Minds on Stage: Greek Tragedy and Cognition 0192888935, 9780192888938

Table of contents :
Cover
Minds on Stage: Greek Tragedy and Cognition
Copyright
Preface
Contents
List of Contributors
1. Introduction
1.1 Cognitive literary studies
1.2 Greek tragedy and cognition
1.3 Questions of methodology
1.3.1 Cognitive ‘theory’ and cognitive ‘science’: ideas or claims to firm knowledge?
1.3.2 Modes of variation: universality or cultural specificity?
1.3.3 The ‘what’ and the ‘how’: does cognitive criticism generate new readings?
1.3.4 Two-way traffic: can cognitive criticism give back to cognitive science?
1.4 The chapters
Acknowledgements
PART I. READING MINDS
2. Mindreading, Character, and Realism: The Case of Medea
2.1 Introduction
2.2 Reading real minds and/or minds on stage
2.3 Realism and character: the case of Medea
2.4 Medea’s multifarious mind
Acknowledgements
3. Reading the Mind of Ajax
3.1 Introduction
3.2 The challenge of Ajax’ ‘deception speech’
3.3 Internal readers of Ajax’ mind
3.4 Human minds in a divinely charged universe
3.5 Shifting mental states and the permanence of change
3.6 Conclusion: reading minds in Athenian tragedy
Acknowledgements
4. Space for Deliberation: Image Schemas, Metaphorical Reasoning, and the Dilemma of Pelasgus
4.1 Introduction
4.2 Literary criticism and cognitive poetics
4.3 Two models of metaphor comprehension
4.4 Metaphor and the mind of Pelasgus
4.5 Embodied experience and dramatic character
PART II. COGNITIVE WORK BY CHARACTERS
5. Attribution and Antigone
5.1 Introduction
5.2 Malle’s attribution theory
5.3 Antigone: cause or reason?
5.4 Antigone’s and Ismene’s attributions
5.5 Antigone’s valuing
5.6 Conclusion
6. ‘Remember to What Sort of Man You Give this Favour’: Looking Back on Sophocles’ Ajax
6.1 Introduction
6.2 The workings of memory
6.3 Memories of the living Ajax
6.4 Debating the memory of Ajax
6.5 Odysseus and the possibility of impartial memory
6.6 Conclusion: looking at memory through Ajax
7. Thinking Through Things: Extended Cognition as a Consolatory Fiction in Greek Tragedy
7.1 Introduction
7.2 Partnering with knowledgeable objects in the Oresteia?
7.2.1 The house of Agamemnon
7.2.2 Electra thinking through things
7.3 Anthropomorphic thinking and embodied memory in Euripides’ Heracles
7.4 Make-believe in drama: extended cognition as an ‘as-if’ fiction
7.4.1 Anthropomorphism as a cognitive reflex
7.5 Conclusion
Acknowledgements
PART III. PERFORMANCE, SPECTATING, AND COGNITION
8. Spectating Ancient Dramas: The Athenian Audience and its Emotional Response
8.1 Introduction
8.2 Theory of conceptual blending
8.3 Emotions of ancient audiences
8.4 Dynamic model of spectating
8.4.1 Interaction, contextualization, and separation
8.4.2 Affect and immersion
8.5 Emotions and conceptual blending theory
8.6 Conclusion
Acknowledgements
9. Gorgias’ ἀπάτη, Sophocles’ Electra, and Cognitive Criticism
9.1 Introduction
9.2 Gorgias fr. B 23 DK and cognitive approaches to aesthetic experience
9.3 An enactive reading of the false messenger speech in Electra
9.4 Aesthetic illusion and deceit, an uncanny entanglement
Acknowledgements
10. Seeing Together: Joint Attention in Attic Tragedy
10.1 Introduction
10.2 Sight invitations as speech acts and catalysts for joint attention
10.3 Capacious invitations and conflated addressees in Libation Bearers
10.4 Joint attention and joint knowledge in Women of Trachis
10.5 Dramatizing the limits of joint attention in Bacchae
10.6 Conclusion
Acknowledgements
11. Generic Expectations and the Interpretation of Attic Tragedy: Some Preliminary Questions and Cognitive Considerations
11.1 Introduction
11.2 Generic expectations in Attic tragedy
11.3 Disappointment of generic expectations: some case studies
11.4 Conclusion
Acknowledgements
12. Situated Cognition: Sophocles, Milgram, and the Disobedient Hero
12.1 Introduction
12.2 Neoptolemus in Sophocles’ Philoctetes: a hero under pressure
12.3 Milgram’s experiments in obedience to authority
12.4 Herbert Clark and the natural tendency to cooperate
12.5 Cooperation in the Philoctetes
12.6 The disobedient hero
12.7 Conclusion
Acknowledgements
Bibliography
Index Locorum
General Index

Citation preview

C O G N I T I V E C L A S SIC S General Editors Felix Budelmann  Ineke Sluiter

C O G N I T I V E C L A S SIC S Cognition is fundamental to knowledge, being, and action. For classicists, a focus on cognition in all its forms re-orientates the study of texts, beliefs, values, artefacts, environments, and practices; in turn, Greco-Roman antiquity offers material that variously supports, extends, and challenges current thinking about the mind. Cognitive Classics promotes interaction between classical scholarship and the range of cognitive studies, from psychology and neuroscience to linguistics and philosophy.

