condition, New York. ––– (1989) Patients and Healers in the Context of Culture: an Exploration of the Borderland Between Anthropology, Medicine, and Psychiatry, Berkeley / London. Manuli, P. (1988) La passione nel De Placitis Hippocratis et Platonis, in P. Manuli and M. Vegetti (eds.), Le opere psicologiche di Galeno, Naples, 185–214. Mattern, S. (2008) Galen and the Rhetoric of Healing, Baltimore.

272

Daniel King

Siegel, R. (1968) Galen’s System of Physiology and Medicine: An Analysis of His Doctrines and Observations on Bloodflow, Respiration, Humors and Internal Diseases, Basel. ––– (1976) Galen on the Affected Parts: Translation from the Greek Text with Explanatory Notes, Basel. Sorabji, R. (2000) Emotion and Peace of Mind: From Stoic Agitation to Christian Temptation, Oxford. Staden, H. von (2000) Body, Soul, and Nerves: Epicurus, Herophilus, Erasistratus, the Stoics, and Galen, in J. Wright and P. Potter (eds.), Psyche and Soma, Oxford. Tieleman, T. (1996) Galen and Chrysippus on the Soul: Argument and Refutation in the De Placitis, Books II-III, (Philosophia Antiqua), Leiden. ––– (2003a) Chrysippus’ On Affections: Reconstruction and Interpretation, Leiden. ––– (2003b) Galen’s Psychology, in Barnes, Jouanna, and Barras (eds.) 2003, 131–161. ––– (2008) Methodology, in Hankinson (ed.) 2008, 49–65. Wenkebach, E. and F. Pfaff (eds.) (1956) Galeni in Hippocratis Epidemiarum librum VI Commentaria I-VIII (Corpus Medicorum Graecorum v.10.2. 2), Berlin.

EMOTIONALITY IN GREEK ART Olympia Bobou 1 INTRODUCTION1 One of the main characteristics of Greek art is the lack of visible emotions in most artworks. The ‘High Classical’ style, exemplified by the Parthenon sculptures and other works that can be associated with Pheidias, is characterised particularly by the calmness and serenity of the gods and mortals portrayed,2 and its influence can be seen in Hellenistic artworks such as the Athena from the Pergamon library,3 the sculptural group from Lykosoura,4 and even in Roman art, as, for example, the Ara Pacis at Rome.5 In this paper, though, I want to explore the cases where emotions were clearly shown, and investigate the representations of emotion in Greek art: (a) how are emotions expressed, (b) what emotions are shown, (c) who is showing emotions, (d) can the emotions shown be categorised as positive or negative, and (e) when are emotions used in Greek art? 2 HOW ARE EMOTIONS EXPRESSED? 2.1 Knowledge of narrative context We find three different ways of representing emotions in Greek art. The first is through knowledge of the narrative context of a scene or art work. For example the Iliad starts with the word ‘anger’, clearly stated as the topic of the poem. Achilles’ anger, first against Agamemnon and his fellow Achaeans, then against Hector, is the topic and the motivating force throughout the Iliad, but he is not the only one of the heroes to experience anger within the poem. The emotion is de1

2 3 4 5

I would like to thank Drs Julia Lenaghan and Maria Stamatopoulou for discussing aspects of this paper with me, Professor Bert Smith for having a look at an earlier draft, and Dr Jonah Rosenberg for helpful remarks. For the classical style in sculpture, see Ridgway 1981; Boardman 1985; Rolley 1999. Modelled after the Athena Parthenos. For the statue, see Smith 1991, 156, fig. 185; Grüßinger, Kästner, and Scholl (eds.) 2011, 559f., cat. nο. 8.2. For the Lykosoura group, see Smith 1991, 240f., figs. 301.13; Kaltsas 2002, 279–281, cat. nos. 584–591; Schraudolph 2007, 190–197, 397, figs. 174k–m. For the Ara Pacis, see Rossini 2007; for the connections between the Ara Pacis and Greek art, especially the Parthenon, see Borbein 1975, 242–266; Kleiner 1978, 753–785; Conlin 1997.

274

Olympia Bobou

scribed as all-consuming, at times righteous or unreasonable, and making heroes face the dilemma of striking out or controlling themselves to the point that divine intervention is required for them to make a decision; for example, in a scene in the Iliad (1.188–222), Athena’s intervention stops Achilles from killing Agamemnon for insulting him. When Achilles finally resumes fighting and goes after Hector, he is a man full of bitter rage (χόλος),6 listening to neither pleas nor reason until he kills Hector. Only then does he stop and retreat to the Greek camp.7 How is this anger of Achilles expressed in Greek art? Following the Iliad, the most obvious example is that of the fight between Hector and Achilles, and there are numerous depictions of this subject.8 One of the most typical ways is to show the two facing each other, with Achilles advancing towards the wounded Hector,9 using the iconography of duelling hoplites.10 A depiction by the Berlin Painter is a good example of this motif:11 the inscriptions identify the two heroes, and locate them securely in the land of myth, separating them from all the other hoplites populating Greek vases at the time (Figure 1). But, unlike the textual Achilles, the Achilles in the Classical images usually shows no emotion: his face betrays nothing; his posture shows a man in control of his body and his actions. None of the cholos, the bitter anger that drives Achilles throughout the Iliad, is present.

Figure 1. Achilles fighting Hector. Red-figure volute-krater by the Berlin Painter, c. 490-460 BCE. British Museum. 6 7

8 9 10 11

For example, Iliad 15.68, 18.337. For cholos see Konstan 2006, 48–56. It is important that the Iliad starts exactly with that emotion: anger (menis), Iliad 1.1; Achilles’ anger against Hector spreads over the second half of the poem, starting from the 18th book. See also Harris 2001, 131–156. See Kossatz-Deismann 1981, esp. 133–138, nos. 558–582 (LIMC I, s.v. Achilleus). Kossatz-Deismann 1981, nos. 558–570. For hoplite and iconography, see van Wees 2000, 125–166; Hanson 2000, 201–232. London, British Museum 1848.8–1.1, Beazley Archive Pottery Database no. 201941: www. beazley.ox.ac.uk/record/6827863E-3495-493D-9AFB-80952FBBD3D9 (accessed on 4 June 2012).

Emotionality in Greek Art

275

However, by identifying the two hoplites as Achilles and Hector, the painter and the viewer acknowledge the text, and by doing so, they understand this scene to be highly emotional. This is not just an advancing warrior, striding forward with purpose; this is Achilles recklessly seeking revenge. This is not just a wounded hero, fighting adrenaline-fuelled to the end; it is Hector, angry at what he recognises as divine betrayal and wanting to die honourably, performing one last great deed. Knowledge of the text transforms the viewer’s reaction to the scene. 2.2 Gestures A more direct way to show emotions is through the use of gestures. In the famous hydria by the Kleophrades Painter showing the Sack of Troy, we see a number of vignettes all taking place during the fall of the city. On one side of the vase, we see Aeneas escaping with his aging father and his young son, and Ajax, son of Oileus, about to rape Cassandra, as well as two women sitting on the ground, pulling their hair. A palm tree divides this scene from the next, where we see Priam, wounded, seated on an altar with the body of Astyanax, his grandson, on his lap. He covers his face with his hands, and does not try to attack the youthful Neoptolemos. The next vignette shows a Trojan woman attacking a Greek man with a pestle (Figure 2). The last image is that of an older woman being led away by two young men, Aithra saved from servitude by her grandsons Demophon and Akamas.

Figure 2. The Sack of Troy. Hydria by the Kleophrades Painter, c. 480 BCE. Naples, Museo Archeologico Nazionale.

276

Olympia Bobou

In these scenes, gestures and composition are used both to represent and to evoke emotions. Priam, old, bloodied, defeated in every sense, covers his face and head. His gesture is one of a man trying to defend himself against his assailant, and the way he is portrayed marks him out as a most pathetic figure, worthy of pity. Yet, what we can see of the man shows him to have a serene expression. The two women behind the palm tear their hair out and hit their heads, using gestures associated with mourning and grieving in Archaic art.12 The Greek victors/assailants are also shown without any discernible emotion. There is no joy in their accomplishment, or lust for flesh or blood. They are calm, methodical, efficient. They look more like statues than men, and a comparison between the figure of Neoptolemos and the warrior statue in the nearcontemporary Foundry Cup (Figure 3) shows how similarly the two are viewed by the vase painters.13

Figure 3. Sculptors working on a statue. Cup by the Foundry Painter, c. 490–480 BCE. Berlin, Pergamon Museum. © Johannes Laurentius.

12

13

Naples, Museo Archeologico Nazionale 81699. Beazley Archive Pottery Database no. 201724: www.beazley.ox.ac.uk/record/5E80388C-3F0B-402B-8AAD-F7D3D71532CF (accessed 30/06/2012). Berlin, Antikensammlung F2294. For the Foundry Cup see Beazley Archive Pottery Database no. 0434: www.beazley.ox.ac.uk/record/A5158FBE-FC09-4880-82AB-3A4C1A699D15 (accessed on 30 June 2012).

Emotionality in Greek Art

277

2.3 Facial expressions The use of facial expressions is the most direct and explicit way of showing the emotions of figures.14 In the Archaic period we see the smiling faces of kouroi and korai,15 but we do not see any other representations of emotion, while in the Classical period, even though the ‘Archaic smile’ disappears, we see a range of emotions making an appearance, on which we will focus later. These are conveyed through the use of two principal markers: the furrowed brow and the open mouth, to which several others can be added. It is interesting to note that we see emotions on display in three of the major monuments of the Classical period. On the temple of Zeus at Olympia we see several figures showing emotion: Herakles in the metopes, Lapiths and centaurs in the west pediment, and Oinomaos in the east pediment, as well as other figures.16 The identification of their emotions, though, is difficult. For example, we see that Herakles has a furrowed brow in three of the metopes, the first (the lion of Nemea), the third (the birds of Stymphalos), and the twelfth (the stables of Augeias). The slightly open mouth is apparent in all the metopes, but body language, postures, and gestures show that it has a variety of functions: from indication of speech (metope 3), to sign of weariness (metope 1, Nemean lion), and marker of physical exertion (metope 4: Cretan bull; metope 5: Keryneian hind; metope 6: girdle of Hippolyte; metope 9: fight with Geryon; metope 11: capture of Cerberus; metope 12). Herakles’ parted lips also serve as a differentiating marker between the hero and Athena, who is represented with divine calmness and closed lips in all the metopes, except the third, where her slightly open mouth is an indication of speech. Considering also the height of the metopes, the slightly parted lips emphasise the mouth and make it a focal point for the viewer’s attention. At Athens, we see representations of emotions in the metopes of the South side of the Parthenon, though the state of preservation does not allow us to judge the extent of emotional displays on the monument.17 The metopes show a Centauromachy, and in most cases where the heads of either Lapiths or centaurs are preserved, we can see that they are showing emotions: anger, rage, fear.18 Emotions are also shown in the sculptural decoration of the temple of Apollo at Bassai, and they are similarly located at the Centauromachy. Most of the centaurs show anger or, sometimes, pain, while their victims, the attacked women, 14 15 16 17 18

On emotions and facial expressions, see Ekman and Friesen 2001, 255–264; cf. Masséglia 2012a, 124f. On kouroi and korai, see Richter 1968 and 1970; Rolley1994; Meyer and Brüggemann 2007. An interesting interpretation of the Archaic smile can be found in Stiebert 2004, 52–54. Ashmole and Yalouris 1967, 8–29. Ashmole sees the Olympia sculptors as the first to show emotion on art: Ashmole and Yalouris 1967, 11. A very selective bibliography on the monument: Brommer 1979; Jenkins 2006, 71–107 and 2007; Delivorrias 2008. For the metopes the most useful publication remains Brommer 1967. The centaurs show emotions in the metopes S1, S2, S4, S5, S7, S8, S9, S16, S26, S29, S30, S31, S32. The Lapiths show emotions in the metopes S4, S9, S16, S27, S30, S31, and it seems likely that most of the figures in the South metopes showed emotions.

278

Olympia Bobou

and their defenders, the Lapith men, are shown fighting back, gesturing wildly or fighting with vigour, but of them only one Lapith man is shown in pain.19 In all these figures we see use of the same visual vocabulary as in the Olympia sculptures, that is, use of furrowed brow and open mouth, more often than not showing the teeth, only more refined and developed. Unlike the Olympia figures, it is possible to discern two emotionally-laden expressions showing pain and anger, without needing explicatory postures and gestures. 3 EXPRESSIVITY VERSUS REPRESENTATION OF EMOTIONS The difficulty of identifying emotions in Greek art is complicated by the fact that the same signs, or markers, are also used in order to show expressivity. Andrew Stewart in his examination of the Tegea sculptures identified the following ‘formal devices’, as he calls them,20 which were used for the display of negative emotions, from the time of the Olympia sculptures: — twisted mouth, open in various degrees and showing, or not, the teeth; — flared nostrils; — eyes that are fixed on an object, wide open or closed; — lined forehead and furrowed bridge of the nose. He comes to the conclusion that their use was ‘eclectic’, the sculptors preferring the effect of emotion rather than physiognomical correctness.21 Stewart’s discussion hints at a problem we have already seen when trying to identify the emotions on the figures from Olympia: whether any display of facial expression is to be understood as a representation of emotions. For example, the term ‘pathetic expression’, most often employed for figures with deep-set eyes fixed on a specific point, open mouth and furrowed brow,22 characterises the figure’s expression as an emotional one, but some examples can show how expressivity is different from representation of emotions. Perhaps the sculptor traditionally most frequently associated with ‘pathetic expression’ in his works is Skopas. We know that he worked at the temple of Athena Alea at Tegea, and the sculptures that were found there were, as Andrew Stewart puts it, ‘carved in one style, but representing different stages of development within it’.23 Stewart’s conclusion is that even though the result ‘would ...

19

20 21 22 23

For the temple and sculpture, see Cooper 1996; Jenkins 2006, 130–150. For the sculptures, see Madigan 1992. Emotions are shown by the centaur attacking the Lapith men in plaques BM 521, BM 522 (Madigan 1992, pl. 42), the centaur attacked on plaque BM 524 (Madigan 1992, pl. 43), both centaurs on plaque BM 526, the centaur on BM 527 (Madigan 1992, pl. 44), one of the centaurs attacking Kaineus on plaque BM 530 (Madigan 1992, pl. 45); but only the Lapith young man bitten by the centaur on plaque BM 527 is showing pain. Stewart 1977, 74. Stewart 1977, 74f. Boardman 1995, 15. Stewart 1977, 73.

Emotionality in Greek Art

279

have been striking enough’, understanding the emotional state of the figures was still subject to seeing their postures, stance, and gestures.24 Indeed, the faces of the Tegea sculptures show tension, general disquiet, but perhaps the most prevalent expression is that of focus on an object or opponent. The East pediment represents the hunt for the Kalydonian boar and the West the battle between Telephos and Achilles on the plain of the Kaikos. Both situations can be the causes of intense emotional states; however, the figures are shown tense and focused as if they were chess players rather than hunters about to kill their prey, or warriors facing each other on a battlefield, whose lives are at stake. There are no grimaces, pronounced muscle movements or other signs of intense physical activity that causes tension. Another example for the use of markers in order to show expressivity rather than emotions can be seen in the Roman copies of the works of Lysippos. One of the other major sculptors of the fourth century, Lysippos is usually associated with depictions of emotions, from his ‘weary Herakles’ to his conquering, almost divine Alexander.25 From the copies we see that Lysippos showed tension and focus through the use of furrowed brow, deep-set eyes, and half-open mouth, as for example in the Herakles Farnese,26 and differentiation between the size of the eyes.27 However, the heroes’ expressions may be tense, but the emotions of weariness or confidence are shown through the stance of the figures (slumped shoulders and lowered head for the Herakles, upturned head for Alexander). These elements in different combinations of open or half-open mouths, knitted brows with or without ridges at the top of the nose, and deep-set eyes, are used also in order to individualise portraits in the fourth century.28 In one of the latest discussions of the topic, Johannes Bergemann does not use the word ‘emotion’; he writes of physiognomies and types, realism and likeness, even though, as we have seen, these are the same markers used for showing emotions. Another case where we see use of emotional markers such as the upturned gaze and a slightly deeper groove between the lips, giving the impression of a barely-open mouth, is in the portraits of rulers. Starting from Alexander the Great, these markers are used in portraits of youthful rulers who want to emulate visually the Macedonian king throughout the Hellenistic period: Demetrios Poliorketes, Attalos I, and Mithridates VI of Pontos are the most prominent rulers employing this particular iconography.29 These devices are also used in the so-called ‘Terme Ruler’.30

24 25 26 27 28 29 30

Stewart 1977, 75. For the portraits of Alexander by Lysippos, see Stewart 1993; Moreno 1995, 148–165. Moreno 1995, cat. nos. 4.14. 1 and 4. Eros of Thespiai: Moreno 1995, cat. nos. 4.15.1–10; Apoxyomenos: Moreno 1995, cat. nos. 4.29.4. See Bergemann 2007, 34–46. For portraits of Alexander the Great, see Bieber 1964; Stewart 1993. For ruler portraits see Smith 1988. For the Terme Ruler, see Meyer 1996, 125–148.

280

Olympia Bobou

Perhaps instead of trying to distinguish between the two (realistic features and emotional markers), we should instead see the two elements as intentionally blurred. The portraits, either honorific or on funerary reliefs, look like real people, and real people have emotions. A touch of grief is as appropriate to the depiction of an old man mourning his son – for instance, in the Ilissos stele. 31 It is expressed through the furrowed brow and slightly parted lips, in the same way as a philosopher is shown with a concentrated, focused expression (for example, the philosopher in the type Capitol-Naples 6134).32 Again the gesture of the old man and his slightly stooped posture differentiate the two, from grieving father to introspective philosopher. 4 EMOTIONS AND REALISM We will return later to expressivity and whether, if under certain conditions it can be also seen as emotionality, that is, displays intended to arouse the viewers’ emotions, but for now let us look one more time at how emotions are shown. As said, the most direct way is through the use of facial expressions, but it is interesting to see how these expressions are shown from a sculptural perspective. In the early Classical art at the temple of Zeus at Olympia, the lined forehead of Herakles in the metopes does not correspond to the movement of the frontalis muscle.33 The same can be observed in the case of the Lapiths in the Parthenon metopes. For example, when the young Lapith in South Metope 30 reacts in pain to the centaur’s pulling him by the hair,34 he shows a lined forehead, ridges at the bridge of the nose, eyebrows angled upwards, and twisted, half-open mouth. These are all markers of the muscular movements, but they are not physiognomically correct depictions of such movements. This is especially obvious in the areas of the forehead and the mouth: when the corrugator supercilii moves, the eyebrows are lowered and ridges are created at the bridge of the nose; on the Lapith’s head, instead, the eyebrows are raised. The upper lip is raised, indicating movement of the levator labii superioris, but the nostrils are not turned upwards, as it happens when the levator labii superioris moves in life. As Stewart observes for the Tegea figures, the artists choose what will give the impression of emotion, rather than show an accurate depiction of it.35 The effect of the Olympia or Parthenon figures is more mask-like, while that of the Tegea figures is hyper-real. These non-realistic approaches are abandoned in the Hellenistic period where we see proper physiognomical expressions of emotions.36 For example, laughter 31 32 33 34 35 36

Kaltsas 2002, 193f., cat. no. 382. Bergemann 2007, fig. 25. http://www.artnatomia.net/uk/index.html (accessed 18/04/2012) was invaluable at seeing the exact movement of the muscles. Brommer 1967, 124f., figs. 229–232. Stewart 1977, 74f. For Hellenistic art and anatomy, see Amberger-Lahrman 1996; for physiognomically correct depictions of emotions, see Queyrel 2005, 170–173.