Minds on Stage Greek Tragedy and Cognition Edited by

F E L I X BU D E L M A N N A N D I N E K E SLU I T E R with the assistance of Bob Corthals

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Oxford University Press 2023 The moral rights of the authors have been asserted All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2023933382 ISBN 978–0–19–288893–8 DOI: 10.1093/oso/9780192888938.001.0001 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Preface It is a pleasure to record the numerous debts of gratitude we incurred in editing this book. Charlotte Loveridge at Oxford University Press provided astute advice as well as encouragement throughout the process. The two readers for the Press were exceptionally helpful and detailed in their constructive criticism; their criticism and suggestions made the book considerably stronger. Louise Larchbourne was a meticulous copyeditor, and Vasuki Ravichandran oversaw the production process with great efficiency. A number of the chapters in this volume started life at a conference on Minds on Stage held 14–16 April, 2016 at Leiden University. The conference was funded by the NWO Spinoza Prize 2010. We are grateful to Leanne Jansen for her or­gan­ iza­tion­al work on that conference and to Alisha Meeder and Anne van Schaik for assisting. We (and our authors) were lucky at an early stage of the editorial process to receive expert input from the following friends and colleagues, listed in al­pha­bet­ ic­al order: Bob Corthals, Max van Duijn, Katherine Earnshaw, Naomi Ellemers, Henkjan Honing, Elizabeth Minchin, Bridget Murnaghan, Ruth Scodel, Evert van Emde Boas. Thanks are also due to the (then) student assistants Henric Jansen and Olivia Monster, as well as the (then) coordinator of the Anchoring Innovation research team, Aniek van den Eersten, who helped us in preparing the manuscript for the press. Last but in no way least, Bob Corthals not only prepared the indexes, but also removed numerous errors and inconsistencies at a late stage.

Contents List of Contributors

ix

1. Introduction Felix Budelmann

1

PA RT  I .   R E A D I N G M I N D S 2. Mindreading, Character, and Realism: The Case of Medea Evert van Emde Boas

25

3. Reading the Mind of Ajax Sheila Murnaghan

43

4. Space for Deliberation: Image Schemas, Metaphorical Reasoning, and the Dilemma of Pelasgus Michael Carroll

60

PA RT I I .   C O G N I T I V E WO R K B Y C HA R AC T E R S 5. Attribution and Antigone Ruth Scodel 6. ‘Remember to What Sort of Man You Give this Favour’: Looking Back on Sophocles’ Ajax Lucy Van Essen-­Fishman 7. Thinking Through Things: Extended Cognition as a Consolatory Fiction in Greek Tragedy Anne-­Sophie Noel

81

98

117

PA RT I I I .   P E R F O R M A N C E , SP E C TAT I N G , A N D  C O G N I T IO N 8. Spectating Ancient Dramas: The Athenian Audience and its Emotional Response Hanna Gołąb

135

9. Gorgias’ ἀπάτη, Sophocles’ Electra, and Cognitive Criticism Jonas Grethlein

153

10. Seeing Together: Joint Attention in Attic Tragedy A. C. Duncan

173

viii Contents

11. Generic Expectations and the Interpretation of Attic Tragedy: Some Preliminary Questions and Cognitive Considerations Seth L. Schein

196

12. Situated Cognition: Sophocles, Milgram, and the Disobedient Hero Bob Corthals and Ineke Sluiter

210

Bibliography Index Locorum General Index

229 257 263

List of Contributors Felix Budelmann  is Professor of Classics at the University of Groningen. He works on Greek literature, with a particular interest in cognition. Current projects include a monograph on the ‘now’ in Greek literature and a co-­edited volume on visualization in Greek and Latin poetry. Michael Carroll is Lecturer in Greek Literature at the University of St Andrews. He is currently completing a monograph on metaphor in Aeschylus that draws on cognitive linguistics. His next research project will be an investigation of the poetic persona in Pindar’s epinicians (again with the help of cognitive linguistics). Bob Corthals  obtained degrees in Classics from Leiden University (MA) and the University of Oxford (MPhil). His research focuses on cognitive approaches to Homeric poetry and tragedy. Another of his interests concerns the development of the Ajax myth in early Greek literature. A. C. Duncan is Assistant Professor of Classics at the University of North Carolina, Chapel Hill. His research and teaching consider the contexts, values, and reception of ancient Greek drama from diverse perspectives. Current and future research projects address materials’ agency in the development of Athenian drama and the modern reception of tragedy in divided societies. Hanna Gołąb is American Council of Learned Societies Barrington Centennial fellow and National Endowment for the Humanities fellow at the American School of Classical Studies at Athens. She has published articles on Greek choral lyric, drama, and epigraphic poetry, but her research interests also encompass ancient conceptualization of space and various practices of place-­making. Jonas Grethlein is Professor of Greek at Heidelberg University. His monographs include The Greeks and their Past (Cambridge University Press 2010), Experience and Teleology in Ancient Historiography (Cambridge University Press 2013), Aesthetic Experiences and Classical Antiquity (Cambridge University Press 2017), The Ancient Aesthetics of Deception (Cambridge University Press 2021), and Mein Jahr mit Achill (Beck 2022). Sheila Murnaghan  is the Allen Memorial Professor of Greek at the University of Pennsylvania. She works in the areas of ancient Greek epic and tragedy, gender in classical culture, and classical reception. Her current projects include Norton Critical Editions of Sophocles’ Antigone and Oedipus at Colonus and a commentary on Sophocles’ Ajax. Anne-­Sophie Noel  is Associate Professor of Greek at the École Normale Supérieure de Lyon and a member of the research team HiSoMA. A philologist and cultural historian of ancient Greece, she devotes her research to the understanding of ancient Greek plays as

x  List of Contributors multisensory performances. Her current research combines affective and cognitive approaches to objects in Greek drama as well as ancient theatrical spectatorship. Seth L. Schein is Professor Emeritus of Comparative Literature, University of California, Davis. He works mainly on Homeric epic, Attic tragedy, and classical receptions. His books include The Iambic Trimeter in Aeschylus and Sophocles: a Study in Metrical Form, The Mortal Hero: an Introduction to Homer’s Iliad, Homeric Epic and its Reception: Interpretive Essays, and commentaries on Sophocles’ Philoctetes and Homer’s Iliad: Book 1. Ruth Scodel,  educated at UC Berkeley and Harvard, retired as D.  R.  Shackleton Bailey Collegiate Professor of Greek and Latin at the University of Michigan in 2019. She is the author of books and articles on Greek literature, particularly on Homer and tragedy, with theoretical approaches derived from narratology, the sociology of Ernest Goffman, Theory of Mind, and attribution theory. Ineke Sluiter is Professor of Greek at Leiden University. She is the PI of a research programme on ‘Anchoring Innovation’. Her research interests further include ancient ideas on language, value discourse in public debates in antiquity, and cognitive approaches to the study of the ancient world and its literatures. Evert van Emde Boas is Associate Professor in Classical Philology at Aarhus University, Denmark. His research focuses on the application of linguistic and cognitive approaches to Greek literature. He is the lead author of the Cambridge Grammar of Classical Greek (Cambridge University Press 2019), author of Language and Character in Euripides’ Electra (Oxford University Press 2017), and co-­ editor of Characterization in Ancient Greek Literature (Brill 2018). Lucy Van Essen-­Fishman  completed a DPhil at Brasenose College, Oxford on the construction of character in the tragedies of Sophocles and spent several years lecturing in Classics at various Oxford colleges. She is now a Lead Policy & Research Analyst at the Higher Education Statistics Agency.

1

Introduction Felix Budelmann

1.1  Cognitive literary studies Along with the cognitive humanities in general, cognitive literary studies are coming of age. What was a number of dispersed activities during the 1980s and ’90s has become a thriving field, with an ever-­increasing flow of publications, frequent conferences, as well as networks and research projects. In his 2004 ‘field map’, Alan Richardson had provisionally divided the terrain into cognitive rhet­oric and conceptual blending theory, cognitive poetics, evolutionary literary theory, cognitive narratology, cognitive aesthetics of reception, and cognitive materialism and historicism.1 A similar map in 2023 would have to accord a substantial portion of space to approaches that try to get away from the notion that the mind operates in isolation, as a kind of computer in the skull. The slogan of the ‘4Es’, which has in recent years assumed considerable currency across the cognitive humanities, emphasizes that the mind is embodied (with ‘Cartesian’ mind-­ body dualism as the go-­to villain), embedded (in its various contexts), extended (to prosthetic devices such as the memory encoded in one’s mobile phone or shopping list), and enactive (viz. constituted by its interaction with the environment). By labelling these approaches—­both the relevant scientific approaches to cognition and in turn the approaches to the cognitive study of literature and other branches of the humanities that draw on them—‘second-­generation’, the field has self-­consciously acquired a history.2 Like feminist criticism, poststructuralist criticism, or New Materialism, cognitive literary studies, and indeed the cognitive humanities as a whole, are a set of loosely connected enterprises rather than a focused programme of research. What holds these enterprises together, and justifies the umbrella term, is a shared interest in cognition, and in dialogue with research into cognition in other 1  Richardson 2004. 2  The secondary literature on 4E cognition and the study of literature, history, and culture is already considerable. Two special journal issues provide useful points of entry from a literary perspective: Kukkonen and Caracciolo 2014a and Morgan et al. 2017. For a short account of the history of the field until 2014, see Caracciolo 2014, 16–23. The most significant publication within Classics is the an­tiquity volume of the Edinburgh History of Distributed Cognition (Anderson et al. 2019), which includes introductions to the study of distributed cognition in the humanities in general and Classics in particular.