Emotionality in Greek Art

281

involves the movement of several muscles: the corrugator supercilii, orbicularis oculi, and procerus in the area of the bridge of the nose and the eyes, the levator anguli oris, levator labii superioris, zygomaticus minor and major in the area of the mouth, and the masseter in the area of the jaws. All these muscles can be seen tensed in the statue of the Drunken Old Woman (see p. 295), as well as in the figures of Satyrs laughing. Pain is expressed through the motion of the frontalis, as well as the corrugator supercilii and the levator labii superioris. Laokoön and his sons are textbook images of how this emotion is transferred and expressed in art.37

5 WHAT EMOTIONS ARE SHOWN IN GREEK ART? 5.1 Introduction We have already seen some of the emotions shown in Classical art, but the focus in this part will be on artworks from the Hellenistic period, where emotions are shown in a clear way through facial expressions, and they do not have to be explained by the figures’ posture or the gestures. There is great scholarly debate about the origin and categorisation of emotions. For the purposes of this paper, I am leaving aside the question of origins, and I am accepting the theory of basic and subordinate emotions.38 According to this theory, emotions can be arranged hierarchically, with intensity of the emotion and the context in which the emotion originates as the two determining factors for their place in the emotional hierarchy.39 The basic emotions identified are fear, sadness, anger, joy, and love, which they were then divided into subcategories (for example, rage, pleasure, and so on).40 5.2 Survey of emotional displays By focusing on works from the Hellenistic period where emotions have been identified we can start answering questions about the use and significance of emotionality in Greek art. A visual survey is necessary in order to have a general overview of the monuments and figures that show emotion (Figures 4–18). From this survey, several things become apparent: first of all, the emotions shown are fairly limited, since, even by adding the different subordinate emotions in our list, we can come up with ten emotions.41 The following emotions can be grouped under the heading ‘fear’: anxiety, distress, and worry. The Persians and Gauls of the Lesser Attalid Group (Figures 4, 5, and 8) show the emotion at its most intense: they are afraid. Ajax from the Pas37 38 39 40 41

For example, see Morelli and Morelli 1977, 369 and fig. 103; Schupbach 2009, for the enduring fascination with Laokoön’s pain. Shaver et al. 2001, 26–56. Shaver et al. 2001, 37f. Shaver et al. 2001, 43–47. Out of over 100 emotions in Shaver et al. 2001.

282

Olympia Bobou

quino group (Figure 7), and Odysseus and his companion from the Blinding of Polyphemus group (Figures 6 and 9) show distress: their situation is bad but not yet desperate. The Old Fisherman is worried, but not yet distressed. We see, therefore, the hierarchy of emotions in action.

Figure 4. A Persian warrior from the Lesser Attalid Group (detail), mid-second century BCE. Rome, Musei Vaticani.

Figure 5 (left). A Persian warrior from the Lesser Attalid Group (detail), mid-second century BCE. Musée Granet, Aix-en- Provence. Figure 6 (right). Odysseus’ Companion carrying a wineskin (detail) from the Blinding of Polyphemus group in Sperlonga, first century BCE/CE. Museo Archeologico Nazionale di Sperlonga.

Emotionality in Greek Art

283

Figure 7 (left). Ajax (detail) from the so-called Pasquino Group, second century BCE. Rome, Musei Vaticani, Museo Pio Clementino. Figure 8 (right). Kneeling Gaul (detail) from the Lesser Attalid Group, mid-second century BCE. Venice, Museo Archeologico Nazionale.

Figure 9 (left). Odysseus (detail) from the Blinding of Polyphemus group in Sperlonga, first century BCE/CE, Museo Archeologico Nazionale di Sperlonga. Figure 10 (right). Statue of an Old Fisherman (detail of cast) from Aphrodisias, late third century BCE. Oxford, Ashmolean Museum.

284

Olympia Bobou

Several emotions can be subordinated under sadness in the classification followed by Phillip Shaver, Judith Schwartz, Doald Kirson, and Cary O’Connor (Figures 11–14). However, the main emotion we see in Greek artworks is agony, or suffering. It can stem from physical causes, as in the Hanging Marsyas, the giants from the Pergamon Altar, or Laokoön and his sons; or mental pain, such as that of Ge pleading for her son’s life.

Figure 11 (left). Hanging Marsyas, second century BCE. Rome, Musei Capitolini. Figure 12 (right). Falling Gaul (detail) from the Lesser Attalid Group, mid-second century BCE. Venice, Museo Archeologico Nazionale.

Figure 13. Details of figures from the Pergamon Altar, mid-second century BCE. Berlin, Pergamon Museum. © Johannes Laurentius. Figure 13a (left). Alkyoneus. Figure 13b (right). Ge.

Emotionality in Greek Art

Figure 13c (left). Opponent of Dione.

Figure 13e (left). Opponent of Triton.

Figure 13g (left). Giant near Themis.

285

Figure 13d. (right). Tityos.

Figure 13f (right). Opponent of Doris.

Figure 13h (right). Opponent of Apollo.

286

Olympia Bobou

Figure 14. Details of figures of the Laokoön Group, second century BCE. Rome, Musei Vaticani Figure 14a. Laokoön.

Figure 14b (left). Laokoön’s younger son.

Figure 14c (right). Laokoön’s older son.

Emotionality in Greek Art

287

Anger is also fairly prominent, and its intensity varies (Figure 15). It is shown by centaurs and giants, as on the Pergamon altar, and different monuments with Centauromachy displays. It is interesting to note that satyrs, heroes, and other figures do not display anger.

Figure 15. Details of figures from the Pergamon Altar, mid-second century BCE. Berlin, Pergamon Museum. © Johannes Laurentius. Figure 15a (left). Porphyrion. Figure 15b (middle). Biting giant. Figure 15c (right): Opponent of Artemis.

Figure 15d (left). Opponent of Phoebe. Figure 15e (middle). Opponent of Apollo. 15f (right): Opponent of one of the Moirai.

Delight, euphoria, and pleasure are all subordinate emotions to joy, and they are obvious in groups such as the Invitation to Dance (Figure 18), or figures such as the Drunken Old Woman (Figure 16), or the Young Centaur with Eros on his back (Figure 17). Satyrs show various levels of intensity – delight, pleasure, joy (expressed with toothy grins and laughter) – whereas the figures of old women show the emotion at its most intense. We also see great variety in the combination of markers (or formal devices) for showing the emotions of the figures. For example, the greatest collection of displays of emotion in Greek art can be found on the Great Frieze of the Pergamon Altar.42 The frieze is 113 metres long and features more than a hundred figures of gods and giants engaged in bitter combat. It can provide us with a good sample of the possibilities for the variation in the use of emotional markers, in42

For the Pergamon altar, see Humann 1959; Rohde, 1981; Queyrel 2005. For the frieze see Hermann 1910; Kähler 1948; Simon 1971; Massa-Pairault 2007.

288

Olympia Bobou

cluding one that is unexpectedly used: almost all the figures on the frieze are shown with parted lips, gods and giants alike. This is an indication of a change: the half-open mouth can no longer be used exclusively as an emotional marker (as it was in the art of previous centuries), but is now a marker of expressivity, dynamism, and life-like behaviour. But the use of other markers is varied, and it is worth mentioning briefly the figures showing emotions along the frieze.

Figure 16. Drunken Old Woman (detail), late third century BCE. Rome, Musei Capitolini.

Emotionality in Greek Art

289

Figure 17 (left). Young Centaur (detail), late third–early second century BCE. Rome, Musei Capitolini. Figure 18 (right). Young nymph from the Invitation to Dance group. Oxford, Ashmolean Museum.

Starting from the East side of the frieze, which was the main side of the altar and portrayed Zeus and Athena to whom the altar was dedicated, the first group we encounter is that of an older giant (Klytios) fighting Hekate. Klytios is shown frowning, with wide eyes and half-open mouth that bares his teeth. However, the lines on the forehead are more indications of age rather than muscular movement, as the cheek- and eye-areas are relaxed. Artemis is fighting a youthful giant, while an older, snake-legged giant is falling between them, trying to fight off Artemis’ dog (Figure 15c). His mouth is half-open with his upper teeth showing. The nostrils are flared, the eyes are wide, and he is frowning slightly. Leto’s opponent, fallen and trying to repel the goddess, is young. His mouth is open, his upper teeth are showing, his nostrils are flared, he is frowning deeply, and his eyes are open wide. The deeply-bulging area over the eyebrows indicates strong muscular contraction. Apollo’s fallen opponent (Figure 13h) is another young giant, who is shown with slightly open mouth and frowning deeply. The rest of the figures from the Apollo and the Zeus groups are badly preserved (Figure 19). Porphyrion, Zeus’ opponent (Figure 15a), is frowning deeply, and the area over his eyebrows is bulging. In the group that features Athena fighting a young giant while his mother Ge watches we have one of the most memorable scenes of the frieze (Figure 20). The young giant is shown with his mouth half-open, his upper teeth showing, nostrils flared, eyes wide and frowning deeply, with the area over the eyebrows bulging. The lower part of Ge’s face does not survive well (Figure 13b), but her eyes, wide and upturned, and her deep frown show her distress. The similarity of her expression to that of the young giant defeated by Athena is striking,

290

Olympia Bobou

but her eyes are wider and not as deep-set as the giant’s, and her frown is less pronounced.

Figure 19. Zeus overpowering his opponents, Pergamon Altar, mid-second century BCE. Berlin, Pergamon Museum. © Johannes Laurentius.

Figure 20. Athena defeats Alkyoneus, Pergamon Altar, mid-second century BCE. Berlin, Pergamon Museum. © Johannes Laurentius.

On the South side of the frieze the first giant whose face is preserved well enough for us to see his expression is the steer-giant. His mouth is half-open, his nostrils flared, his eyes wide, and he is frowning. Phoebe’s opponent (Figure 15d) is shown frowning and with slightly narrowed eyes. Two giants from the West side are showing emotion: the first is the giant next to Triton, and the second is the one next to Doris (Figures 13e-f). Both are youthful and have almost the same expression: parted lips, eyes looking upwards, and

Emotionality in Greek Art

291

slight frowns. The first has his eyes open wide, while the second’s eyes are smaller and slightly narrowed. Finally on the North side we see emotions in the snake-legged giant biting his opponent’s arm (Figure 15b), the giant in front of a winged goddess, and the giants fighting two female goddesses. The first is an older giant, open-mouthed, with teeth showing, flaring nostrils, and wide eyes, who is frowning. The second is also bearded, with open mouth with the upper teeth showing, and wide eyes. The third is bearded, open-mouthed, with narrowed eyes, and frowning, while the last is a young giant, open-mouthed, with visible upper teeth, flared nostrils, and wide eyes. Another example is the Lesser Attalid Group. Pausanias describes the dedications of Attalos on the South Wall of the Acropolis:43 statues depicting a Gigantomachy, an Amazonomachy, a battle against the Persians, and a battle against the Gauls. Since he did not specify which Attalos dedicated them, much of the scholarly debate has focused on the question of the monument’s patron and date.44 What survives is a group of ten figures, Roman copies of the Greek originals, showing giants, Amazons, Gauls, and Persians about to be defeated or to fall in battle. All have slightly parted or half-open mouths, but in the figures that are shown as still living the nostrils are slightly flared and frowns are visible. In the dead or dying figures the forehead is relaxed and the eyes are closed or halfopen.45 The Dead Persian (Naples) has his mouth slightly open, with the upper teeth visible. His nostrils are still flared, the eyes are half-closed, and the forehead is relaxed. The Dead Amazon (Naples) has her mouth slightly open, with the rest of her face relaxed and her eyes closed. The Dead Giant (Naples) has his mouth halfopen, but the rest of his face is relaxed and his eyes are half-open. The last figure from the group in Naples has an ancient head, but one that did not originally belong with the figure but came from another statue. The Kneeling Persian (Aix-enProvence, Figure 5) has his mouth half-open, with upper and lower teeth visible, flared nostrils, eyes wide open, and a deep frown. The Kneeling Persian (Vatican, Figure 4) has the mouth slightly open, with upper teeth visible, flared nostrils, slightly narrowed eyes, and a deep frown. The Kneeling Gaul (Paris) has his mouth slightly open with his upper teeth showing, and his eyes wide. The Dead Gaul (Venice) has his mouth slightly open with his upper teeth showing, relaxed nostrils, eyes half-open, and he is frowning slightly. The Kneeling Older Gaul (Venice, Figure 8) has the mouth half-open, revealing both upper and lower teeth, flaring nostrils, eyes wide open, and a gentle frown. The Falling-back Gaul (Venice) has the mouth half-open, showing upper and lower teeth, relaxed nostrils, eyes quite wide open, and a slight frown.

43 44 45

Pausanias 1.25.2. For example, Stewart 2004, 218–220, prefers Attalos I; von Prittwitz 2007, 241–271, esp. 267–270, prefers Attalos II. Stewart 2004 is the definitive study of this group.

Olympia Bobou

292

What is interesting here is not just the variety of ‘formal devices’, but also their use in dying or dead figures, including the open-mouth motif, which is usually considered a sign of emotionality or dynamism. As Andrew Stewart observes, the variation of markers creates the range from fear of defeat to dying to death, which reflects not so much the emotional states of the depicted figures (especially since dead figures have no emotions), but is instead used for invoking emotions in the viewers.46 6 WHO SHOWS EMOTIONS? The previous question is directly connected to this one: who shows emotions in Greek art? Through the survey of artworks we see that even though there is little variation in the actual emotions shown (anger, rage, fear, worry, distress, joy, pleasure), there is great variation in the figures showing them: heroes and their adversaries, old men, and women. In fact, it seems that Olympian gods, and mortal, young and middle-aged men and women are not allowed to show any emotion at any point. Only Ge shows pain at the imminent death of her children, but she is not an Olympian and is shown as a mature woman. This brings us back to the question of expressivity versus representations of emotion. We have seen how the same markers can be used to show expressivity, age, and emotional states, but the survey of artworks shows that in portraits of mortal men and women, they should be interpreted as age and expressivity markers, with the potential of causing emotional responses from the viewers (as we will see later), rather than markers of emotional displays. Emotional displays belong to the world of myth, but the survey of artworks also shows the limitations of emotions. We see that very few artworks show emotions (Lesser Attalid Group, Pergamon Gigantomachy frieze, Blinding of Polyphemus, Scylla group), or have emotions as their subject (Old Fisherman, Laughing Old Woman, Invitation to Dance group, Pasquino, Laokoön, Hanging Marsyas). Even though they appear in high-quality and highly visible monuments, they are not a common feature of Greek art. Their strangeness and rarity would have been even more obvious in the Hellenistic world, before the time of the creation of numerous, obviously much-admired and desired Roman copies. Next we see that the emotions of older people are shown as more intense than those of younger people. This is especially obvious in the Lesser Attalid Group, where we have younger and older Persians and Gauls together, and in the groups showing Odysseus tricking Polyphemus and fighting Scylla. This can be attributed perhaps to the blurring of physiognomical and emotional markers that we have noticed already in the art of the fourth century: since the way of showing one’s age is through wrinkles, and wrinkles are also used for showing emotions, an older man in an emotional state looks like one with particularly pronounced wrinkles. 46

Stewart 2004, 144–170, 237–241.

Emotionality in Greek Art

293

Another apparent point is that fairly consistently older men are shown in anger, while younger men display pain. The only exceptions are Laokoön and Marsyas, both characterised as men in later middle age. These emotions are all shown in situations of heightened physical activity and anxiety or both, as in battles. Joy, on the other hand, shown both by young satyrs or centaurs or young and old women, is connected to physical activities that have preceded it or that will follow it. For example, the Drunken Old Woman is no longer drinking, but her activities have led her to her state of happiness, while the Satyr and the Nymph of the Invitation to Dance Group are anticipating pleasurable physical activity. These observations can explain why Achilles is not showing great emotions when he is lifting the body of Penthesileia, unlike Menelaos when rescuing the body of Patroclus. Achilles is both the perfect hero and a young hero. So if the survey of artworks really shows the artistic conventions of representing emotions, then Achilles is not allowed to show any emotions, since pain is an emotion unbecoming to a young hero. 7 ATTITUDES TOWARDS EMOTIONALITY The survey also shows a change of attitudes: from the display of emotions as simple ‘facts’, visual statements of a figure’s state at a specific situation, to displays of emotionality. In the Classical period both Greeks and their adversaries show emotions. However, throughout most of the Hellenistic period, when emotionality becomes prominent in artworks, as we shall see, emotions can be associated with barbarians, giants, centaurs, or satyrs, as well as fishermen and women, all figures that represent the Other in Greek (male, normative) mentality.47 In the first centuries BCE/CE, we see Greek heroes showing emotions again, although in a limited way and under certain conditions. Even though the intensity and nuance of the emotion varies from case to case, they can be categorised under fear, sadness, anger, or joy.48 This shows not just the limited range of emotions represented, but also how negative emotions are preferred to positive ones when it comes to displaying them. This raises the questions: why are emotions used in Greek art? Where are emotions used? In order to answer them, we must first look at the types of narrative scenes where emotions are used. Almost all of the scenes derive from the world of myths and legends: the battles between gods and giants, centaurs and men, Greeks and Trojans, or satyrs and nymphs. Only the Drunken Old Woman and the Fisherman do not seem to belong to that world of myths and legends, although suggestions have been made

47

48

For the figure of the Other and Otherness in ancient Greek thought and art, see Cohen 2000; Sassi 2001. See also Hölscher (ed.) 2000a, esp. Dihle 2000, Schnapp 2000, Osanna 2000, Giuliani 2000, Hölscher 2000b, Heinemann 2000, and Schneider 2000. Following Shaver et al. 2001, 43–47 for the classification of basic emotions.