Felix Budelmann, Introduction In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0001

2  Felix Budelmann subjects. It is important to recognize how loose this definition is. Cognition itself is understood very broadly, and has, with the dismissal of computational models of the mind, ceased to stand in opposition to affect: emotions, intuitions, and contextual embedding are all within the remit of cognitive studies, so long as there is a focus on the way agents acquire knowledge of their environment and negotiate their place within it. The particular ‘cognitive’ subjects, moreover, with which cognitive literary studies enter into dialogue, form themselves a wide range that includes philosophy and linguistics as much as neuroscience and cognitive psychology, and the ancestry of the cognitive humanities includes not just fields of study in the sciences but also reader-­response criticism in literature and phenomenology in philosophy. For its part, psychology, which has a prominent presence in the cognitive humanities, more prominent certainly than neuroscience,3 opens the door to the social sciences because the dividing line between cognitive and social psychology is often blurry. With this multiplicity of associated dis­cip­ lines comes a variety of modalities. The engine of much of the work is, as it should be, what scholars of literature are good at—­reading texts and exploring issues of methodology—­but cognitive literary studies also draw on, and occasionally produce, empirical findings, ranging from psychological questionnaires to fMRI scans. The texts studied are both literary and theoretical, and the work produced variously puts forward readings of specific texts or makes systematic claims about literature and literary reading. Some cognitive critics build on poststructuralism while others position themselves in opposition, some home in on specific texts and their contexts while others foreground transhistorical developments. Even in its second generation, this is still very much a field in flux. A number of thoughtful critiques of the aspirations and achievements of cognitive literary studies are not just testimony to a sense of incipient establishment, but also an indication that a great deal remains to be worked out.4 It will be a theme of this Introduction that much of the best work in cognitive literary studies has an air of adventure and experimentation. Classics has been catching up fast with the cognitive humanities—­the publication of three wide-­ranging collected volumes in 2018 and 2019 representing a significant milestone—­but it is too early still to speak with confidence about the particular shape cognitive literary studies, or indeed cognitive studies in general, will take in Classics.5 One tentative suggestion may perhaps be made neverthe3  The limited role of neuroscience in cognitive literary studies is bemoaned by Armstrong 2013. His book is a good starting place for thinking about what is and isn’t possible. More has been done with performance than with matters of interpretation; see e.g. Falletti et al. 2016. 4  For early critiques see Adler and Gross 2002 and Jackson 2003. A more recent account of the challenges faced by the field is Bruhn 2011, and with a focus on spectatorship see McGavin and Walker 2016, ch. 2. Alber et al. 2018 present critical dialogues between cognitive and ‘unnatural’ narratologists. The fullest critical assessment in Classics to date is Sharrock 2018. 5  Three volumes: Lauwers et al. 2018, Anderson et al. 2019, Meineck et al. 2019b. For a bibli­og­ raphy of relevant work see https://cognitiveclassics.blogs.sas.ac.uk/, which permits some tentative

Introduction  3 less: arguably, cognitive frameworks feel less radical and less of a deviation from normal scholarly practices in Classics than they do in some other humanities disciplines, for two reasons. One is that many of them have precursors in an­tiquity.6 Aristotle, with his interest in mimesis and in the formal means that tragedy and epic use to create certain psychological effects, might be regarded as a forerunner of cognitive criticism,7 and the point is often made that much ancient psychology is less dualist, and in that sense more in line with current thinking, than the ‘Cartesian’ model that dominated modernity for a long time.8 Neither the notion that literature is usefully thought about in terms of mental processing nor the idea that the mind is embodied marks a major change for classicists. The second reason is closely related. Compared to English or Modern Languages, Classics rarely challenges the prevalence of the implicit model according to which the criterion for assessing a reading is whether it is one that (typically ancient, sometimes modern) readers and audiences would have accepted. Radically impersonal and decentred intertextuality à la Kristeva, for example, never properly took hold, especially among Hellenists, and the relation between texts is normally thought of, though not necessarily articulated, as a connection in the mind of the reader (variously conceived) or sometimes the author.9 Only rarely does intention altogether leave the picture, even though it usually remains unnamed. In some ways, therefore, cognitive criticism is simply an extension of what classicists like to do with literature anyway. This creates the need to guard against serving old wine in new bottles, but it is also, and above all, an opportunity for sharpening, testing, and extending long-­held critical instincts and adding conceptual robustness.

1.2  Greek tragedy and cognition The majority of collected volumes and special journal issues in cognitive literary studies define their remit in terms of a specific approach or theme, which they explore across a mixed group of texts. Recent examples include: cognitive literary studies and the ethical and pedagogical function of literature, dialogues between observations, e.g. the prominence of work on metaphor, and the co-­presence of first- and second-­ generation work (on which see further pp. 6–7 below). See also the accounts in Cairns 2019 and Meineck et al. 2019a. Some areas within Classics, notably philosophy and religion, and perhaps also archaeology, have distinctive, and more established, profiles in their modes of engagement with cognitive science. 6  Cf. the related remarks, and references, in Cairns 2019, 18–20. 7  Note for example the prominence of Aristotle in Lowe 2000. 8  See especially Gill 2019, 156–62, comparing embodiment in ancient Stoicism and modern enactivism, and Ostenfeld 2018, who makes the case for the distinctiveness of Aristotle and Plato when viewed in the context of the modern mind–­body debate. 9  These are of course tendencies rather than universal norms; for key references see Baraz and Van den Berg 2013, 2–3.

4  Felix Budelmann unnatural and cognitive narratology, situated cognition and culture, continuities and breaks between cognitive, aesthetic, and textualist approaches to literature, theatre and cognitive neuroscience, second-­generation cognitive approaches to literature, and the intersection of cognition, literature, and history.10 In leaning the other way—­specific corpus but varied themes and approaches—­this volume is not altogether alone, but the decision requires explanation.11 In part, we are responding to the state of the field within Classics: we hope that Minds on Stage, as well as contributing to the study of Greek tragedy, may serve as an introduction to cognitive literary studies for classicists by showcasing, in the context of a specific and well-­known corpus, a wide range of approaches (and it is for the same reason that a sizeable portion of this Introduction is devoted to general methodological issues).12 What is sacrificed by way of focused investigation of a particular aspect of cognition is gained, we suggest, by enabling a view of tragic cognition in the round and providing access to what is still an emerging field. For more than one reason, Greek tragedy is an obvious choice for such a volume. Drama and performance (above all Shakespearean) have long been a particularly dynamic area in cognitive literary studies. Early pioneering work by individual scholars, notably Ellen Spolsky and Mary Thomas Crane, eventually led to the first multi-­author volume in 2006 and the first general treatment of cognition and spectating soon after.13 We now have two dedicated series, Bloomsbury Methuen’s Performance and Science and Palgrave Macmillan’s Cognitive Studies in Literature and Performance, which range across performance, language, character, embodiment, intertextuality, consciousness, emotion, memory, kinesic intelligence, and more, and at the time of writing stand at nine and twenty-­one titles respectively, as well as a Routledge Companion to Theatre, Performance, and Cognitive Science.14 More important, some of the most creative and methodologically astute thinking in cognitive literary studies has emerged and is still emerging from work on drama and performance. Apart from Spolsky’s continuing contributions, the work of Raphael Lyne, on rhetoric, metaphor, and intertextuality in Shakespeare and beyond, and Evelyn Tribble, on memory and

10  Bruhn and Wehrs 2014, Kukkonen and Caracciolo 2014a, Falletti et al. 2016, Morgan et al. 2017, Stopel 2017, Alber et al. 2018, Easterlin 2019. 11  Edited volumes with similar kinds of scope include Shakespeare and Consciousness (Budra and Werier 2016), Cognitive Joyce (Belluc and Bénéjam 2018), and Cognition, Mindreading, and Shakespeare’s Characters (Helms 2019). 12  In that respect, it resembles Meineck et al. 2019b, but with a literary, and indeed corpus-­specific, rather than general Classics remit. 13  Spolsky 1993, Crane 2001, Spolsky 2001; Spolsky’s work in particular is wide-­ranging, and goes far beyond drama. Multi-­author volume: McConachie and Hart 2006. General treatment: McConachie 2008. For a slightly fuller account of the history of cognitive approaches to drama and performance up till 2016, see Blair and Cook 2016, 11–13. 14  Kemp and McConachie 2018.