294

Olympia Bobou

that the figures belong in the Dionysiac world or the world of poetry, and so, they too are set apart from the world of mortals. However, in the numerous mythological monumental narratives known from the Greek world, only in these particular ones do we see the emotions of the figures portrayed. For example, we do not see them in the Amazonomachy friezes from the Mausoleum,49 or the Telephos frieze from the Pergamon altar,50 or even the Gigantomachy at the Lagina frieze.51 We do not see emotions in sculptural groups such as that of Achilles and Penthesileia,52 or Artemis saving Iphigeneia either.53 One answer could be that displays of emotions are nothing more than an option or a tool in the arsenal of sculptors trying to create ever more impressive images, and do not have any other significance. But perhaps that is not the whole answer. Visibility of the figures as a criterion for the use of emotionality is also not providing us with a fully satisfactory answer. It can explain why emotions are not used in the Lagina figures, since the height of the frieze would have made it difficult seeing the emotions on the faces of the giants. But it does not explain the lack of emotions on Achilles’ face from the Achilles and Penthesileia group. A chronological arrangement offers some clues as to the use of emotions in monumental artworks. The first depiction of emotions in monumental art occurs in the fifth century BCE, in the figures from the temple of Zeus at Olympia, then at the Parthenon metopes, and finally at Bassai. The sculptures of Athena Alea at Tegea are the most representative artwork of the fourth century, while the Drunken Old Woman and the Old Fisherman are dated at some point in the third century BCE. Monuments connected with the Attalid dynasty, as well as the original of the Laokoön and the Hanging Marsyas statues are dated in the second century BCE, while the Invitation to Dance group is dated between the third and second centuries BCE. Finally, the Polyphemus and Scylla groups are dated at some point in the first century BCE/first century CE. Perhaps the most interesting characteristic of the art of the fifth and the first half of the fourth century BCE is that emotions are shown on the faces of Greeks, including heroes.54 Heroes only show emotions again in late Hellenistic art, for example Ajax in the Pasquino55 or Odysseus and his companion in the Scylla group.56 However, there is a difference. In the Classical period, it was usually young, beardless men in pain (Lapiths on the Parthenon metopes or the Bassai frieze), or heroes showing effort and determination (Herakles on the Olympia me49 50 51 52 53 54 55 56

For the Mausoleum reliefs, see Cook 2005; Jenkins 2006, 203–235; Lucchese 2009. For the Telephos frieze see Dreyfus et al. 1996-97. For the Lagina frieze see Schober 1933; Junghölter 1989; Baumeister 2007. Vorster 2007, 315–319, figs. 316–320, with earlier bibliography. Vorster 2007, 316–319, figs. 322a–e. This applies for Classical art in general, as emotions are also shown on the faces of heroes and Greeks in pottery of the same period. For the Pasquino group see Vorster 2007, 312–314, figs. 311a–b. For the Scylla group see Vorster 2007, 325–327, figs. 335a–g.

Emotionality in Greek Art

295

topes). In the Hellenistic period, it is heroes showing distress and focus (Ajax, Odysseus), or their companions showing pain (Scylla group), or anxiety (Polyphemus group). All these men are also bearded, indicating their older age. Women also show no emotion in Classical or fourth century art, either when they are attacked (for example, the female Lapiths at every Classical Centauromachy) or when they are fighting, as is the case with Amazons. They are shown laughing in Hellenistic art, both as old, mortal women and young nymphs, or dying, as Amazons. Generally they are not showing suffering, either mentally or physically, with the exception of Ge at the Pergamon altar.57 These points can explain the lack of emotions in works such as the Mausoleum Amazonomachy: the frieze is located relatively high up on the Mausoleum, and created at the time when heroes no longer show emotions and women have not started to show them. To summarise: emotions start being shown through facial expressions and not gestures in the beginning of the fifth century, but they are shown realistically only from the third century BCE onwards. Greek heroes, mostly beardless, show emotion in the fifth and first half of the fourth century BCE, as well as in groups (like the so-called Pasquino) surviving from the first century BCE, first showing pain and later showing fear. Women only show emotion in Hellenistic art, if they are old or belonging to the mythological sphere. Centaurs, satyrs, giants, Trojans, and then Persians and Gauls, also show emotions (pain, anger, fear, and joy) from the fifth to the first century BCE. Emotions are more likely to be shown on monuments where the figures will be highly visible. Negative emotions are more common, and they are associated with stressful situations, such as battles, while positive ones are rare and not shown in situations that demand strenuous physical activity. 8 POSITIVE OR NEGATIVE EMOTIONS? Andrew Stewart’s analysis of the Lesser Attalid Group makes clear the role and importance of the Roman viewer in order to understand the significance of the group in the Imperial period.58 By looking at specific works of art, it is possible to detect the relationship between viewer and artist through the artwork in the Hellenistic period, and the shift towards emotionality. The first case is that of the Drunken Old Woman (Figure 16).59 She is shown laughing: she throws her head back, her mouth is open, revealing her teeth, the corners of the mouth are upturned, and her eyes are lined with strong laughter lines. She has been identified as a prostitute,60 or a cult agent of Dionysos,61 or a 57 58 59 60 61

For example Dirke, from the Farnese Bull group, is not shown with any facial emotions, even though her gesture is one of supplication. For the group see Pozzi 1991. Stewart 2004, 136–170, 237–241. More recently on the statue, see Masséglia 2012b, 413-430. Zanker 1989. Wrede 1991, 163–188, esp. 173–179.

296

Olympia Bobou

statue representing the female equivalent of a satyr,62 to name but a few of the interpretations given. The identification influences the way the figure is understood: for Paul Zanker it is odious and negative,63 for Henning Wrede and Christian Kunze it is positive and connected to the joy that Dionysos gives to his followers.64 These differing views alert us to a change with which emotions are used by Hellenistic sculptors: sharing and conveying the subject’s inner world or character is no longer enough; they also demand the viewers’ reaction and interaction. These statues are designed to engage the viewers. Another statue that is just as confusing to us and has invited numerous interpretations is that of the Old Fisherman (Figure 10), the original of which also dates from the late third century BCE.65 The expression is one of worry and weariness: slightly frowning, with open nostrils and half-open mouth (showing both upper and lower teeth in the Vatican copy), he looks as if he is struggling to walk. Most scholars agree on the ‘banausic’ quality of the Fisherman.66 The figure’s popularity in the Imperial period67 does not help us understand the function of the statue in the Hellenistic period, although the suggestion that it was a votive is the most plausible. Literary texts show us possible ways of understanding the figure: a pathetic representation of society’s poorest,68 a scathing look at less clever people,69 an awe- and pity-inspiring votive.70 All these responses, modern and ancient, perhaps show that the Fisherman was created more for evoking emotions and reactions, rather than for portraying a specific emotional state. The final example is that of the portrait of Alexander by Lysippos. We have already mentioned how the emotional and physiognomical markers used for creating Alexander’s distinctive image were taken up and used in portraits of his youthful successors who wanted to emulate him (p. 279). This appropriation of features and attributes, as well as the numerous copies of the type, show that the portrait was successful and had a strong resonance with viewers and patrons. In this case, Alexander’s image cannot be separated from what Alexander as a figure meant to the people: looking at the image leads to thinking of the figure, which leads to repetitions of the image, that again lead to thinking of the figure.

62 63 64 65 66 67 68 69 70

Kunze 1999, 73–82, esp. 69–80. Zanker 1989, 42. Wrede 1991, 175f.; Kunze 1999, 80. Kunze 1999, 53f. Himmelmann 1980, 84; Laubscher 1982, 49–59; Bayer 1983, 42; Kunze 1999, 54. Bayer 1983, 248–255, for copies of the type. See also Kunze 1999, 54 note 44. Himmelmann 1980, 87f. Laubscher 1982, 49–59. Kunze 1999, 68f.

Emotionality in Greek Art

297

9 CAUSES OF AND REACTIONS TO EMOTIONS Reacting to the emotions of others is a normal human response, and, consequently, artworks showing emotion are bound to cause reactions. However, some artworks are designed especially for evoking emotions in the viewers. These are sculptures such as the Fisherman and the Drunken Old Woman, but also ruler portraits. So, if emotions are meant to be seen and to evoke reactions, what were supposed to be the expected reactions? What is the point of visualising anger or suffering? In order to try answering this question, we must look briefly at the emotions we find described, explained and categorised in ancient Greek thought. Plato recognises several emotions, which are initially considered as parts of the immortal soul, but in his later works these are relocated to the mortal soul.71 In Timaeus he talks of pleasure, distress, confidence, and fear, while in Laws he adds shame, friendly love, and hate, and in Philebus anger, longing, love, mourning, jealousy, and envy.72 Emotions can come about when either pleasure or distress or both are present;73 for example, laughing at a friend’s ignorance is caused by envy, which is distressing, and amusement, which is pleasurable.74 What is interesting for us here is how mixed emotions, such as anger, longing, or jealousy, can be used for understanding the situation of others in comedy and tragedy, both on and off stage.75 Furthermore, anticipation of a negative or positive emotion is accompanied by a mental image of oneself enjoying or feeling distress.76 Aristotle discusses emotions mainly in two works: the Nicomachean Ethics and the Rhetoric.77 He recognises fear, confidence, desire, anger, pity, pleasure, and distress,78 but it is in the second book of the Rhetoric that he describes the emotions in detail, and shows how they can best be used by a speaker.79 He pairs them in opposites: anger–mildness, friendly love–hate, fear–confidence, shame– graciousness, pity–indignation, and envy–emulation, based on whether the primary sensation that can be associated with them is painful or pleasurable.80 For him, both imagination and sight can arouse emotions in the viewer.81 In both Plato and Aristotle we observe that fear and pity are aroused when we see something we recognise and we fear happening to ourselves. Aristotle is more 71 72 73 74 75 76

78 78 79 80 81

Knuuttila 2004, 13–16. Plato, Timaeus 69d; Laws1.647a–d, 649b–c, 2.653a–c, and 3.699c–d; Philebus 47e. Knuuttila 2004, 16f. Plato, Philebus 49d-50a. On these mixtures of emotions, see Knuuttila 2004, 23. Plato, Philebus 50b–d; Knuuttila 2004, 23f. Plato, Philebus 40a. Knuuttila 2004, 23. For a detailed analysis of emotions in Aristotle see Konstan 2006. Aristotle, Nicomachean Ethics 2.6, 1106b18–23; Knuuttila 2004, 25. Fortenbauch 2008, 29–47, esp. 33; Konstan 2006, 33–38. Knuuttila 2004, 28–33, Konstan 2006, 38–40. For example, fear is ‘a certain pain or disturbance at the appearance of an approaching destructive or painful evil’ (Aristotle, Rhetoric 2.5, 1382a21–2); Knuuttila 2004, 36–38; Konstant 2006, 29–32.

298

Olympia Bobou

detailed: we feel fear for someone who is like us,82 and pity is felt for those who suffer undeservedly (and who are not relatives), and to whom something that we fear happening to ourselves is happening.83 The Stoics divided emotions into primary ones: pleasure, pain, desire, and fear, under which all other emotions can be categorised. The first two relate to the present, while the others to the future, and they can be further connected to good and evil. So pleasure and desire are related to good experiences, one belonging to the present and the other to the future, while pain and fear are relevant to unpleasant/evil experiences, present and future respectively. Reaction to emotions, good or bad, is also part of what constitutes emotions in Stoic thought.84 Stoic theory also advises freedom from emotions (apatheia), since emotions are effectively disturbances of the soul.85 The Epicureans see two main emotions: pleasure and pain. Instinctively, people seek pleasure and avoid pain. The type of pleasure experienced by the person differs depending on whether pain is absent or removed: when no pain is present, then the person feels a ‘static pleasure’ (ataraxia), whereas when the pain is removed, then the pleasure is ‘kinetic’. Happiness is the state in which a person has satisfied all the natural urges that could cause suffering to the body (such as hunger or thirst), and the mind (such as misguided quest for glory or wealth), and has achieved ataraxia. All emotions, such as envy, hatred, contempt, pity, and distress, are signs of weakness and dependency on others, and ultimately can prevent the person following the Epicurean way of life from attaining pleasure.86 So, depending on the philosophical school one followed, negative emotions were to be controlled or eradicated, while positive ones were to be accepted, if not encouraged. Seeing the emotions of others, regardless of whether the emotions were seen in real people or in literary works, and one has to also suppose, works of art, could trigger an emotional response in the viewer, although the exact causes of the reaction varied from school to school. Some emotions were directed towards specific individuals, such as anger or envy, while others, such as fear or pity, were triggered because the viewer could empathise with others. Implicit or explicit in the texts is the commonality of experience and state between viewer and object. For example, a father feels pity for another man’s orphan because he too can imagine his own children becoming orphans or because he had experienced that state himself. Even though philosophy offers a taxonomy and explanation for emotions, literature offers their depiction and exploration as motives for actions (or lack of them). Tragedy, in particular, is ‘the imitation of an action that ... by means of fear and pity accomplishes the catharsis of such emotions’,87 and tragic figures need 82 83 84 85 86 87

Aristotle, Poetics 13, 1453a3–6; Konstant 2006, 129–155. Knuuttila 2004, 39; Konstan 2006, 201–218. Knuuttila 2004, 51–53. Knuuttila 2004, 63–80. Knuuttila 2004, 80–85. On Stoic thought on emotions see also Brennan 1998, 21–70; Irwin 1998, 219–241; Graver 2007. Aristotle, Poetics 6, 1449b27–28.

Emotionality in Greek Art

299

not be Greek in order to arouse the viewers’ emotions. Hellenistic poetry also offers examples of interest in figures outside of cultured society: goatherds, shepherds,88 and fishermen89 who live in nature; the Cyclops is also transformed from a wild mountain of a man into a love-struck shepherd.90 Women and their world also become a worthy topic of exploration.91 Finally, although later than the sculptures, the literary descriptions of paintings or sculptures (ekphraseis) offer evidence for a changed attitude towards some of the mythological monsters: lovely and maternal female centaurs take care of their children,92 while the satyrs are always ready to leap, play, and laugh, even when tragedy strikes among them.93 10 EMOTIONAL RESPONSES TO ARTWORKS We have recognised a variety of emotions in Greek art, and it is worth examining them through the lens of the different philosophical schools in a chronological order, starting from the fifth century BCE. First of all, we identified Herakles showing emotions based on a number of artistic devices; however, the comparison with the lists of emotions drawn up by philosophers shows that an ancient viewer would not necessarily recognise Herakles as being in an emotional state. A Stoic would perhaps see in Herakles an expression of the propatheiai, the preliminary, physiological responses that lead to emotional responses, while a Platonist or an Aristotelian could perhaps see Herakles’ actions in the framework of the emotional responses that led him to the point of having to perform his labours (for example, anger that led to killing his family), but would not see emotions in Herakles’ figure. On the other hand, Athena’s blank expression, which was not considered indicative of emotion, would have been recognisable to an ancient viewer as a representation of calmness (praotes) or even friendship or friendly love, as much as a god can feel philia towards a non-god. We also recognised emotions in some of the figures on the pediments. First of all, we saw the anger of centaurs. An ancient viewer would also have recognised it as such, but, according to Aristotle, anger is a response to a slight.94 Plato, also considers anger among the ‘foolish councillors’ of the mortal soul,95 so even though he does not dwell on its origin, he highlights its negative associations. This 88 89 90 91 92 93 94 95

For example, Theokritos, Idylls 1 (‘Thyrsis’), 4 (‘The Herdsmen’), 5 (‘The Goatherd and the Shepherd’). Theokritos, Idyll 21. Theokritos, Idyll 11. Herondas, Mimes 1, 4, 5, 6, 7, and 9. Philostratos, Imagines 2.3 (‘female centaurs’). Philostratos, Imagines 2 (‘Marsyas’). Aristotle, Rhetoric 2, 1378b. For the different types of slight, see Viano 2003, 85–97, esp. 88–90. Plato, Timaios 69d.

300

Olympia Bobou

is not the case with the anger of the centaurs; they instigated the fight between themselves and the Lapiths, and even if one accepts that the anger they show is in response to their feeling pain or being slighted when they are attacked back, they are still the initial wrong-doers. So, the centaurs’ anger is an emotion that should not exist; it is a manifestation and consequence of an unruly, wild nature. As such, its expression acts as a counterpoint to the calmness and dignity displayed by most of the Lapith men and women, and sends a clear message about the wrongness of unjustified anger, and lack of self-control. This message is repeated over and over at every Centauromachy in Greek art. The pain of the young Lapiths (West Pediment Figure Q in Olympia, South Metopes of the Parthenon, Bassai Frieze) is interesting. Both Plato and Aristotle would see pain as a reaction to physical stimuli, and the loss of the soul’s natural balance. As such, it is not an emotion, but a cause of emotions. Understanding pain as a stimulus instead of an emotion would mean that the Lapiths’ physical response to the centaurs’ actions could be seen as arousing fear or distress in them, following Aristotle’s definition of fear as ‘due to a mental picture of some destructive or painful evil in the (imminent) future’, or the Stoic definition of distress as fear for something bad that is happening in the future. The fear or distress of the Lapiths has several consequences: first of all, fear is the opposite of confidence. According to Aristotle confidence is an emotion that is associated with youth and recklessness, with going into battle knowing that you are well-armed and with many supporters. To see the Lapiths showing fear is to see how dangerous their situation is, since it is not an emotion one would associate with them. Secondly, as mentioned earlier, seeing something undeserved happen to someone good rouses pity in the viewer, and pity is also directed towards the young, the women, and the old, and not towards men in their prime. This means that the distress of the Lapith young men is meant to evoke pity in the viewer; it can be compared to the physical pain experienced by Philoktetes in Sophocles’ homonymous play, or by Herakles in the Trachiniae, which is also used as a means of evoking pity for the heroes’ suffering.96 But, at the same time, unlike the pity one feels for heroic figures in tragedy, an emotion that can be processed through words and resolved at the end of the play, there is no evident resolution or ending in artworks. The viewer has to supply that in his or her mind while confronted by a perpetual, frozen image showing a negative emotion. Or does perhaps the youthfulness of the figure negate or makes less apparent its heroic status through emphasis on vulnerability? What if the figure feeling pain

96

See Budelmann 2007, 443–467, esp. 443–446, for the emotions that the representation of physical pain can evoke to tragedy viewers, among which pity is the prevalent one.

Emotionality in Greek Art

301

is a bearded, famous hero such as Patroclus in a cup by the Sosias Painter?97 This is a conundrum that Classical Greek art refuses to answer. Instead, we see the pain (and fear) of heroes disappearing from art, to be replaced by confidence, such as that of the figures in the Athena Alea temple. Confidence is also apparent in ruler portraits of the fourth century BCE and the Hellenistic period, such as those of Alexander, Demetrios Poliorketes, and Attalos I. It is a positive emotion, but its primary importance lies in the message it is supposed to convey to the viewers: the image is meant to impress, inspire, and instil confidence; it detracts from the ruler’s youthfulness and possible lack of experience; after Alexander the Great, it also carries connotations of assimilation with this new, modern-day hero, and so becomes appropriate even for older rulers who want to be shown as youthful.98 Portraits of older rulers, such as Ptolemy or Seleukos, convey calmness or mildness, similar to that of the gods. They are knowledgeable and experienced, and promise to be just to those who respect, follow and obey them. Laughter is generally considered an indication of a positive emotion such as joy or calmness. Together with feasting and amusing oneself Aristotle considers it one of the markers of people who are in, or entering a state of calmness. It is also a characteristic of young people, not old. So, the laughter of the Drunken Old Woman could be explained as a positive emotion, a sign of her enjoyment of the feast she’s been, and that of the Satyr and Nymph from the Invitation to Dance group provides a sign of their amusement and entertainment. However, there is a problem with such a reading. The main subject of philosophical inquiries – and the ideal viewer and reader in ancient Greece — is a freeborn man of means in his prime. For him, both satyrs and women are considered Other, different, wild, inferior, unbalanced, voracious, controlled by appetites rather than reason. Thus, their laughter is associated not with the positive calmness of the ideal viewer but with wildness and lack of restraint. At the same time, satyrs and wine belong to the world of Dionysos, the god who transcends boundaries, frees one from self-control and allows mortals to approach the divine through ecstasy. Under these conditions, even the laughter of Others is accepted, understood, and transformed (again) into something positive. A similar transformation of the laughter of satyrs into a natural emotion can be seen in literature. In Euripides’ Cyclops, the laughter and delight of Satyrs is clearly connected to the gifts of Dionysos, wine, music, and dance. Even though it is in the nature of satyrs to pursue them, they are not shown as intrinsic parts of the satyrs’ nature. By the second century CE, in Philostratos’ 20th Painting on Satyrs approaching the sleeping, beautiful youth Olympos, the satyrs are described as ‘ruddy’ and ‘grinning’, while their actions are those of daring stalkers: they 97

98

Berlin, Antikensammlung F 2278. On that, see Junker 2012, 1–18. Beazley Archive Pottery Database no. 200108: www.beazley.ox.ac.uk/record/2E23511B-953D-4807-86A8-7126BB40 FAFF (accessed 4 July 2012). For portraits of Attalid rulers shown as young men, see von der Hoff 2011, 122–130, esp. 130.