Introduction  5 attention in early modern theatre-­ making, may be singled out, but this is al­together a remarkably vibrant field.15 For its part, Greek tragedy, as a much-­read corpus, well-­furnished with editions and commentaries and not in need of spadework, has long served classicists as a testing ground for new methods. All major theoretical developments of recent decades have spurred work on Aeschylus, Sophocles, and Euripides, in many cases earlier and more extensively so than for other Greek authors, and indeed a body of cognitive criticism has been steadily building up for several years now. With hindsight, the first important contribution was probably N. J. Lowe’s cognitively inflected treatment of the classical plot, with a chapter on tragedy.16 Since then, Ruth Scodel, Felix Budelmann, and Pat Easterling, as well as Ineke Sluiter and colleagues, have drawn on Theory of Mind to discuss how characters are interpreted by the audience and interpret one another, Colleen Chaston has explored the cognitive function of props, and Douglas Cairns’ wide-­ranging work on metaphor and the emotions has covered tragedy alongside other ­genres.17 Peter Meineck surveys a range of cognitive approaches in his book-­ length account of the affective powers of Greek drama in performance, and a variety of approaches is on display also in the tragedy-­related chapters in the three recent collected volumes, discussing, respectively, synaesthesia, metaphor, madness, emotion, the mask, collective cognition, and attention.18 The cutting edge in cognitive criticism continues to be located outside Classics, but classicists, and not least so scholars of Greek tragedy, have been working to close the gap. It is unsurprising that Greek tragedy is proving fertile ground for this kind of work. Minds play a large role in the genre, both those of the characters and those of the spectators. The characters’ cognitive achievements and more often limitations are almost constantly in the spotlight, emblematically so in recognition and deception scenes and in the interpretation of oracles. For their part, the spectators know more than the characters but can themselves be subtly misdirected, have to come to terms with events that are deeply disturbing ideologically and existentially, and, most fundamentally, have to ‘suspend disbelief ’ to turn a man with a mask into Helen of Troy. Both dimensions are well-­represented in this volume, and often it is precisely the combination—­character cognition as an object of audience cognition—­that provides the critical fulcrum. Between them, the chapters presented here discuss many of the major topics in the study of Greek tragedy, grouped into three parts. Part I tackles the notions of

15  Lyne 2011, Lyne 2016, Tribble 2011. 16  Lowe 2000, ch. 8. 17  Scodel 2009, Budelmann and Easterling 2010, Chaston 2010, Sluiter et al. 2013, Cairns 2016 (which synthesizes several strands of Cairns’ work), Cairns 2020. 18  Angelopoulou 2018, Dobson 2018, Meineck 2018a, Meineck 2018b, Budelmann 2019, Meineck 2019, Noel 2019b, and (in a fourth volume) Budelmann and Van Emde Boas 2020. Note also the primarily linguistic pieces, some of them cognitively inflected, in Martin et al. 2020, and cognitive approaches make several appearances in De Temmerman and Van Emde Boas 2018a.

6  Felix Budelmann character and characterization, exploring the ways in which spectators make sense of characters, and characters make sense (or fail to make sense) of one another. The section opens with a theoretical discussion (2), followed by two essays focused on individual plays (3, 4). Part II homes in on specific cognitive modes of relating to one’s surroundings, and discusses causation and motivation (5), memory (6), and the relationships characters form with inanimate objects (7). Touched on intermittently throughout the volume, the cognitive dimensions of performance and spectating become the focus of study in Part III. A pair of chapters on make-­believe and engagement with theatrical fictions (8, 9) is followed by an essay on communal spectatorship (10), and the final two contributions discuss the steering of audience expectation in terms of, respectively, genre (11) and conceptual frameworks (12). (See below, pp. 18–21, for summaries of each chapter). An alternative mode of categorizing the contributions, and articulating their variety, is by cognitive approach taken or discussed. Several engage aspects of cognitive psychology—­Theory of Mind (chs 2, 3), mental imagery (9), joint attention (10), situated cognition (12), predictive processing (11)—whereas others look to forms of social psychology: attribution (5), memory (6), anthropomorphism (7). A further group draws on more conceptual work that comes out of philosophy and other humanities thinking: prototype theory (11), image schemata and blending (4, 8), thing studies and extended cognition (7), as well as enactivism (9).19 This (inevitably schematic) list of approaches prompts two observations. First, it is obvious that the definition of ‘cognitive’ adopted in this volume is capacious even by the traditionally capacious standards of the term. Social psychology is cognitive only in a loose sense, and the same is true for thing studies, which often sit under the umbrella of posthumanism. In part, our motivation has been breadth of coverage, but a further consideration is equally important. The prom­ in­ence that Greek tragedy accords to the minds of characters is to a large degree generated through social interaction: it is above all when the characters observe, deceive, accuse, pity, and resist one another that their cognitive efforts, achievements, and failures command our attention. The world of Greek tragedy is an essentially social one, and the appeal to the knowledge and methods of the social sciences is a natural response to this sociality. The second observation points in a similar direction. A number of chapters (esp. 7, 9, and 12) employ one or another of the 4E frameworks that have been so influential in the cognitive humanities recently (see above), but several others revisit concepts, such as Theory of Mind and image schemata, that were well-­ established already at the time of Richardson’s 2004 field map mentioned earlier. Here, too, the aspiration of the volume to provide a broad purview of different cognitive tools is a factor, but again there is a further consideration. The 19  What is not represented is (a) empirical work on responses to classical texts, such as Budelmann et al. 2016 and Van Emde Boas (forthcoming), and, except in passing, (b) neuroscience (cf. nn. 3 and 22).