302

Olympia Bobou

watch him, contemplate touching him, leave him gifts, and the most clever of them touches Olympos’ flute. In the 22nd Painting on Midas, satyrs are described in greater detail, as creatures that enjoy dancing, laughing, and seducing women, and their pictorial representation expresses that by showing them sturdy and impetuous, with prominent ears and thin waists. This behaviour is shown even more emphatically as a part of satyrs’ nature in the Second Painting by Philostratus the Younger: the satyrs who watch the approaching death of Marsyas mourn for him, but at the same time they show their noble, high-minded spirit and their natural tendency to leap and play. With laughter we see yet another example of how a specific emotion can be perceived and have multiple meanings, all belonging to the dominant cultural and social framework of understanding emotional displays. The Old Fisherman’s expression is also an emotional one. The usual Platonic or Aristotelian divisions do not seem applicable in his case, unlike Stoic distress. It is also an expression that can arouse pity; even if the Old Fisherman is nothing like the ideal viewer, old age is an unavoidable phase of the human life that can be instantly recognised by all. Even though it has been suggested that his features indicate stupidity, the popularity of the figure, with the appearance of poor fishermen in Hellenistic poetry,99 betrays interest in others – non-ideal subjects – and perhaps recognition of the pitiful state of old men who have no option but to keep on working or die.100 In the second century BCE Hellenistic monuments such as the Lesser Attalid Group and the Pergamon Altar we see fear and anger. The anger of the giants can be compared to that of the centaurs, and is an uncomplicated emotion both in its identification and reception. Dealing with fear is more difficult, since it poses another question: can an ideal viewer be moved to pity, or experience fear himself, when confronted with the fear of a non-Greek, non-human, non-male? On first thought, the response would be an unequivocal ‘no’. Most of the textual evidence suggests that pity was not an emotion that could be felt when it came to the suffering of those who deserved it. Also, when confronted with real Gauls or Persians, an ideal Greek viewer would probably find them unlike himself, and his pity would probably not be aroused. His admiration at their brave behaviour during battle – especially that of Gauls – could perhaps have been mixed with incredulity and contempt, like Aristotle, who condemned Celtic fearlessness as madness, because they feared nothing.101 But then there is the tragic response, and the response of Aristotle to tragedy: incidents that cause fear and pity lead to catharsis, and they do not always occur when Greeks are portrayed on stage. By viewing the misfortunes or tragic deeds of others, one can become emotional oneself, and find some common ground between oneself and the tragic characters’ experiences. One could argue that the gi99 Theokritos, Idyll 21. 100 Theokritos, Idyll 21.63-67. For depictions of non-ideal, non-citizens or disfigured, poor people in Hellenistic art, see Stewart 1997, 224–228. 101 Aristotle, Nicomachean Ethics 3.7, 1115b24–8; Knuuttila 2004, 45.

Emotionality in Greek Art

303

ants are not tragic figures, but the inclusion of Ge and her depiction as a mother begging for her son’s life emphasises the youthfulness (and inexperience?) of the young giants.102 Also, Ge as the instigator of the battle can be seen as a tragic figure: her wish for avenging her fallen children, the Titans, leads her to lose her other children, the giants, as well. In this way, the Ge-Alkyoneus-Athena group is transformed into a theatrical group, inviting the viewer to see (and to interpret) it as he would a scene from tragedy. This change blurs the boundaries between the proper reaction at the pain of a person suffering unworthily, and the pity one would be inclined to feel regardless of the person’s worthiness because of their tragic circumstances.103 Then there is the Stoic response, which would find all the depictions of distress and fear in the monuments as examples of why emotions cannot be allowed to hold sway over a man’s life. The Pergamon Altar, with its portrayal of the ultimate hybris story, the attack of giants against the Olympians, could be a prime example of negative responses and judgements that lead to ruin. One thing is certain: in a politically charged dedication from the Attalid king in return for favours granted by the gods, the display of emotions cannot be accidental or just for the purpose of creating dramatic tension. One of the Attalid aims in the decoration of the altar is the glorification of their dynasty, by proclaiming their connection with Herakles, the mortal without whom the gods would not have won against the giants.104 The gods and their allies, including Herakles, all show the slightly parted lips and the focused gaze, which have been identified as a sign of emotionality in other figures, and in this case can be interpreted as confidence – and righteousness, one would add. The victory of the Olympians is a given – they have their mortal ally and their cause is just. In this context, the fear and anger of the defiant and defeated giants heightens the impression of confidence and calmness one sees displayed in the Olympian gods and Herakles. It contrasts their wild, temperamental nature with the selfcontrol of divinity. Seeing the emotional displays on the altar through the Stoic lens adds further meaning to them.105 To see the giants succumbing to their emotions is to see how emotions can imbalance the soul and disfigure the body. In this manner the display of such passions can teach the importance of restraining emotions by negative example: succumb to your emotions, and you become like a giant. In artworks of the first centuries BCE/CE we see the return of the fear of the Greek hero. Odysseus’ companions in the Polyphemus and Scylla groups show their anxiety over what is about to happen (imminent death in the case of the companions about to be killed by Scylla), as well as their insecurity and lack of 102 103 104 105

For the inexperience of young warriors being pitiful, see Konstan 2001, 79. That this kind of pity was not unknown among Greeks, see Konstan 2001, 81, 87–90. For Attalid glorification on the Pergamon Altar, see Queyrel 2005, 126–136. Other scholars have identified Stoic influences on the Altar. According to Massa-Pairault 2007, the frieze’s design is reflecting Stoic ideas on cosmology. See also Queyrel 2005, 136– 138.

304

Olympia Bobou

knowledge about the future (the companions helping Odysseus blind Polyphemus have no idea if their stratagem will be successful). Unlike the Classical Centauromachy scenes in which the Lapiths were engaged in one-to-one combat with the centaurs in the metopes, or in paratactically arranged groups, as in the Bassai frieze, the Scylla and Polyphemus groups are designed as multi-figural objects, with multiple points of view. The mythological narrative is emphasised and the emotions of the figures are prominent, as if they invite the viewer’s responses. The same can be said about the Laokoön group. Different versions of the myth survive and, depending on them, Laokoön could be the unwitting victim of the gods,106 or the priest whose impure acts justify his punishment.107 His children could also be understood as innocent victims, or their death could be seen as added punishment for the sins of the father. Their fear could or could not provoke pity, although the existence of a (now lost) tragedy by Sophocles on the topic108 indicates that pity was sought. 11 WHEN ARE EMOTIONS USED? Combining Greek ideas about emotions with artworks that show emotion, even on such a brief and limited way, shows some interesting things. First of all, that from the fourth century BCE onwards, art-works were designed and created for provoking reactions and emotions, and not just for displaying them. All the Hellenistic statues we have seen invite the viewer to respond and then think about his responses, and the emotions are put centre-stage precisely for this reason. Secondly, they invite mixed responses at the same time. The prospective viewer is informed by his prejudices and his education, and, depending on his social position, culture, age and so on, he has different ways of approaching the same monument, that need not be exclusive. In some cases, we see that polysemy is actively sought, and so it is possible that creating ambivalence and contradiction in the viewer was part of what made a monument successful, interesting, discussed, influencing, and copied. Thirdly, while some of the monuments offer a narrative as complete as possible, for example the Pergamon altar with the battle with defeated giants and victorious gods, others, such as the Drunken Old Woman, leave the viewer alone with the figure; others put the viewer in the place of accidentally coming across a scene, for example the punishment of Marsyas or the invitation of a satyr to a nymph. The ideal viewer’s position is not fixed. On one occasion he is a spectator, 106 This is the most popular version, and the one followed by Virgil in the Aeneid 199–227. 107 According to one version, Laokoön slept with his wife in the sanctuary of Apollo (Apollodoros, Epitome 5.18). According to another, this impure act took place at the sanctuary of Poseidon (Servius, Commentary on the Aeneid 2.201). A different version sees Laokoön as the priest of Apollo who married and had children, even though he should have remained chaste (Hyginus, Fables 135). 108 Pearson, Claverhouse Jebb, and Iedlam1917, ΙΙ 38–47.

Emotionality in Greek Art

305

in another he is put in a position where interaction with the image is possible, and in the third he is an intruder and a voyeur. All three attitudes find their literary expressions in texts, admittedly of different periods, which can be used as examples of the different ways of positioning a viewer in relation to an event or object. These are: first, drama choruses, where the actors comment on what they see or hear; second, poems such as Herondas’ fourth mime, where the women visiting Asklepios’ sanctuary comment on the life-likeness of the statues; and third, texts such as Lucian’s Amores, in which we hear the story of the young man who managed to lock himself in the temple of Aphrodite when he finally became tired of admiring her statue from afar.109 The reciprocal relation between sculptures and texts can also be seen in the connection between the Laokoön and the other Sperlonga groups and the texts that describe the fall of Troy and the foundation myth of Rome. Scholars disagree on which texts are followed: Virgil’s and Ovid’s versions have been considered the most likely,110 although a fourth century CE date and aesthetic has recently been suggested.111 They also disagree on whether the statues reflect the texts, or the texts, most notably the Aeneid,112 have been influenced by the sculptural depictions of the myth.113 In the end, it is a question that cannot be answered with certainty, although Virgil’s engagement with other works of art, combined with his sensitivity towards them, perhaps hints at his descriptions being influenced by the statue rather than the other other way around. 114 However, what is undeniable is that the emotionally-laden and emotion-inducing spectacle is connected to an emotionally-charged description. The examination of the meaning and reception of emotional responses adds two further criteria for when display of emotion can be used in art. First, it can be used for maximum impact upon the viewers. Emotions affect the viewers regardless of whether the person suffering them is a Greek or non-Greek, and to have them so prominently displayed induces strong emotional reactions. Second, emotions are also mostly used in mythological scenes that are instantly recognisable, such as Centauromachies or Gigantomachies, or that have instantly recognisable protagonists, such as satyrs. Therefore, works of art that show emotions are instantly connected to this sphere of high mythological drama – or comedy. The Lesser Attalid Group is a strong example of how the battles between Persians and Greeks or Gauls and Greeks are compared and connected to the mythical Gigantomachies and Amazonomachies, and the use of emotion is another tool for connecting mythical and real. 109 Lucian, Amores 15f. 110 For Virgilian connections to the Sperlonga groups, see Hampe 1972; for Ovidian see Andreae 1994. 111 See Squire 2009, 230–232. 112 See Weis 2000, 111–165, esp. 126–128. 113 Howard 1989, 417–422, esp. 422. 114 For a survey of artworks in Virgil’s work, see Fairclough 1906, 59–68. For Virgil’s sensitivity to art works see Highbarger 1942, 87–89; Bartch 1998, 322–342. See also Jones 2011 on how Roman painting (and literature) influenced Virgil’s Eclogues.

306

Olympia Bobou

The apparent lack of emotions in some cases is also interesting. First of all, the ideal blank expression of the gods can also be understood as showing divine emotion, though whether that is Aristotelian calmness or Stoic apatheia is up for discussion (and probably would have been so in antiquity). In the case of the gods at the Pergamon altar or images of heroes created by artists such as Skopas or Lysippos, we even see limited use of emotional markers, while the narrative context of the images (for example, battle scenes, weariness after the Twelve Labours) offers the explanation: confidence. Divine confidence also inspires confidence in the viewers themselves. As we have seen, most mythological women also show no emotion in sculpture, but men and women also show no emotions in portrait statues, honorific statues, or dedications, votive reliefs or funerary monuments. In the case of statues, instead of a narrative of events, we have a narrative of virtues, constructed through the use of posture, gestures, and clothes. In reliefs, whether funerary or votive, we have either a narrative of virtues, or a narrative of events (sacrifice, death in battle, domestic scenes), that are ‘told’ with the addition of other figures, props and attributes. In some cases, we can see emotions such as confidence or bravery expressed through a man’s expression, but with women it is always virtuous behaviour that is shown (modesty, propriety, fecundity). The almost complete lack of emotionality in depictions of women is especially interesting when contrasted with the literary portrayals of women, mythological and mortal. There, women are constantly in thrall of their emotions, whether falling in love or plotting revenge, as Medea does in numerous texts, and their behaviour is ruled by emotions. If texts either try to reinforce the stereotype of unruly female conduct, or to teach restraint and self-control through the use of negative examples, statues try to instruct by displaying positive values and virtues. The repetition of motifs and types reinforces positive behaviour, in a manner reminiscent of animal training, where treats and praises entice the animal into behaving in the right way.115 12 CONCLUSIONS The aim of this paper was to answer the following questions: (a) how are emotions expressed, (b) what emotions are shown, (c) who is showing emotions, (d) can the emotions shown be categorised as positive or negative, (e) when are emotions used in Greek art? The survey of artworks shows that the answers are not clear-cut or self-evident. Emotions are shown through the use of specific markers, such as open mouths and furrowed brows. Until the fourth century BCE the surface of the face is 115 See Xenophon, On Horsemanship II, for the training of horses. But see also Trimble 2011, who studies the replication and dissemination of the Large Herculaneum Woman in the Imperial period, and proposes that the type was used for creating and emphasising the image of an elite woman receiving public honours, 150–205, 259.

Emotionality in Greek Art

307

treated like a mask upon which the emotional markers can be inscribed, but from the third century BCE we see sculptors displaying their understanding of human anatomy, and the facial expressions correspond to exact movements of the facial muscles. Almost everyone can show emotions in Greek art, but mythological figures such as centaurs, giants or satyrs are more prevalent in the visual record, and the artworks where emotions are shown span from the fifth to the first centuries BCE/CE. Young Greek heroes show emotions in the fifth century, and older heroes in the first century BCE. Women rarely show emotions. Gods and rulers show limited emotions in the Hellenistic period. Most of the figures show negative emotions, such as fear or anger or distress. However, the display of joy cannot be easily categorised as negative or positive. Gods and rulers show only positive emotions. The display of emotions is ruled by a complex set of rules that range from practical (for example, stylistic considerations, great visibility), to symbolic (emotions as markers of wildness, intended to evoke reactions, having an instructive function). The survey of monuments shows that, while the ‘symbolic’ values of emotionality were prioritised, they were enforced through the use of practical devices such as displaying them on areas with greater visibility and accessibility. Form and function were intertwined. Emotions were not to be taken lightly, considering the effect they could have on the viewers. For this reason, their use is carefully prescribed, inserted in narratives, and has a didactic purpose. BIBLIOGRAPHY Amberger-Lahrman, M. (1996) Anatomie und Physiognomie in der hellenistischen Plastik: dargestellt am Pergamonaltar, Mainz. Andreae, B. (1994) Praetorium Speluncae: Tiberius und Ovid in Sperlonga, Mainz. Ashmole, B. and N. Yalouris (1967) Olympia: The Sculptures of the Temple of Zeus, London. Bartch, S. (1998) Ars and the Man: The Politics of Art in Virgil’s Aeneid, Classical Philology 93, 322–342. Baumeister, P. (2007) Der Fries des Hekateions von Lagina: Neue Untersuchungen zu Monument und Kontext, Istanbul. Bayer, E. (1983) Fischerbilder in den hellenistischen Plastik, Bonn. Beazley Archive Pottery Database: http://www.beazley.ox.ac.uk/pottery/default.htm. Bergemann, J. (2007) Attic Grave Reliefs and Portrait Sculpture in Fourth-Century Athens, in P. Schulz and R. von der Hoff (eds.), Early Hellenistic Portraiture, New York/Cambridge, 34– 46. Bieber, M. (1964) Alexander the Great in Greek and Roman Art, Chicago. Boardman, J. (1985) Greek Sculpture: The Classical Period, London. ––– (1995) Greek Sculpture: The Late Classical Period, London. Bol, C. (ed.) (2007) Die Geschichte der antike Bildhauerkunst III. Hellenistische Plastik, Mainz. Borbein, A. (1975) Die Ara Pacis Augusti. Geschichtliche Wirklichkeit und Programm, Jahrbuch des Deutschen Archäologischen Instituts 90, 242–266. Brennan, T. (1998) The Old Stoic Theory of Emotions, in J. Sihvola and T. Engberg-Pedersen (eds.), The Emotions in Hellenistic Philosophy, Dordrecht/London, 21–70.

308

Olympia Bobou

Brommer, F. (1967) Die Metopen des Parthenon, Mainz. ––– (1979) Die Pathenon-Skulpturen, Mainz. Budelmann, F. (2007) The Reception of Sophocles’ Representation of Physical Pain, American Journal of Philology 128, 443–467. Chaniotis, A. (ed.) (2012) Unveiling Emotions. Sources and Methods for the Study of Emotions in the Greek World, Stuttgart. Cohen, B. (2000) Not the Classical Ideal: Athens and the Construction of the Other in Greek Art, Leiden. Conlin, D. A. (1997) The Artists of the Ara Pacis: The Process of Hellenization in Roman Relief Sculpture, Chapel Hill/London. Contreras Flores, V. (2005) ARTNATOMY/ARTNATOMIA: www.artnatomia.net. Cook, B. F. (2005) Relief Sculpture of the Mausoleum at Halicarnassus, Oxford. Cooper, F. A. (1996) The Temple of Apollo Bassitas 1.3. The Architecture, Princeton. Delivorrias, A. (2008) The Parthenon Frieze: Problems, Challenges, Interpretations, Athens. Dihle, A. (2000) Die Philosophie der Barbaren, in Hölscher (ed.) 2000, 183–203. Dreyfus, R. et al. (1996/97) Pergamon: The Telephos Frieze from the Great Altar, San Francisco. Ekman, P. and W. V. Friesen (2001) Facial Signs of Emotional Experience, in W. G. Parrott (ed.), Emotions in Social Psychology, Philadelphia, 255–264. Fairclough, H. R. (1906) Vergil’s Relations to Graeco-Roman Art, The Classical Journal 2, 59– 68. Fortenbauch, W. G. (2008) Aristotle and Theophrastus on Emotions, in J. T. Fitzgerald (ed.), Passions and Moral Progress in Greco-Roman Thought, London. Giuliani, L. (2000) Die Giganten als Gegenbilder der attischer Bürger im 6. und 5. Jarhundert v. Chr., in Hölscher (ed.) 2000, 263–286. Graver, M. R. (2007) Stoicism and Emotion, Chicago/London. Güssinger, R., V. Kästner, and A. Scholl (eds.) (2011) Pergamon – Panorama der antiken Metropole, Petersberg. Hampe, R. (1972) Sperlonga und Vergil, Mainz. Hanson, V. D. (2000) Hoplite Battle as Ancient Greek Warfare, in H. van Wees (ed.), War and Violence in Ancient Greece, London, 201–232. Harris, W. V. (2001) Restraining Rage, Cambridge, Mass. Heinemann, A. (2000) Bilderspiele beim Gelage. Symposiast und Satyr im Athen des 5. Jahrhunderts v. Chr., in Hölscher (ed.) 2000, 321–349. Hermann, W. (1910) Die Friese des Grossen Altars, Berlin. Highbarger, E. L. (1942) Virgil and Roman Art, Classical Weekly 36, 87–89. Himmelmann, N. (1980) Über Hirten-Genre in der antiken Kunst, Opladen. Hölscher, T. (ed.) (2000a) Gegenwelten zu den Kulturen der Griechen und der Römer in der Antike, Munich-Leipzig. ––– (2000b) Feindwelten-Glückswelten: Perser, Kentauren und Amazonen, in Hölscher (ed.) 2000, 287–320. Hoff, R von der. (2011) Bildnisse der Attaliden, in Güssinger, Kästner, and Scholl (eds.) 2011, 122–130. Howard, S. (1989) Laocoön Rerestored, American Journal of Archaeology 93, 417–422. Humann, C. (1959) Der Pergamon Altar endeckt, beschrieben und gezeichnet, Dortmund. Irwin, T. H. (1998) Stoic Inhumanity, in J. Sihvola and T. Engberg-Pedersen (eds.), The Emotions in Hellenistic Philosophy, Dordrecht/London. Jenkins, I. (2006) Greek Architecture and its Sculpture, New York. ––– (2007) The Parthenon Sculptures in the British Museum, London. Jones, F. (2011) Virgil’s Gardens: The Nature of Bucolic Space, London. Junghölter, U. (1989) Zur Komposition der Lagina-Friese und zur Deutung des Nordfrieses, Frankfurt. Junker, K. (2012) Interpreting the Images of Greek Myths, Cambridge.

Emotionality in Greek Art

309

Kähler, H. (1948) Der grosse Fries von Pergamon, Berlin. Kaltsas, N. (2002) Sculpture in the National Archaeological Museum, Los Angeles. Kleiner, D. E. E. (1978) The Great Friezes of the Ara Pacis Augustae. Greek Sources, Roman Derivatives, and Augustan Social Policy, Mélanges de l’École Française de Rome 90, 753–785. Knuuttila, S. (2004) Emotions in Ancient and Medieval Philosophy, Oxford. Konstan, D. (2001) Pity Transformed, London. ––– (2006) The Emotions of the Ancient Greeks, Toronto/London. Kossatz-Deismann, A. (1981), s.v. Achilleus, in Lexicon Iconographicum Mythologiae Classicae, vol. I, 37-200. Kunze, C. (1999) Verkannte Götterfreunde, Mitteilungen des Deutschen Archäologischen Instituts, Römische Abteilung 106, 73–82. Laubscher, H. P. (1982) Fischer- und Landleute, Mainz. Lucchese, C. (2009) Il Mausoleo di Alicarnasso e i suoi maestri, Rome. Madigan, B. C. (1992) The Temple of Apollo Bassitas 2. The Sculpture, Princeton. Massa-Pairault, F.-H. (2007) La gigantomachie de Pergame ou l’image du monde, Athens. Masséglia, J. (2012a) Emotions and Archaeological Sources: A Methodological Introduction, in Chaniotis (ed.) 2012, 131–150. ––– (2012b) Reasons to be Cheerful? Conflicting Emotions in the Drunken Old Women of Munich and Rome, in Chaniotis (ed.) 2012, 413–430. Meyer, H. (1996) The Terme Ruler. An Understudied Masterpiece and the School of Lysippos, Bullettino della Commissione Archeologica Comunale di Roma 97, 125–148. Meyer, M., and N. Brüggemann (2007) Kore und Kouros, Vienna. Morelli, A. and G. Morelli (1977) Anatomia per gli artisti, Faenza. Moreno, P. (1995) Lisippo: l’arte e la fortuna, Milan. Osanna, M. (2000) Zwischen Dorern, Ionern und Indigenen: Die Achäer und die Anderen im archaischen Großgriechenland, in Hölscher (ed.) 2000, 245–262. Pearson, A. C., R. C. Jebb, and W. G. Hedlam (1917) The Fragments of Sophocles. Volume 2, Cambridge. Pozzi, E. (1991) Il Toro Farnese. La ‘Montagna di Marmo’ tra Roma e Napoli, Naples. Prittwitz, H.-H. von (2007) Die Hellenistische Plastik von 160 bis 120 v. Chr., in Bol (ed.) 2007, 241–271. Queyrel, F. (2005) L’autel du Pergame: Image et pouvoir en Grèce d’Asie, Paris. Richter, G. M. A. (1968) Korai, London. ––– (1970) Kouroi, London. Ridgway, B. S. (1981) Fifth Century Styles in Greek Sculpture, Princeton. Rohde, E. (1981) Pergamon: Burgberg und Altar, Berlin. Rolley, C. (1994) La sculpture grecque I. Des origines au milieu du Ve siècle, Paris. ––– (1999) La sculpture grecque II. La période classique, Paris. Rossini, O. (2007) Ara Pacis, Milan. Sassi, M. M. (2001) The Science of Man in Ancient Greece, Chicago. Schnapp, A. (2000) Pourquoi les Barbares n’ont-ils point d’images?, in Hölscher (ed.) 2000, 205– 216. Schneider, R.M. (2000) Lust und Loyalität. Satyrstatuen in hellenistischer Zeit, in Hölscher (ed.) 2000, 351–389. Schober, A. (1933) Der Fries des Hekateions von Lagina, Baden. Schraudolph, H. (2007) Beispiele hellenistischer Plastik der Zeit zwischen 190 und 160 v.Chr., in Bol (ed.) 2007, 189–239. Schupbach, W. (2009) Laocoon and the Expression of Pain. Wellcome Library for the History and Understanding of Medicine. http://www.wellcome.ac.uk/en/pain/microsite/culture3.html [accessed 29 May 2012].

310

Olympia Bobou

Shaver, P., J. Schwarz, D. Kirson, and C. O’Connor (2001) Emotion Knowledge: Further Exploration of a Prototype Approach, in W. G. Parrott (ed.), Emotions in Social Psychology, Philadelphia, 26–56. Simon, E. (1971) Pergamon und Hesiod, Mainz. Smith, R. R. R. (1988) Hellenistic Royal Portraits, Oxford. ––– (1991) Hellenistic Sculpture, London. Squire, M. (2009) Image and Text in Graeco-Roman Antiquity, Cambridge/New York. Stewart, A. (1977) Skopas of Paros, Park Ridge. ––– (1993) Faces of Power: Alexander’s Image and Hellenistic Politics, Berkeley/Oxford. ––– (1997) Art, Desire and the Body in Ancient Greece, Cambridge. ––– (2004) Attalos, Athens and the Acropolis, Cambridge. Stiebert, M. (2004) The Poetics of Appearance in the Attic Korai, Austin, TX. Trimble, J. (2011) Women and Visual Replication, Cambridge. van Wees, H. (2000) The Development of the Hoplite Phalanx. Iconography and Reality in the Seventh Century, in H. van Wees (ed.), War and Violence in Ancient Greece, London, 125– 166. Viano, C. (2003) Competitive Emotions and Thymos in Aristotle’s Rhetoric, in D. Konstan and N. K. Rutter (eds.), Envy, Spite and Jealousy, Edinburgh, 85–97. Vorster, C. (2007) Die Plastik des späten Hellenismus – Porträts und rundplastische Gruppen, in Bol (ed.) 2007, 273–330. Weis, H. A. (2000) Odysseus at Sperlonga: Hellenistic Hero or Roman Heroic Foil?, in N. T. de Grummond and B. S. Ridgway (eds.), From Pergamon to Sperlonga: Sculpture and Context, Berkeley/London, 111–165. Wrede, H. (1991) Matronen im Kult des Dionysos, Mitteilungen des Deutschen Archäologischen Instituts, Römische Abteilung 98, 163–188. Zanker, P. (1989) Die Trunkene Alte, Frankfurt.

PICTURE CREDITS Figure 1: Figure 2:

Figure 3:

Figure 4: Figure 5:

Figure 6:

Figure 7:

Figure 8:

Red-figure volute-krater by the Berlin Painter depicting Achilles fighting Hector, c. 490-460 BCE. London, British Museum. Credits: © Trustees of the British Museum. Hydria by the Kleophrades Painter showing the Sack of Troy, c. 480 BCE. Naples, Museo Archeologico Nazionale. Credits: © Soprintendenza Speciale per i Beni Archeologici di Napoli e Pompei. Cup by the Foundry Painter showing sculptors working on a statue, c. 490–480 BCE. Berlin, Pergamon Museum. Credits: © b p k / Antikensammlung, Staatliche Museen zu Berlin. Photo: Johannes Laurentius. Detail of a Persian warrior from the Lesser Attalid Group, mid-second century BCE. Rome, Musei Vaticani. Credits: © Musei Vaticani. Photo: author. Detail of a Persian warrior from the Lesser Attalid Group, mid-second century BCE. Musée Granet, Aix-en- Provence. Credits: © DAI Rome, neg. nr. D-DAI-ROM39.30. Odysseus’ Companion carrying a wineskin from the Blinding of Polyphemus group in Sperlonga, first century BCE/CE. Museo Archeologico Nazionale di Sperlonga. Credits: © DAI Rome, neg. nr. D-DAI-ROM-65.1897. Ajax (detail) from the so-called Pasquino Group, second century BCE. Rome, Musei Vaticani, Museo Pio Clementino. Credits: © DAI Rome, neg. nr. D-DAI-ROM 96VAT2524B. Kneeling Gaul (detail) from the Lesser Attalid Group, mid-second century BCE. Venice, Museo Archeologico Nazionale. Credits: © DAI Rome, neg. nr. D-DAIROM-68.5023.

Emotionality in Greek Art Figure 9:

Figure 10:

Figure 11: Figure 12:

Figure 13:

Figure 14:

Figure 15:

Figure 16: Figure 17: Figure 18: Figure 19:

Figure 20:

311

Odysseus (detail) from the Blinding of Polyphemus group in Sperlonga, first century BCE/CE, Museo Archeologico Nazionale di Sperlonga. Credits: © DAI Rome, neg. nr. D-DAI-ROM- 65.94. Statue of an Old Fisherman (detail of cast) from Aphrodisias, late third century BCE. Oxford, Ashmolean Museum. Credits: © Cast Gallery, Ashmolean Museum. Photo: David Gowers. Hanging Marsyas, second century BCE. Rome, Musei Capitolini. Credits: ©Archivio Fotografico dei Musei Capitolini. Falling Gaul (detail) from the Lesser Attalid Group, mid-second century BCE. Venice, Museo Archeologico Nazionale. Credits: © Concessione del Ministero per i Beni e la Attività Culturale. Details of figures from the Pergamon Altar, mid-second century BCE. Berlin, Pergamon Museum. 13a: Alkyoneus; 13b: Ge; 13c: Opponent of Dione; 13d: Tityos; 13e: Opponent of Triton; 13f: Opponent of Doris; 13g: Giant near Themis.13h: Opponent of Apollo. Credits: © b p k / Antikensammlung, Staatliche Museen zu Berlin. Photos: Johannes Laurentius (figures 13a–13e, 13g–13h) and Jürgen Liepe (figure 13f). Details of figures of the Laokoön Group, second century BCE. Rome, Musei Vaticani. Credits: © DAI Rome. 14a: neg. no. D-DAI-ROM- 33.1313. 14b: neg. no. DDAI-ROM- 64.904. 14c: neg. nr. D-DAI-ROM- 64.905. Details of figures from the Pergamon Altar, mid-second century BCE. Berlin, Pergamon Museum. 15a: Porphyrion; 15b: Biting giant; 15c: Opponent of Artemis; 15d: Opponent of Phoebe; 15e: Opponent of Apollo; 15f: Opponent of one of the Moirai. Credits: © b p k / Antikensammlung, Staatliche Museen zu Berlin. Photos: Johannes Laurentius (figures 15a–b, 15d–f) and Jürgen Liepe (figure 15c). Drunken Old Woman (detail), late third century BCE. Rome, Musei Capitolini. Credits: © Archivio Fotografico dei Musei Capitolini. Young Centaur (detail), late third–early second century BCE. Rome, Musei Capitolini. Credits: © Archivio Fotografico dei Musei Capitolini. Young nymph from the Invitation to Dance group (detail). Oxford, Ashmolean Museum. Credits: © Cast Gallery, Ashmolean Museum. Zeus overpowering his opponents, Pergamon Altar, mid-second century BCE. Berlin, Pergamon Museum. Credits: © b p k / Antikensammlung, Staatliche Museen zu Berlin. Photo: Johannes Laurentius. Athena defeats Alkyoneus, Pergamon Altar, mid-second century BCE. Berlin, Pergamon Museum. Credits: © b p k / Antikensammlung, Staatliche Museen zu Berlin. Photo: Johannes Laurentius.

FEELING LOW Social Status and Emotional Display in Hellenistic Art Jane Masséglia 1 INTRODUCTION In September 2008 an article appeared in the Greek philosophical magazine Cogito, which centred on a question raised by Yannis Tsividis, Professor of Electrical Engineering at Columbia University in New York.1 The question was a seemingly straightforward one: Why is laughter almost non-existent in ancient Greek sculpture? It prompted a number of different answers from distinguished scholars, some of them providing sociological answers, some art historical, some technical, and often with support from ancient texts. The article caught my attention for three reasons: firstly, because with only a single exception (Brunilde S. Ridgway), the scholars whose answers were reprinted in the article treated ‘ancient Greek sculpture’ as a single, unchanging entity; secondly, that for some of the respondents, the suppression of emotions was philosophically associated with idealisation and propriety, and so inherently with socially ‘desirable’ (and, implicitly, educated) individuals; and finally, that the methodological issues raised by Tsividis’ question could apply equally to the study of expressions of emotions in art, not simply laughter. In addition to my own interest in the body language protocols of Hellenistic society, these three considerations combined to give rise to this study. With respect to the first point, this general notion of ‘ancient Greek Sculpture’, there are obvious caveats. Those familiar with either the sculpture or the literature mentioned in the article (which ranged from kouroi to Hellenistic art, and from Homer to Aristotle respectively) will know that 700 years makes a huge difference to the appearance, as well as to the social significance of the objects. And so to search for explanations for the art of the Archaic period in the works of Aristotle is not without its problems; likewise using the plays of fifth-century BCE Athens to understand the sculpture of the Hellenistic period. If we are to understand the emotional expression of any art material, the textual support needs to come from a historically and socially relevant context. Certainly, the 700-year generalisation about ‘ancient Greek sculpture’ in the Cogito article may be excused on the grounds that the respondents were simply using the phrasing of the question itself. Or, more probably, that this generalisation was suitable for the in1

Kindi 2008.

314

Jane Masséglia

tended readership of Cogito, the layman philosopher. But within this more specialised collection of papers, we can afford to be slightly more particular in our contextualisation. The material discussed here is from the Hellenistic period (c. 330-30 BCE), the time (amongst other things) of Alexander’s successors, of kings and courts, of civic benefactions and public honours, new religious festivals, expensive education, philosophical schools, and a new visibility for women of status. And while the Hellenistic world can be said to encompass areas as different as modern Libya and Afghanistan, Bulgaria and Egypt, I will concentrate largely on figurative images from the Mediterranean and Aegean regions. 2 EMOTIONAL EXPRESSION IN FACE AND BODY One of the most eye-catching remarks in Vasso Kindi’s article was that by Brunilde Ridgway, who observed that Hellenistic art was different from Classical: ‘Even the Hellenistic period, for all its alleged realism, may show deformity ... but not true emotion.’ What comprises ‘true emotion’ in an artificially constructed figure is hard to imagine. Perhaps the intention was to suggest that the depictions were somehow unconvincing, or unlike the displays of emotions by people in the real world. Whatever reservations I might have about Ridgway’s statement, it does prompt important considerations about emotions in ancient art in general: emotional expressions can never be entirely convincing in ancient art, because they are static, snap-shot images, while emotional expressions in real life are composed of much longer sequences of actions and gestures, and can be accompanied by additional information such as language, tone of voice, and a complex web of contextual nuances. If we are to talk of emotional expression in Hellenistic art we must first be aware that there was a conventionalised repertoire used by artists that does not correspond absolutely to real-life behaviour. It had only to be recognisable, and its use and interpretation were governed by cultural convention and familiarity. But we, the modern viewers, are viewing this ancient art on the basis of very different cultural conventions, conventions that prioritise different physical processes in the communication of emotion. In the United Kingdom, for example, our politicians stand behind lecterns, occasionally raising a hand or fist, but public speaking is not a ‘whole body’ performance. And while purely corporeal means of communication such as dance are undergoing a revival, appreciation is largely focussed on technical complexity. ‘Interpretive dance’ (that which ‘channels’ the emotions of a musical score) has become a synonym for pretentious nonsense. The face is for many of us the bare minimum of emotional display (to which the rapid rise of the emoticon is an interesting concomitant). But emotion is not simply conveyed through the face. It is expressed through the whole body, and even through orientation and proximity to other individuals and objects. That different cultures may strike a different balance in the use of face and body to communicate emotions is an important concession if we are to understand Hellenistic sculpture. Even figures with idealised, passive faces (often labelled ‘unemotional’) can

Feeling Low: Social Status and Emotional Display in Hellenistic Art

315

communicate their emotions very successfully with their bodies. Mythological art demonstrates this particularly effectively: if we take the well-known example of the so-called ‘Slipper Slapper’ group for example, in which Aphrodite raises her sandal to fend off the advancing Pan,2 we see the closed mouth and relaxed brows conventionally employed in images of ‘good’ divinities (giants, centaurs, and other forces of chaos have their own rules). But the body language is highly successful in communicating emotions. From the flexing of muscles and the positioning of arms and legs, we understand precisely how each party feels about the other. From this alone we can see that, in studying emotional expressions in ancient art, we need to look beyond facial cues, and to look at body language, orientation and proximity as some of the primary means of communication.3 3 SELF-CONTROL AND THE HELLENISTIC ELITE Historians of the Hellenistic Age have rightly observed the importance of discretion and self-control in the philosophical literature of the period. These texts were not simply abstract treatises for discussion by specialists, but underpinned the education of Hellenistic elite males, in particular in public speaking, the defining means of self-presentation among one’s peers. Sōphrosyne, in particular, is the term often used for this kind of self-restraint and moderation. Through Homer, Theognis, Aeschylus, Plato, Aristotle, and Hellenistic oratory, it gradually took on a political meaning, representing a form of shrewdness which meant not only the regulation of the self, but the successful administration of public affairs, and the ability to make the right hoices.4 In particular, the late fourth century BCE saw an increase (in authors including Aristotle, Demosthenes, Aeschines, and Isocrates)5 in the use of sōphrosyne in opposition to hybris,6 that impulsive and inappropriate behaviour which resulted from a lack of discipline.7 Considering the great value set on a philosophical education and being a useful member of the citizen body, it is not surprising that sōphrosyne should manifest itself in the public images of the same group. This phenomenon has already been treated by the art historian Paul Zanker in his study of the Hellenistic grave reliefs from the city of Smyrna (modern Izmir).8 Among the highly standardised 2 3 4 5 6 7 8

From House of the Poseidoniasts of Berytos, Delos. Athens NAM inv. 3335. Kaltsas 2002, no. 617. Cf. Dillon 2007, 69, on the importance of the body in portraiture. North 1979, 89–91, 135, and passim. On its meaning and development up to and including Plato, see Rademaker 2005. E.g. Aristotle, Politics 1334a25–28; Demosthenes 24 (Against Timokrates) 75; Aeschines, Against Ktesiphon 168; Isocrates, On the Peace 119. North 1979, 102f. On an earlier, Archaic precedent for this opposition, see Rademaker 2005, 76–85. Cf. Aeschines, Aigainst Timarchos 7, 20, 139, on sōphronizein as the disciplining function of the Law. Zanker 1993; Burn 2004, 52: ‘brain plus moral responsibility for [images of] the statesman;’ cf. Zanker’s 1995 study of ‘celebrity’ intellectuals.