Introduction  7 ‘second-­generation’ approaches constitute a response to a narrowly mind-­centred view of cognition, which pays scarce attention to bodies, physical environments, and other people. It is perhaps in part because the more extreme versions of such a view have never achieved real dominance in Classics, as indeed they would be alien to most ancient thought, that classicists can still find value in earlier frameworks at the same time as engaging with the 4E agenda. For classicists, embodiment and contextualization are a given more often than they need to be defended against internalism.

1.3  Questions of methodology The continued fluidity of cognitive literary studies is such that methodological questions are never far below the surface, in this volume as elsewhere, and this indeed is one of the attractions of the field. In the remainder of this Introduction, I will discuss four of the central issues. While the primary focus will continue to be on cognitive literary studies, several aspects of the discussion apply to the cognitive humanities more widely. The tone will be exploratory, as I believe circumstance demands.

1.3.1  Cognitive ‘theory’ and cognitive ‘science’: ideas or claims to firm knowledge? The epistemological status of cognitive criticism requires little elaboration where the body of work that is being engaged is speculative or theoretical—­such as for example Fauconnier and Turner’s ‘conceptual integration theory’, which makes several appearances in this volume. Here it is self-­evident that the truth claims of the resulting argument will be no different from those in other articles published in literary studies. Some contributors in fact flag the non-­empirical nature of the material on which they draw by using the label ‘cognitive theory’ rather than ‘cognitive science(s)’ (and I am using ‘cognitive studies’ as a neutral term in this Introduction). Such kind of cognitive literary work shares an interest with the scientific study of the mind—­cognition—­but does not share its epistemology, and like all literary criticism should be judged for its coherence, rigour, and interest. The issue becomes more complicated where evidence-­ based findings are involved, as is the case in empirical studies of readers’ responses to literary texts and, above all, in literary discussions that draw on empirical work in psychology and neuroscience.20 Undeniably, part of the appeal of the cognitive humanities 20  For the former, there exists a dedicated journal, Scientific Study of Literature. See also n. 19 above, and beyond literature the field of ‘neuroaesthetics’.

8  Felix Budelmann derives from the engagement with empirically obtained observations and the attendant truth claims. For some, in fact, this appeal is bolstered by the hope that the empirical grounding of science can be mobilized to overcome the perceived excesses of poststructuralism, but one does not need to be motivated negatively in this way to be intrigued, even excited, by the sense of solidity gained from dealing with scientific evidence.21 Yet any such excitement needs to be tempered by caution. When literary ­scholars draw on a ‘hard’ finding, the claims that they are themselves able to make will for two reasons often be rather soft. First, the data need to be interpreted. A finding itself may be beyond doubt (typically because it recurs when the ex­peri­ ment is repeated) yet its interpretation may not. The debate over ‘mirror neurons’ is a good illustration: while it is a fact that some of the same neurons are activated both when we observe and when we carry out a particular action, the extent and the interpretation of this phenomenon, known as ‘motor resonance’, are much less certain. Mirror neurons are fascinating to anybody interested in theatre because they might seem to explain why seeing a play (or watching a ballet, or perhaps even reading a book) can be such a vivid experience, but they only go so far. The connection between neural activity and conscious experience is elusive at best, and at the non-­neural, phenomenological level, there is certainly a substantive difference between the experience of doing something oneself and seeing or imagining somebody else do it. Mirror neurons are useful as an emblem of the intricate ties between motion enacted and motion perceived, and suggestive as traces of the pre-­conscious dimension of action, but they do not give us a shortcut for capturing, let alone explaining, the experience of theatre-­going. For scientists, mirror neurons are a subject of fast-­moving research programmes and lively debate. For scholars in the humanities, they are best taken not as a ready ex­plan­ ation but as an impetus to exploration, and mutatis mutandis the same is true for many other empirical observations about cognition.22 A second reason for tempering one’s excitement about cognitive truths is that the gap between the ‘hard’ findings on offer and many of the issues that concern literary scholars when they think about Greek tragedy is huge. Even though the scientific understanding of cognition has grown at an immense pace in recent