316

Jane Masséglia

repertoire of images, we find a number of corporeal conventions intended to communicate this personal control, and many of these extend beyond the Smyrnean grave reliefs, appearing in funerary and honorific civic art across the Hellenistic world. Among these body language conventions is the ‘arm sling’ which is recognisable in portrait statues of orators, poets, and even in a pot painting showing individuals exchanging goods,9 as a kind of uniform of civic affairs. Other stock poses that show restrained arm gestures include what I term elsewhere the ‘Lapel Hold’, the ‘Belt Roll’, and the ‘Wreath Hold’.10 Each of these poses essentially represents a variation of the same theme: a static, forwardorientated figure wearing a himation, with passive facial expression, wide stance, and at least one arm bent at the elbow enhancing the impression of width across the shoulders. Indeed, so conventionalised was the repertoire for images of Hellenistic elite men that many grave reliefs and honorific images are strikingly similar. And even when confronted with a more complex image, such as the relief scene of a votive procession now in the Louvre (Figure 1), a closer inspection reveals that even this is a composite of the various stock types.

Figure 1. Votive relief from Attica (c. 350-300 BCE) showing men and youths approaching Demeter.

Similarly, funerary and honorific images of women of the Hellenistic civic elite are largely forward-orientated, facially passive, draped (in this case heavily), and seemingly emotionally contained.11 Indeed the quintessentially Hellenistic pose for women, the misleadingly but conventionally named pudicitia, with one arm 9 10 11

Paris, Louvre inv. CA1852 = Beazley Archive 206122. Masséglia forthcoming. Dillon 2007, 63f., 77.

Feeling Low: Social Status and Emotional Display in Hellenistic Art

317

wrapped tightly around the waist and the other raised to face or veil (Figure 2), is usually discussed by art historians, among them R. R. R. Smith, Paul Zanker, and Sanne Houby-Nielsen,12 as embodying a female form of sōphrosyne.13

Figure 2. Statue of Baebia from Magnesia on the Maeander (late second century BCE).

The emotional passivity of these figures, both male and female, is particularly striking in the context of grave reliefs. Expressions of loss, sadness, etc., are almost imperceptible in many Hellenistic examples. The uncertainty surrounding the precise meaning of the common hand-to-face gesture, whether it represents thinking or mourning (or both), or the precise emotional connotations of dexiosis 12 13

Smith 1991, 84; Zanker 1993, 225; Houby-Nielsen 1997, 242f. A highly popular term in Greek funerary inscriptions for women. cf. North 1966, 253; 1977, 40 (on gendered usage). Theocritus 28.14, with sōphrōn as simply ‘respectable’; Menander, Epitrepontes 702 and Samia 129, meaning ‘chaste’ for either gender. North 1966, 243, notes 1f.

318

Jane Masséglia

(the handclasp), is testament to the subtlety of the emotional expression, at least to our eyes. Emotional passivity is also starkly apparent when we look at grave reliefs showing more than one standing figure. Hellenistic grave reliefs present a new fashion for individuals shown side-by-side, but entirely self-contained, with little or no interaction between figures. The mere inclusion of the figures in the same relief is often the only apparent means of expressing emotional relationships. The popularity of this new frontal, isolated arrangement in grave reliefs certainly stresses the self-control narrative inherent in the individual figures, by offering an alternative to the more sentimental, interactive dexiosis formula.14 But there is also a second, highly significant motive behind this choice: These frontorientated grave reliefs are intended to mimic the frontal-facing statues awarded to civic benefactors from their grateful polis. Not only does the language of grave inscriptions often ape that of honorific decrees, but as Paul Zanker has pointed out in the case of the Smyrna reliefs, the images could also include carved wreaths (imitating those awarded by the polis to her great citizens) and the words ho demos (‘the people [supply: award this to X]’), and, in some cases, the statue base itself is also imitated.15 Emotional restraint characterised the honorific portraits of the civic elite, and subsequently in the reliefs that imitated them, often at the expense of any emotional reference to the funerary context of the latter. 4 STATUS AND EMOTIONS: HELLENISTIC SNOBBERY? If history is dominated by the study of the wealthy and those in power, then history of art shows a similar tendency towards the honorific and commemorative pieces connected with named individuals and those socially significant enough to acquire these images. These are the individuals on which notions of ‘the citizen intellectual’ and the ‘modest wife’ are based and that dominate the study of Hellenistic society. But if we look at the images of the lower classes, or rather the ‘other’ classes, we find a far greater range both of body language narratives and of emotions being displayed: a powerful Gaul, supporting the body of his dead wife in his arms, pierces his own chest in anguish;16 a cripple with his spine projecting like a huge dorsal fin, sits on the floor, his shoulders drooping and his mouth grimacing in discomfort (Figure 3); a peasant boy, losing at a game of knucklebones, grabs the hand of his opponent and angrily sinks his teeth into it;17 a dwarf dancer, providing exotic entertainment of the wealthy, throws his head back in exuberance (Fig14

15 16 17

Dillon 2006, 66, rightly notes that the number of figures on grave reliefs in the Hellenistic period is less than the Classical, i.e. that the depiction of several generations of one family falls out of fashion. Zanker 1993, 214f. Large Gaul group, Rome, Terme inv. 8608. Smith 1991, fig. 118; cf. Burn 2004, 153f. Biter group, London, British Museum inv. GR 1805.7-3.7 (Sculpture 1756). Smith 1991, fig. 172.

Feeling Low: Social Status and Emotional Display in Hellenistic Art

319

ure 4b); an African jockey leans forward on his horse as he excitedly urges it on;18 a paidagogos wraps his arm affectionately around his charge who stands smiling beside him (Figure 5);19 a peasant puffs his cheeks out in pain and concentration as he removes a thorn from his foot – one should note the idealised face of the statue that this parodies;20 nurses cuddle and kiss their protégés;21 drunken old women are merry,22 very merry (Figure 6), and sometimes even nauseous. 23

Figure 3. Talismanic bronze figurine of a hunchback, of unknown provenance (mid-third century BCE).

18 19 20 21 22 23

The Artemisium Jockey, Athens National Museum inv. Athens National Musem 15177. Pollitt 1986, fig. 159. An emotional relationship parodied in the erotically charged Pan and Daphnis statue group, Museo Archeologico Nazionale di Napoli inv. 6329. Smith 1991, fig. 160. Peasant Spinario statuette, Berlin inv. TC 8626. Smith 1991, fig. 172; Monumental Spinario, London British Museum inv.1880.7-1.7 (Sculpture 1755). Nurse figurine, Athens National Museum inv. 3978. Pfisterer-Haas 1989, fig. 55. Cheerful Old Woman with wine jar, Dresden inv. ZV 1633. Pfisterer-Haas 1989, fig. 146; Masséglia 2012 (with bibliography). Nauseous Woman, Paris, Louvre 2936 (MNC 1916). Bieber 1961, fig. 586. For a detailed study of the emotions of the drunken old women figures, see Masséglia 2012.

320

Jane Masséglia

Figure 4: Figurine pair of dwarfs from the Madhia shipwreck (mid-second century BCE).

Figure 5 (left): Hellenistic terracotta figurine of smiling paidagogos and child (Bolschaia Blinitza). Figure 6 (right): ‘The Drunken Old Woman of Munich’. Roman copy of an original from the third century BCE.

Feeling Low: Social Status and Emotional Display in Hellenistic Art

321

This represents just a sample of the images usually grouped under art-historical headings such as ‘Genre Figures’, ‘Grotesques’, and ‘the Rococo’.24 But what they really are is the not-rich, non-elite people, with great variety of postures and emotional expression. What makes such lower class pieces particularly interesting for historians is that they were not commissioned by the groups that they represent. Instead they are projections of certain social expectations of the wealthier, art-buying element of society onto these figures. What we have, therefore, is a situation in which it appears that emotional display was more appropriate to lower status figures. Could this be the result of a different attitude towards emotions, as something more suitable for the ‘lower classes’? Does an education that involves the mastery of the self necessarily create a distinction between the ‘stiff upper lip’ of the elite and the unhindered emotional narratives of the others? Or are there other constructions at work here? In order to test the strength of the relationship between social status and emotional display, I would like to propose a number of additional factors, some interrelated, that should be born in mind when using figurative imagery as evidence in the study of emotions. 4.1 Specific Identity The first indication that such emotional restraint is not simply the result of status is that, beyond the corpus of grave reliefs and public statuary, we have many depictions of individuals of the well-to-do citizen class, that are emotionally expressive, and interactive. We may think particularly of the large corpus of Hellenistic Tanagras (mould-made, painted terracotta figurines), which is dominated by images of well-dressed ladies of leisure in colourful fabrics and jewellery engaged in a wide range of activities. Among those activities that never appear in honorific monuments and grave reliefs are their conspiratorial chats,25 their dandling babies,26 and, for the younger girls, cheerful games of ephedrismos.27 While in dress and status they are not dissimilar from the elite women in funerary and honorific statuary, their body language narrates emotions of affection and interpersonal relationships. What could account for this difference? One important distinction between Tanagra figurines and the emotionally restrained sculpture is that the little terracottas do not represent particular women. They are not linked, either through portrait likeness, or through an accompanying inscription, with a named individual. One reason that the emotional expressiveness and interaction that makes these little figures so appealing, even acceptable, is their anonymity. In the case of the Tanagra figurines, there are also other factors which may explain this emotional difference, some of which I will address 24 25 26 27

Pollitt 1986, 127. Terracotta group of seated women, London British Museum inv. 2274 (= C529). Woman with baby group, Berlin, Antikensammlung inv. TC 7946. Kleiner, Parlasca, and Linfert 1984, 43c. Ephedrismos group, Ny Carlseberg Glyptotek, inv. 904. Nielsen and Østergaard 1997, fig. 83.

322

Jane Masséglia

below (§ 4.6), but it is already clear that it cannot be explained away as a difference in status. 4.2 Locations of display and narrative context Another consideration regarding emotional expression might be the location or narrative environment in which the figure is supposed to be imagined. The figures in honorific statues and grave reliefs are set in public locations. Sometimes this is paralleled in the internal location of reliefs, with the occasional mock-statue base and even architectural details such as gymnasium doors and herms stressing the public nature of the scene. But even when the individual is shown at home and at leisure, they show an emotional restraint appropriate for the public context in which the image was viewed. Conversely, the well-dressed and interactive little Tanagras are often shown in domestic scenes, such as the young women sitting on a couch in the terracotta group in the British Museum.28 The recent excavations in Priene have revealed rooms with shelves, niches, and attachments that seem to have been intended for figurines.29 If these rooms are representative of wider fashions, then these Tanagras were also set up in a domestic environment. Tanagras were also popular in inclusion in burials, again, not on public display. And the same can be argued for the highly expressive little entertainer figures that are so often described as ‘grotesques’. These figures are themselves too small for public display, and their own internal narratives suggest a similarly private context: dwarf dancers like those in Figure 4 should be imagined in a symposion environment, such as we see in a vase of the Classical period showing a dwarf dancing on a symposion table.30 Similarly, if the much-discussed Drunken Old Woman type, with copies in Munich and Rome (Figure 6), is to be imagined engaged in a religious celebration in a sanctuary, then her expressions too (whether or not they are ‘too much’ or not) result from the celebratory context afforded by her environment. In these cases, it is not the status of the figure that dictates the emotional expression, but the fact that certain behaviour was more appropriate to certain locations and contexts. What is significant is that the elite chose to internally replicate the restraint of the public environment within their honorific statuary and funeral commemorations. The corpus of Classical grave reliefs, stressing domestic locations and familial emotions, shows that this had not always been the case. 4.3 Reliability of the evidence A second reservation about attributing emotional restraint simply to status is the misleading state of the evidence. If we consider the pudicitia pose for example, the pose that appears in most Hellenistic art handbooks to illustrate the citizen 28 29 30

As above, note 25. E.g. in House 15. Rumscheid 2006, 60. Red-figure krater, Zurich Arete Gallery. Dasen 1993, pl. 55.1.

Feeling Low: Social Status and Emotional Display in Hellenistic Art

323

woman, we see well-preserved examples such as Baebia from Magnesia (Figure 2), or Kleopatra from Delos,31 with arms drawn in out of harm’s way. And in grave reliefs too, the closed body form of the pudicitia makes it eminently wellsuited to this kind of carving. It is a neat, compact form without any awkward limbs extended away from the body. This must have made it a popular choice for artists and patrons as a less risky use of expensive stone. But, if we look beyond the pudicitia to the wider range of choices available for women in the freestanding honorific monuments, we can see that the impression of the handbooks is misleading: as J. Cordelia Eule has shown, there were other popular schemes, and two of them, exemplified by the named statues of Moschine and Nikeso,32 stood with their arms extended in a votive gesture. If we take the figure of Nikeso from Priene, for example, we see a badly damaged figure, without head or arms, with its potential for emotional expression greatly compromised. But the ancient viewer, seeing her complete, with arms held away from body, brightly painted and measuring over two metres tall, is unlikely to have thought of her as passive.33 The modern preference for the restrained pudicitia in the handbooks does indeed reflect their popularity.34 But it is not impossible that part of this popularity is founded on practicality (which led in turn to an established fashion), and that the format was further prioritised within the surviving evidence by the modern scholar searching for an attractive, well-preserved illustration. Again, status is not the issue here, but instead the likelihood of the female figures losing their arms prior to archaeological study. 4.4 Deliberate exception Even if we accept the general principle that emotional restraint was a socially prized quality, there is no reason to suppose that there were not also those who, for whatever reason, preferred a different kind of image. The grave relief of two young people (brother and sister or a young couple?), set up in around 100 BCE somewhere on the Ionian coast, is one such example (Figure 7).35 The relief shows an unusual tender gesture of affection, although one that from its clumsy execution was surely an experimental commission for the art31 32 33

34 35

Kleopatra beside her husband Dioskourides, Delos Museum inv. A 7763, A 7799, A7997a, in situ, House of Kleopatra (A 7763). Smith 1991, fig. 113. Moschine: Izmir inv. 582; Eule 2001, 53. Nikeso: Pergamon Museum, 1928; Smith 1991, fig. 111. Cf. Aeschines, Against Timarchos184, regarding Solon’s law that adulteresses were not permitted to dress up or attend public sacrifices. Honorific statues of well-dressed women set up in sanctuary spaces must have had connotations of the converse: that these were not adulteresses, but well-behaved and moderate. Zanker 1993, 222. The inscription is from the 2nd/3rd cent. CE, from the stele’s second use. I.Smyrna 376: Μηνώφιλος Ἀπολλονίδου κὲ Ἀπφιὰς Ἀπολλωνίου Τρυφαίνῃ τῇ θυγατρὶ καὶ Μηνωφίλῳ τῷ ὑιῷ, χρε (‘Menophilos of Apollonides and Apphias of Apollonios, to their daughter Tryphaina and their son Menophilos. Farewell’).

Jane Masséglia

324

ist.36 It is the only one of its kind that we have, combining the standard adult, restrained, front-orientated forms with the young woman’s raised hand resting on her partner’s shoulder. This relief represents an individual’s choice to diverge from the convention, not only representing emotional interaction between those depicted, but also simultaneously revealing the heightened emotional state of those that commissioned it: by breaking away from the norm, the commissioners were themselves expressing the nature of their loss as something that went beyond the ordinary.

Figure 7 (left). Stele of two young people, probably from a site on the Ionian coast (c. 100 BCE). Figure 8 (right). Grave stele from Rhodes, depicting two women embracing (second century BCE).

4.5 Local Fashion The effect of exception can also be exploited by larger groups in forming a sense of local or regional identity. On the island of Delos, for example, the handshake (dexiosis) scheme, physically and emotionally connecting loved ones, remained

36

Zanker 1993, 227.

Feeling Low: Social Status and Emotional Display in Hellenistic Art

325

popular even into the Hellenistic period.37 But the island of Rhodes, for example, had within its local repertoire an entirely different emotional response to commemoration. In what appears as an inversion of the wider fashions of the Ionian coast, some second-century BCE grave reliefs of Rhodes show couples, sometimes man and wife, sometimes two women, embracing each other in profile (Figure 8). Their collapsed shoulders, heads nuzzling in against one another’s necks, and the women fully muffled in heavy cloaks, narrate emphatic expressions of love and grief. The restrained, frontal-facing type also existed on the island, but this ‘embrace-type’ stood alongside it as a very different alternative. Choosing the local type may have represented an expression of local pride and identity, or perhaps it did appeal to a genuinely more emotional clientele, but importantly, there appears to be no difference in status between those who chose the front-orientated types at Smyrna, the handshakes at Delos, and the embrace type at Rhodes. 4.6 Function The distinctions that I have drawn between images with portrait and generic identity, between those for public and private display, and those with conventional and exceptional elements, could all in some respects be treated under the heading of function.38 The intended use of an object nearly always leaves a mark on its appearance. I would like to focus more specifically here on four different functions that figurative art could serve, and how this relates to the emotional expression of the individuals depicted: Commemoration, Votive Offering, Talisman, and Tool. Commemoration I have already described how the location of display affected the emotional expression of the images, but without making explicit why it makes such a difference. The answer lies in the fact that certain locations were reserved for certain kinds of art, and the location for commemorative images (that is either honorific figures or grave reliefs) was traditionally a highly visible civic space such as the agora, the theatre, a sanctuary or, in the case of funerary art, a cemetery. Being such a public display of individual tastes, set up alongside similar commemorations of others’ lives and deeds, it is significant that the Hellenistic attitude was, in general, highly conservative. Rather than seeking to be distinguished from other pieces, there appears to have been a general preference for conformity in commemorative art. The threat of public scrutiny in posterity seems to have encouraged an understandable ‘play it safe’ attitude towards figurative images, with the great majority focusing on the desirable connotation of emotional restraint and sōphrosyne that has already been discussed. And since only those with either suitable connections or financial means received such commemorative images, we

37 38

Zanker 1993, 229. Which was the (sensible, it seems to me) explanation offered by Andrew Stewart to answer Yannis Tsividis’ question concerning the apparent absence of laughter in Greek art.

326

Jane Masséglia

can see how this emotional restraint has become particularly associated with high status. Votive Offering Perhaps surprisingly, there does not appear to have been the same predilection for emotional restraint in images given as offerings in sanctuaries and graves, so long as they were not commemorative portrait images. If the Drunken Old Woman figure in Munich (Figure 6) was indeed originally a votive dedication, then her exuberance would be a wonderful example of this on the monumental scale, but the phenomenon is more obvious (and more secure) in those smaller votive items that were included within burials. Particularly interesting in this case are the terracotta figurines of paidagogoi and nurses (both low status groups), usually shown alongside their freeborn protégées, which were popular in children’s burials. In this corpus, we find a wide range of emotional expressions, both corporeal and facial, by both the elderly servant figures and the higher status children. We see smiles (Figure 5), grimaces of irritation, and patient resignation,39 as well as affectionate hugs and even kisses. The inclusion of these figures within graves, sometimes in large numbers,40 and even with multiple versions of the same figure group, suggests some sort of religious (or perhaps less dogmatically, superstitious) function. Significant too that the figures which dominate in child burials (far outweighing images of parents, such as the mother-and-baby groups) are of those individuals who were defined by their roles as guardians of children. This phenomenon can be explained by the desirability of these figures’ emotional expressions in fulfilling a kind of protective role for the deceased and a reassuring function for those placing them in the graves. In all of these votive objects, the emotional expression is related to the function of the object, whether celebration or protection, and indeed can be seen to enhance its efficacy. Talisman Another example of where emotional expression can be thought to enhance efficacy is in the use of objects for their talismanic properties, and in particular for warding off the Evil Eye. As John Clarke has examined in the Roman context in Looking at Laughter,41 there was an intimate connection in antiquity between emotions and protection from malign forces. To laugh in the face of danger, even to mock it, is a means to disperse its potency. This power can be harboured in art images, making them apotropaic, in two ways: firstly, figures can be shown behaving in an irreverent way, smiling and laughing and performing rude gestures. Secondly, as in the case of the hunchback in Figure 3, the body of the figure can be so strange, even deformed, as to encourage the viewers of the object to laugh 39 40 41

Grimacing paidagogos group, Berlin inv. TC 7084. Schulze, 1998, fig. 15.3; Resigned paidagogos group, from Pergamon, sine loco. Schulze, 1998, figs. 14.1–2. Schulze 1998, 46, 49: in Pella, tholos grave Θ was found to contain thirty-four examples of the same figurine group, including different firings. Clarke 2007, 63–81.

Feeling Low: Social Status and Emotional Display in Hellenistic Art

327

themselves, or to feel relief at their own comparative health. In some cases, these two methods can be combined, and we see deformed individuals such as dwarfs (Figure 4), in this case a pair of dancers, performing a talismanic ‘Tuck-for-luck’ dance42 with broad smiles and enthusiastic energy. The combination of emotionprovoking bodily appearance, and their own emotionally expressive behaviour, enhances their efficacy. Of course, deformity had significance for social status. It was desirable in entertainers, but not, it seems, in those who wished to engage in public life. Status was an important part of why these individuals were chosen, but it was not the reason they were depicted in moments of emotional expression. This instead was more closely connected with the function of the object. Tool Finally, it is worth considering the role that the ergonomics of the object could have on the emotional expression depicted. In the case of freestanding statuary, its ability to be seen and not collapse were really the only ergonomic considerations. But if we look at figurative utensils, we can observe the effect of practicality: the monumental Drunken Old Woman figure that we have already seen (Figure 6) was also replicated in a number of small, terracotta wine jugs including one that is roughly contemporary with the Hellenistic figure (Figure 9), and two later, North African examples from the second century CE.43

Figure 9. Wine jug of the ‘Drunken Old Women’ type, from Skyros (late second/early first century BCE). 42 43

Masséglia forthcoming. Roman jug from North Africa, Kunsthandel Deutschland inv. 1975. Salamonsen 1980, fig. 43a. Jug from Sousse, Utrecht University inv. BS 77.1; Salamonsen 1980, fig. 41a.

328

Jane Masséglia

Each of these little wine jars shares the same adaptation that compromises the degree of emotional expression: while in the monumental version, the old woman’s head is thrown back in drunken enjoyment, in the case of the jugs, the placing of the spout on the top of the head necessitates that the chin is lowered. We can see in the Skyros jug (Figure 9) that the flat top of the head provides a better location for this spout, and prevents unsightly encroachment onto the forehead. This, of course, affects the intensity of the emotional expression. To make a more plausible overall composition, the wild open mouth that accompanies the thrown-back head of the monumental version becomes a more modest smile in the Skyros jug. But so that these adaptations do not completely undermine the intended expression, the inscription on the bottom makes it explicit, essentially recalibrating our reading of the more subtle signals: ‘This old woman sits here full of joy holding her wine.’44 In comparing the monumental figure with the Skyros jug, we have no difference in status, indeed no difference in subject, and yet there is a different intensity of emotional expression because one is a utensil that must fulfil other requirements related to its use. 5 EMOTIONS IN HELLENISTIC ART: A TIME AND A PLACE I hope that the limitations of the propositions made above (§ 4), that emotional expression in Hellenistic art correlates with the status of the individual depicted, are now evident. Undoubtedly, great store was set by an appearance of emotional restraint in certain contexts, which rested heavily on notions of education and ‘access’, and this was showcased in portrait art that was for public consumption. But even then, as the embrace-reliefs of Rhodes exemplify, this was not always for everyone. Moreover, attempting to compare the emotional range of public art with that of low-status figures who served an entirely different purpose is not methodologically sound. Instead, I have tried to emphasise the importance of asking both social and archaeological questions about the material: What was the object for? Whom does it represent? How representative is the archaeological sample we have to work with? The discussion which followed the Cogito article was interesting and provocative but there were also a number of problems, already embedded in the question which presupposed homogeneity. These kinds of oversimplifications do not do justice to the material, either as art or historical evidence. Just as an epigrapher, papyrologist, or historian, the archaeologist must ask questions about audience, intention, and preservation when responding to images, images that, after all, are not photographic records, but contrived products in every sense. But this should not be disheartening. On the contrary, observing where and in what form conventions arose gives us precious insights into social values. In the case of Hellenistic material, the restrained images of the civic elite that were popular in places like Smyrna were nonetheless potent expressions of social anxiety and civic pride. The appearance of emotional relationships in places 44

IG XII.8.679: Γραὺς ἥδε οἰνοφόρος κε|χαρηµέ(νη ὧδ)ε κάθηται.

Feeling Low: Social Status and Emotional Display in Hellenistic Art

329

like Rhodes, in non-portrait images like the Tanagras, and among lower status individuals like the paidagogoi and nurses, does not suggest a blanket form of ‘emotional snobbery’, but rather a Hellenistic appreciation of the variety of emotional states and, above all, a finely tuned sense of occasion. BIBLIOGRAPHY Bieber, M. (1961) The Sculpture of the Hellenistic Age, New York (revised edition). Burn, L. (2004) Hellenistic Art: from Alexander the Great to Augustus, London. Clarke, J. R. (2007) Looking at Laughter: Humor, Power, and Transgression in Roman Visual Culture, 100 B.C. – A.D. 250, Berkeley. Connelly, J. B. (2007) Portrait of a Priestess: Women and Ritual in Ancient Greece, Princeton. Dasen, V. (1993) Dwarfs in Ancient Egypt and Greece, Oxford. Dillon, S. (2007) Portraits of Women in the Early Hellenistic Period, in R. von den Hoff and P. Schultz (eds.), Early Hellenistic Portraiture: Image, Style, Context, New York, 63–83. Eule, J. C. (2001) Hellenistische Bürgerinnen aus Kleinasien: weibliche Gewandstatuen in ihrem antiken Kontext, Istanbul. Fischer, J. (1998) Der Zwerg, der Phallos und der Buckel, Chronique d’Égypte 73, 327–361. Higgins, R. (1986) Tanagra and the Figurines, London. Himmelmann, N. (1983) Alexandria und der Realismus in der griechischen Kunst, Tübingen. Houby-Nielsen, S. (1997) Grave Gifts, Women, and Conventional Values in Hellenistic Athens, in P. Bilde (ed.), Conventional Values of the Hellenistic Greeks, Aarhus, 220–262. Kaltsas, N. (2002) Sculpture in the National Archaeological Museum, Athens, Athens. Kindi, V. (2008) Why is Laughter Almost Non-Existent in Ancient Greek Sculpture?, Cogito (Greece) 8 [reprinted in Eurozine 2008: http://www.eurozine.com/articles/2008-09-18-kindien.html (accessed 22 June 2010)]. Kleiner, G., K. Parlasca, and A. Linfert (1984) Tanagrafiguren: Untersuchungen zur hellenistischen Kunst und Geschichte, Berlin. Laubscher, H. P. (1982) Fischer und Landleute: Studien zur hellenistischen Genreplastik, Mainz. Masséglia, J. E. A. (2012) Reasons to be Cheerful? Conflicting Emotions in the Drunken Old Women of Munich and Rome, in A. Chaniotis (ed.), Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, Stuttgart, 413–430. ––– (forthcoming) Body Language in Hellenistic Art and Society, Oxford. Mehrabian, A. (1971) Silent Messages: Implicit Communication of Emotions and Attitudes, Belmont. Nielsen, A. M. and J. S. Østergaard (1997) Hellenism: Catalogue: The Eastern Mediterranean in the Hellenistic Period: Ny Carlsberg Glyptotek, Copenhagen. North, H. F. (1966) Sophrosyne: Self-knowledge and Self-restraint in Greek Literature, Ithaca NY. ––– (1977) The Mare, the Vixen, and the Bee. Sophrosyne as the Virtue of Women in Antiquity, Illinois Classical Studies 2, 35–48. ––– (1979) From Myth to Icon: Reflections of Greek Ethical Doctrine in Literature and Art, Ithaca NY. Pfisterer-Haas, S. (1989) Darstellungen alter Frauen in der griechischen Kunst, Frankfurt, 40–46. ––– (1994) Die bronzenen Zwergentänzer, in G. Hellenkemper Salies, H. von Prittwitz, and G. and G. Bauchhenss (eds.), Das Wrack: Der antike Schiffsfund von Mahdia, Cologne, 483 and 504. Pollitt, J. J. (1986) Art in the Hellenistic Age, Cambridge. Rademaker, A. (2005) Sophrosyne and the Rhetoric of Self-restraint: Polysemy and Persuasive Use of an Ancient Greek Value Term, Leiden. Rumscheid, F. (2006) Die figürlichen Terrakotten von Priene: Fundkontexte, Ikonographie und Funktion in Wohnhäusern und Heiligtümern im Licht antiker Parallelbefunde, Weisbaden.

330

Jane Masséglia

Salomonsen, J. W. (1980) Die Trunkenbold und die Trunkene Alte, Bulletin Antieke Beschaving 55, 65–135. Schulze, H. (1998) Ammen und Pädagogen: Sklavinnen und Sklaven als Erzieher in der antiken Kunst und Gesellschaft, Mainz am Rhein. Smith, R. R. R. (1991) Hellenistic Sculpture: a Handbook, London. van Bremen, R. (1996) The Limits of Participation: Woman and Civic Life in the Greek East in the Hellenistic and Roman Periods, Amsterdam. Zanker, P. (1993) The Hellenistic Grave Stelai from Smyrna: Identity and Self-image in the Polis, in A. W. Bulloch, E. S. Gruen, A. A. Long, and A. Stewart (eds.), Images and Ideologies: Self-definition in the Hellenistic World, Berkeley. ––– (1995) The Mask of Socrates: the Image of the Intellectual in Antiquity, translated by H. A. Shapiro, Oxford.

PICTURE CREDITS Figure 1:

Figure 2:

Figure 3:

Figure 4: Figure 5:

Figure 6: Figure 7:

Figure 8:

Figure 9:

Votive relief showing men and youths approaching Demeter. Attica; c. 350-300 BCE; marble; H 63 cm, W 109 cm. Paris, Louvre, Ma 756. Credit: © RMN (Musée du Louvre). Photo: Hervé Lewandowski. Statue of Baebia. Magnesia on the Maeader; late second century BCE; marble; H 215 cm. Istanbul Archaeological Museum, 605. Credit: Deutsches Archäologisches Institut. Photo: W.Schiele. Talismanic bronze figurine of a hunchback. Unknown provenance; mid-third century BCE; bronze; H 6.6 cm. Museum für Kunst und Gewerbe Hamburg, 1949.40. Credit: Museum für Kunst und Gewerbe Hamburg. Figurine pair of dwarfs from Madhia shipwreck; mid-second century BCE; bronze; 30 cm and 32 cm. Bardo Museum, F.213 & 215. Credit: LVR-Landesmuseum Bonn. Figurine of smiling paidagogos and child. Bolschaia Blinitza grave 4; Hellenistic; terracotta; H 14 cm. Hermitage, BB 173. Credit: © The State Hermitage Museum, St Petersburg. Photo: Yuri Molodkovets. ‘The Drunken Old Woman of Munich’. Roman copy of original from the third century BCE; marble; H 92 cm. Munich Glyptothek, 437. Photo: Author. Stele of two young people. Ionian coast (?); c. 100 BCE; marble; H. 136 cm. Vienna, Grinzing Cemetery. Photo: W. Kubitschek, Jahreshefte des Österreichischen Archäologischen Instituts 29 (1935), fig. 38. Grave stele depicting two women embracing. Rhodes, Korakonero Necropolis; second century BCE; marble; H 51 cm, W 30 cm, D 2 cm. Formerly Rhodes Arch. Museum. Photo: J. Kontis, Τὸ Ἔργον τῆς ἐν Ἀθήναις Ἀρχαιολογικῆς Ἑταιρείας 1961, 220, fig. 223. Wine jug of the ‘Drunken Old Women’ type. Skyros; late second/early first century BCE; terracotta; H 25.5cm. National Archaeological Museum, Athens, 2069. Credit: © Hellenic Ministry of Culture and Tourism /Archaeological Receipts Fund.

THE IMPRINT OF EMOTIONS SURROUNDING THE DEATH OF CHILDREN IN ANTIQUITY Chryssa Bourbou 1 THE BIOARCHAEOLOGY OF CHILDREN1 Both the best of mortals and those who are nobodies love children; they differ in material things; some have property and some do not; but the whole race is child-loving. Euripides, Hercules Furens 634–636.

In recent years, children, once invisible in archaeological studies, have come into scholarly consideration.2 These studies attempt to bring the social role of children to the fore, and to determine their interaction with the natural and cultural environment in which they lived. The treatment of the younger segment of the population in both life and death differs between societies. Although no strong evidence exists that any human society treats the death of a child lightly, many past and modern societies assign different values and meanings to dead offspring. Can emotions towards a dead child, such as fear, love, affection, grief, or despair be traced in the archaeological record? How can we detect and interpret the expression of emotions in the past other than in the testimonials of written records and artistic representations? Greek grave inscriptions, especially epigrams, often refer to the grief felt by parents and relatives for the death of a child.3 Papyri of the Hellenistic and Imperial periods hold special interest as they offer direct insight into how people in antiquity felt and dealt with everyday occurences and misfortunes.4 Oracular consultations, for example, illustrate worried parents’ fears about the fate of their unborn children; similarly, private letters among family members offer diverse references to parental affection for or disinterest in children. The letter of a soldier to his wife clearly demonstrates the mixture of his tenderness and worry when he writes: ‘as for the child, keep an eye on him as you would on an oil lamp, since I am worried about you’.5 1

2 3 4 5

I would like to thank Angelos Chaniotis for inviting me to contribute to this volume. I am also grateful to a number of friends and colleagues for our fruitful collaboration and long conversations on the role of children in antiquity: Mrs Euthymia Kataki, Professor Maria Liston, Professor Susan Rotroff, and Professor Petros Themelis. Scott 1999; Sofaer-Derevenski 2000; Kamp 2001; Dasen 2004; Bakke 2005; Baxter 2005a and 2005b; Wileman 2005; Guimer-Sorbets and Morizot 2010. E.g., Vérilhac 1978 and 1982; for a few examples, see Chaniotis 2012b, 107–112. Kotsifou 2012a and 2012b. Kotsifou 2012a, 76.

332

The Imprint of Emotions Surrounding the Death of Children in Antiquity

This expression of affection is the direct opposite of the emotional detachment manifested in the famous letter from Oxyrhynchus wherein a husband advises his pregnant wife to keep the child if it is a male and to expose it if it is female.6 Childhood is as much a biological stage in the human life-cycle as a social construct loaded with powerful cultural experiences and intense emotions. In many cases biological age is consistent with social age, and these two elements may be linked through careful consideration of how the various stages in children’s lives (when they crawl, walk, and connect with other children and adults) correspond to alterations in their cultural identity or enable them to form their social networks and relationships.7 Until the application of gender-based approaches to archaeology in the 1990s, children were considered ‘variables’ rather than ‘cultural actors’, hidden on the periphery of past societies. At the same time, the study of children in biological anthropology became more focused, providing information on aspects such as growth and development, health and disease, and mortality patterns. Grete Lillehammer was one of the first scholars who made children visible in the archaeological record, 8 and has proposed that through the study of burials, artefacts, and human skeletal remains as well as through ethnographic parallels, it is possible to reconstruct the role of children in the adult world better. Although the osteological study of children’s remains occurred independently and at a much earlier period than theoretical concepts of childhood in archaeology, both disciplines have now reached a level of sophistication that encourages the progress of bioarchaeology.9 It appears reasonable to ask exactly what type of information can be extracted through the bioarchaeological study of children’s remains and burial contexts. Ideals and practices surrounding the death of a child reflect the adult society which is responsible for creating the mortuary record through which we view children.10 The manner chosen for disposing of the remains of children who have died varies depending on age and growth, and reveals cultural and social perceptions in any given society. For example, new-borns are often treated differently from older children, or they may be regarded as not yet fully human and so ineligible for burial rites, while children under a certain age are often excluded from adult burial places. The question, of course, arises about the relationship between ritual and emotion in these cases. One may assume that the different ritual treatment of dead bodies is connected with differences in the emotions expected to be aroused, displayed, and controlled, as well as with different social and cultural mechanisms of coping with death. However, the relationship between ritual and emotions is not without tension.11 Not all who treated dead individuals in a certain manner felt the 6 7 8 9 10 11

Rowlandson 1998, no. 230; Kotsifou 2012a, 77. Sofaer-Derevenski 2000, 9–11; Lewis 2007, 1–13. Lillehammer 1989. Bioarchaeology studies the human biological component of the archaeological record through a multidisciplinary approach; see Larsen 1997; Buikstra 2006. Baxter 2005a, 94. Chaniotis 2010.

Chryssa Bourbou

333

same emotions; rituals did not always succeed in arousing the desired emotions; and different rituals do not necessarily imply differences in feelings. Today, studies of the biological remains of children are receiving much more attention, as they provide the most direct evidence of their existence in the past and allow the bioarchaeologist to visualise them in their cultural and natural environment. At present, reliable methods have been adopted for estimating the age at death of children from skeletal remains – that is, dental development, epiphyseal closure, and diaphyseal long bone length.12 We are still unable to make reliable sex determinations due to the absence of the secondary sexual features present on the adult skeleton (on the skull and pelvis), but DNA analysis of ancient skeletal material may contribute in the future.13 Bioarchaeologists still face a number of challenges in the study of children’s remains from archaeological contexts. These challenges include the applied methodology – that is, the lack of standardised terminology for the various age categories which complicates comparative studies, and the preservation and representation of the sample, influenced by the fragile nature of immature remains, burial conditions, excavation techniques, and funerary practices. Although the vast majority of pathological conditions observed on the adult skeleton can be also detected in children – dental diseases, metabolic and infectious conditions, traumatic incidents, etc. – it should be noted that chronic diseases take time to develop in the bones, and children usually died before any lesions appeared thus further obscuring the actual cause of death. Advances in bioarchaeological studies, such as the refinement of techniques for detecting age, research on the diagnostic criteria for various pathological conditions,14 and the publication of specialised textbooks in non-adult osteology,15 have contributed much to limiting these biases. Our understanding of how families and societies responded and felt towards children has been greatly enhanced through the combination of biological and cultural data. Anna Lagia, for example, has investigated the bioarchaeological evidence for non-adult individuals from the Kerameikos and Plateia Kotzia cemeteries in Athens, focusing on the Late Archaic through to the Early Imperial periods.16 By comparing temporal and regional trends in non-adult (0–14 years of age) representation with a model life table, she attempted to assess the degree to which the bioarchaeological patterns observed reflect culturally significant patterns, or whether they are the result of preservation and recovery. Similarly, in recent decades, human skeletal remains have been used for the identification of infanticide in the archaeological record. Although such a practice would leave no trace on the bones, studies of the age-at-death patterns – natural deaths tend to exhibit a rather more diffuse age distribution, while infanticide

12 13 14 15 16

Ubelaker 1989; Scheuer and Black 2000. Lewis 2007, 13. Ortner et al. 2001; Lewis 2004. Scheuer and Black 2000. Lagia 2007.

334

The Imprint of Emotions Surrounding the Death of Children in Antiquity

shows a strong peak at perinatal age –, has shed more light on cases where infanticide was suspected.17 In this chapter we explore evidence for the demonstration of emotions towards children in antiquity, which is visible archaeologically in burial practices. These mortuary remains (both material and biological) offer the best possibility of studying emotions towards children as projected in the archaeological context. 2 WHERE DID ALL THE CHILDREN GO? In the line of social transitions, death can occur a long way down or, in many unfortunate cases, shortly after birth. According to ancient texts, the burial of children followed specific regulations; Pliny the Elder (first century CE) for example, clearly stated that a child should be buried, as cremation could not be used before the development of the first teeth.18 Strictly outlined rites of passage for children existed in antiquity, providing the socio-cultural context for the systematic ritual introduction of the new-born to the family, important members and groups in society, and the gods. In Athens, for example, formal acceptance into the household (amphidromia) took place on either the fifth or seventh day after birth.19 Although we do not have detailed accounts of the ceremony, we can reconstruct its outline. The primary event was the father’s carrying of the new-born around the hearth, the household’s ritual centre. Around the same time, the birth of a child was announced publicly by decorating the door with wreaths and branches adorned with wool. Female children and those of lower-status families were sometimes named during the amphidromia, while wealthier families named their male offspring on the dekate (the tenth day after birth), accompanied by an animal sacrifice performed by the father in a public sanctuary. A child’s first public appearance may have occurred at the Anthesteria festival, when the child turned three. According to custom, the child’s presentation was marked by a gift of choes – small vases depicting children.20 Did such clearly distinguished rites of passage affect the fate of those who died before effective initiation? Can we identify other factors possibly affecting the interment of children? The following examples are specifically chosen to address social and cultural circumstances under which no formal burial service was performed, or cases in which children were buried in other, less distinguished places.

17 18 19 20

Smith and Kahila 1992; Mays 1993; Mays and Eyers 2011. Natural History 7.72. Parker 1983, 51, proposed that the amphidromia also served to purify the child from the pollution of birth. Garland 1985; Hamilton 1992; Golden 2003; Wise 2005.

Chryssa Bourbou

335

2.1 Babies in Wells From a modern point of view, the discovery of a well filled with children’s remains can be considered atrocious. However, the archaeological record includes a good number of adult and non-adult burials in wells, and the excavators, noting the ‘atypical’ nature of these deposits – that is, outside the formal cemetery area, or in its vicinity but clearly separated from it – suggest that they were used for the burial of slaves; social outcasts; victims of war, famine, natural disaster, or epidemics; or as sacrificial pits.21 In an effort to understand the motives and emotions behind this act and if it reflects possible social exclusion, my discussion will focus on the case of Messene, and its closest parallel at Athens.22 In both cases, the deposits include the skeletal remains of children, together with faunal remains (primarily of dogs) and numerous sherds. During the 2004–2005 digging campaign, a well deposit dating to the Hellenistic period (third–second century BCE) was brought to light in the Agora of Messene (Figure 1).23 The well is situated close to a public building identified as the old Bouleion of the city, a few metres north of the Doric temple of the goddess Messene.24 At a depth of three to four metres, a great number of commingled nonadult human remains and animal bones were recovered, as well as many sherds, mainly of pointed amphorae, of local manufacture, and cooking pots.25 A count of the human bones resulted in a minimum figure of 262 individuals, as represented by the left femur. Based on the measurements of the intact long bones, the human bones belonged to prematures, neonates, and stillborns. The macroscopic and radiological examination revealed no observable signs of long-term disease or trauma that could have contributed to their death.26 Nevertheless, pathological conditions associated with the risks of pregnancy and childbirth affecting the mother-infant pair could have contributed to the early death of the new-borns – that is, poor maternal health and nutritional status, inadequate care during pregnancy and delivery, premature birth, neonatal tetanus, etc. It is not possible to identify cases of miscarriage or abortion positively, but such possibilities cannot be excluded. The Athens Agora well (165–160 BCE) contained the remains of c. 450 infants and 130 dogs, buried in large pots, such as basins. The on-going study of the 21 22

23 24 25 26

For a thorough discussion, see Papadopoulos 2000. Another similar example is the well deposit (FK 153) at the Sebasteion in Eretria (Schmid 1997; Chenal-Velarde 2006), although canine bones (1100) outnumber those of the infants (nineteen individuals). The well also contained a hoard of silver and bronze coins dating to the first half of the third century BCE, possibly corresponding to the Chremonidean War between 267 and 261 BCE. The study by Chenal-Velarde focuses on the analysis of the dog remains and the significance of their presence in association with the infant skeletons. Themelis 2004, 42, fig. 5, pl. 16b; Themelis 2005, 54f. Messene was the deified queen of the land who, according to epigraphic evidence, gave her name to the new capital of Messenia in 369 BCE. Analysis of the abundant canine remains recovered from the well is still in progress. Bourbou and Themelis 2010.

336

The Imprint of Emotions Surrounding the Death of Children in Antiquity

collection has so far identified a number of premature babies; others with pathological conditions, mainly bacterial meningitis; and some born with disfigurations, such as a cleft palate.27

Figure 1. The well in the Agora of Messene.

Figure 2. Messene: Ground plan of the K3 Grave Monument and its precinct.

27

Rotroff et al. 1999; Rotroff and Liston forthcoming.

Chryssa Bourbou

337

It has been argued that the bodies of adolescents and children over the age of three were commonly placed in burial grounds along with adults, but infants under that age comparatively rarely.28 Infants in the ancient Greek world, and elsewhere, constituted a special category of the dead, and were given a different treatment than adults in both life and death. It thus seems possible that those who died before the age marking their formal recognition as a member of society were buried in less distinguished places. Disposal in a well may appear rather cruel, but before jumping to assumptions based on modern ideas about social exclusion, additional aspects should be considered. Messene has an unusually large number of intra muros grave monuments associated with public buildings. The grave monuments K1, K2, and K3, along the west stoa of the gymnasion dating to the late third century BCE, contained the inhumed members of elite Messenian families. The largest grave monument, K3, contained eight cist graves, symmetrically arranged around a central square theke, and was surrounded by a precinct wall.29 Inside this precinct, along its west wall, behind the chamber of K3, twenty-five child pot burials were brought to light, (Figure 2), accompanied by four burials of dogs (Figure 3).

Figure 3. Messene: Dog burial no. 29 from K3 Grave Monument.

28 29

Sourvinou-Inwood 1995. Themelis 1999, 97, fig. 7, pls. 63g, 66a; Themelis 2000a, 97–99, fig. 7, pls. 61b, 62; Themelis 2000b, 124–136, fig. 114. Recent discussions of this find: Schörner 2007, 245–247; Fröhlich 2008, 210–219.

338

The Imprint of Emotions Surrounding the Death of Children in Antiquity

It can be argued that, in this case, the Messenians buried their deceased new-borns formally, although in places other than the designated burial ground – as they were within the precinct of this grave monument, which somehow placed them under the protection of the heroised eminent Messenian family buried there. Thus, it also seems probable that the non-adults recovered from the well had been initially buried in pots and amphorae at a specific burial ground which at some point in time changed in use; so the pot burials were subsequently transported and redeposited into the well. The large number of non-adult remains might also suggest that the well continued to receive dead infants as a place known to mothers and local midwives for the disposal of infants who died at birth. The discovery at the Agora well of infants demonstrating congenital anomalies – that is, cleft palate – can be associated with specific notions and emotions towards a defective child. Cleft palate, a midline defect of the palate that permits open communication between the oral and nasal cavities, is the result of arrested development during embryogenesis.30 Infants born with cleft palate had a smaller chance of surviving in antiquity, since it was impossible for them to suckle well and breathe normally, making them vulnerable to respiratory infections.31 Although the death of many of the affected infants should be expected in ancient times, superstitious attitudes towards malformed individuals must have also taken their toll. The birth of a deformed child tended to be considered as the unfortunate result of divine punishment inflicted upon the sinful parents, who had to bear the burden of such a child – though occasionally this was tempered by compassion for the child as an innocent victim. Practical considerations must also have weighed in upon the decision to reject a congenitally deformed new-born, since the rearing of such a child was economically demanding. This notion that it is only worth raising a perfect (anatomically and mentally) new-born is attested in Soranus’ treatise on Gynaecology (first century CE), where he clearly states that the child should be ‘perfect in all its parts, members, and senses’.32 Congenitally deformed children were occasionally raised – and served as a constant reminder of divine displeasure – but in most cases the pressure to expose such a child must have been substantial. As such, a violent fate awaited deformed new-borns, being the most probable victims of infanticide.33 The presence of canine remains in both wells can be explained by their association with death, healing, and purification rituals following childbirth.34 Dogs were essential to the comfort and dignity of the dead, escorting them on their trip 30 31 32 33 34

Aufderheide and Rodríguez-Martín 1998, 58; Ortner 2003, 456–459; Barnes 1994, 171–175. Roberts and Manchester 2005; Barnes 1994, 174. Soranus 2.6 [26]; see also Temkin 1991, 79f. Garland 1995, 13–16. Trantalidou 2006. By the Late Hellenistic period, pets were buried and many were even given their own gravestones and epigrammatic epitaphs (Preston Day 1984, 29). A well-known example of the first century CE is the burial of a dog in a carefully constructed cist grave found in the cemetery of Amalias and Panepistimiou Street at Athens, wearing a richly decorated leather collar and accompanied by two glass bottles (Parlama and Stampolides 2000, 157, T. 82, cat. 162–164).

Chryssa Bourbou

339

to the underworld, or as guards of Hades and symbols of death. Of special importance is the association of dogs with childbirth, since dogs were sacrificed after a successful childbirth to deities associated with fertility, pregnancy, and childbirth; that is, to Eileithyia and Artemis Orthia.35 Dogs are also reported as sacrificial animals for easing childbirth or in funeral/ritual contexts.36 As a medium between the living and the dead, and associated with deities assigned to the protection or easing of childbirth, it is also possible that dogs were thrown into the wells following purification rituals of birth and death, or even the killing of deformed newborns. 2.2 A special place for the dead? Ancient writers noted the symbolism inherent in pot burials (enchytrismoi), the pot taking the role of the uterus. 37 Such burials using domestic and trade pots for the interment of non-adults were found – depending on the age – at specifically reserved zones within the cemetery, or scattered among adult burials.38 The cases of the Kylindra infant cemetery at Astypalaia and the Chania infant and adult burial ground are discussed here because they raise a number of questions regarding the choice of who was buried in these burial grounds and the reasons why. The large number of pot burials excavated at Kylindra in Astypalaia, which by the end of September 2009 reached the remarkable number of 2,754 intact and fragmentary pots, places this cemetery among the very few known examples of large children’s cemeteries worldwide.39 The site was continuously used from the late Geometric to the Imperial period for the exclusive burial of non-adults in pots, few of them with accompanying goods. Traces of burial rituals are also occasionally reported. Study of the human skeletal remains has been undertaken since 2000 by a team from UCL under the supervision of Dr Simon Hillson.40 This up-to-date anthropological study has revealed that some of the pots held one individual, although pot burials with two (probably twins) or three individuals are also reported. The osseous remains represent mainly new-borns and foetuses or prematures, although older children (the oldest being three years old) have also been identified;41 no pathological conditions have been recorded. 35

36 37 38 39 40

41

In ancient Messene, Artemis Orthia was honoured by women as a protector of children in a small sanctuary with a prostyle temple to the northwest of the Asklepieion; see Themelis 1994. Preston Day 1984; Mylonas 1956; 1962, 480; Vermeule 1979, 46, 103; Scholz 1937, 37; Wapnish and Hesse 1993, 55–80; Gräslund 2004, 168–176. Dasen 2002; Dasen and Ducaté-Paarmann 2006. See Guimier-Sorbets and Morizot 2010. Michalaki-Kollia 2010a. In a recent lecture given by Maria Michalaki-Kollia (21 March 2012), the number was found to have increased to 2,770 pot burials. The Astypalaia collection offers an exceptional opportunity for a detailed study of growth based on dentition and the skeletons. Preliminary results show considerable variation in these babies’ state of development: see Hillson 2009. It is not clear yet why these older children were not buried in the necropolis of Katsalos where adult and child burials have been found.

340

The Imprint of Emotions Surrounding the Death of Children in Antiquity

This infant cemetery is situated on the margins of the ancient city, along the city-wall. It is also quite distant from Katsalos’ designated necropolis, where adults and older children are found. Having missed out the cultural rites of passage undergone by adults and older children, these infants were placed outside the confines of the communal burial ground. This impressive congregation of pot burials constitutes a confusing riddle. Piece by piece, information derived from different data sets is pointing to a number of hypotheses. Epigraphic evidence from the island attests to the cults of Artemis Lochia and Eileithyia,42 and, particularly during the Hellenistic era, of foreign deities such as Atagartis.43 If a connection exists between these inscriptions and the infant cemetery, it might be possible that this unique assembly of enchytrismoi formed a special burial place close to a sacred area or sanctuary dedicated to Artemis Lochia, where women came to bury their dead children and thus put them under her protection. However, numerous questions still puzzle researchers involved in the analysis of this unique find: were only babies who died at birth buried there? Is it possible that unwanted, aborted, exposed, or abandoned babies, or even the victims of infanticide, were included? Were all babies from Astypalaia, or can we suspect different origins? Was this a local custom or a custom brought by the colonists of the eighth century BCE? The thorough analysis of this exceptionally large sample is expected to add considerably to our understanding of the attitudes and emotions surrounding the death of new-borns and infants. During three field seasons (2005, 2009 and 2011), the rescue excavation at the National Stadium in the city of Chania on Crete has revealed an extended burial ground dating to the Classical–early Hellenistic period (fourth–third centuries BCE), south-east of the already known area used as the ancient cemetery of Kydonia.44 To the north, a road came to light, probably leading to the burial area or providing an exit from the city. Sixty pot-burials, nineteen primarily adult burials in ‘pits’ (some with multiple interments) and two tile graves have been recovered.45 The pot burials usually had no accompanying goods, although occasionally some were retrieved from inside or outside the pots. The majority of the ‘pit’ burials, on the other hand, had one or more accompanying grave goods (that is, clay vessels, clay figurines), and four silver charoneia (coins placed in the mouth of an individual as payment for the ferryman) have been also recovered. Preliminary data on the human skeletal material indicate that the pots contained the remains of premature and full-term babies, while the burial in pot A41 most proba-

42 43

44 45

The inscriptions date to the fourth–second centuries BCE, much later than the earliest pot burials, which date to the late Geometric period; Michalaki-Kollia 2005. Michalaki-Kollia 2005 and 2010b. It should be noted that during the 2010 excavation season a figurine of the Egyptian deity Bes, known as a protector of households and, in particular, mothers and children, was found inside the pot burial of an infant (~1.5 years old); see Michalaki-Kollia 2010a, 182. See Chatzidaki 1987; Niniou-Kindeli et al. 1988; Niniou-Kindeli 1990; Andreadaki-Vlazaki 1998, 858–860. For the finds from the 2005 and 2009 field seasons, see Kataki 2012.

Chryssa Bourbou

341

bly held the remains of twins.46 Twin burials in the bioarchaeological record are occasionally reported,47 though the method of their disposal varies, thus partially reflecting differences in cultural attitudes towards twins.48 Ancient medical writers and biologists viewed the birth of twins either as an infrequent yet natural phenomenon, or as a monstrosity. In mythology, twins often had a double paternity, human and divine,49 but in any case their birth was considered an exceptional event often associated with impurity, adultery, and illegitimacy, stimulating relevant emotions of despair, guilt, shame, fear for the consequences, etc. 50 In our case, however, the death of these twins followed the same ritual regulations as any of the other neonates who died before effective initiation: disposal was in a pot and at the same resting place, presumably outside the formal burial area.

Figure 4. National Stadium of Chania: Burial 36, evidence of trephination (detail).

Of special interest are Burial 25, which included 12 individuals, one with clear evidence of trephination (Figure 4), and Burial 36, that of an adult individual with iron fetters. Although the study of the finds (artefacts and human bones) is still in progress, the first question that has immediately arisen is why babies and the aforementioned adults were buried in the same area. If a special place of burial is suspected for these children, based on our previous discussion regarding the fulfilment of specific rites of passage, perhaps we could also suspect that these adult individuals buried in the same area once lived on the periphery of society. The 46

47 48 49 50

Estimation of age revealed that the remains belonged to preterm babies (