21  The idea that cognitive science provides ammunition against poststructuralism was important mostly in the earlier stages, and has receded as the field has developed and the influence of poststructuralism waned across the humanities. See for example Richardson and Steen 2002, 1–2, introducing an early special issue on cognitive literary studies, and the rejoinder by Adler and Gross 2002, 202. Notably, however, already at that stage Spolsky 2002 was looking for interactions between cognitive science and poststructuralism, an aspiration she has restated since. 22  Mirror neurons are briefly discussed in this volume by Grethlein, pp. 158–9. For a balanced review of the evidence with a view to spectatorship, see Garner 2018, 145–84; see also the index of Falletti et al. 2016, s.v. ‘mirror neurons’. Banks and Chesters 2018 is an excellent collection of essays exploring Renaissance literature, including drama, through the lens of kinesic intelligence. Uithol et al. 2011 usefully discuss the different uses of the term ‘resonance’ in the literature.

Introduction  9 years, science still has a long way to go before it will be able to answer many of the complex questions that might interest the literary critic. What exactly happens in our brain when we hear an allusion? What difference does alliteration make to the way a line of verse is processed? What is the neuroscientific basis of make-­ believe? We do not know. Considerable creativity as well as restraint are therefore required for making meaningful as well as defensible connections between the scientific article one reads and the tragic text. Does all this mean that, for better or worse, in the final reckoning cognitive studies always offer ‘just’ another set of ideas, even if those ideas are derived from scientific knowledge? I believe that different views can legitimately be taken on this issue. One way of testing one’s sensitivities is by asking whether a discussion of a literary text is invalidated if an apparently secure empirical finding on which the analysis rests is subsequently revised—­by no means a far-­fetched scenario in view of the speed at which cognitive science is moving, and indeed topical in view of the ‘replication crisis’ in psychology.23 On the one hand, it would be perverse to claim that there will be no difference. The cognitive sciences offer literary scholars knowledge about the functioning of the mind that may be used to produce better grounded and more finely nuanced thinking about the minds of authors, readers, characters, or performers. If such knowledge is superseded by subsequent research, the use we make of such knowledge in our thinking about literature must inevitably be affected (and nothing wrong with that, so long as the provisionality of many scientific findings, and new findings especially, is always kept in mind—­such is the development of knowledge in the sciences as indeed in the humanities). Despite the importance of interpretation that I have emphasized, there surely is a difference between a reading of a tragedy, or a theory about the workings of theatre, that appeals to scientific fact and one that appeals to (for example) aesthetic judgement—­both have their place but they are not the same—­ and it therefore cannot be without consequence if apparent fact stops being fact. On the other hand, much depends on how a particular scientific finding is being used for literary purposes. Literary critics have been employing psychoanalysis productively long after mainstream psychologists had turned their backs on it and Freud had become a no-­no. Many examples could be cited of exciting work by classicists that draws inspiration from a body of thought without committing to the factual truth of that body of thought. There is no reason why a version of this cannot be the case with cognitive science: if contemplation of scientific research allows us to see more in a text, or if scientific research stimulates us to devise new approaches to texts, what we have seen or what approaches we have devised will not disappear if the scientific research is challenged.

23  The replication crisis was sparked when a study that attempted to replicate a number of famous experiments failed in half of the cases. For an account, see Shrout and Rodgers 2018.

10  Felix Budelmann Arguably, then, the most important conclusion is that cognitive criticism needs to strive for maximal methodological clarity—­about the nature of the cognitive material used, the nature of the connections made, and the nature of the conclusions drawn.24 ‘Ideas or factual truths?’ is a good question to ask exactly because the answer, at least in the current state of the field, needs to be worked out afresh for each cognitive-­literary engagement.

1.3.2  Modes of variation: universality or cultural specificity? It is a truism that the study of cognition, whether in philosophy, psychology, neuro­sci­ence, history, or literature, will always involve the balancing of universality and (cultural, individual, etc.) specificity. The particular balance depends on the aspect of cognition that is being studied—­ethical values vary more substantially than fight-­or-­flight reflexes—­but as already the Greeks knew, a balance there always is.25 In fact, the idea of ‘balance’, and indeed the binary casting of the question in the section header, are probably simplistic, and it may be better to think of the relationship in terms of a system: for the cognitive critic, as for the cognitive scientist, the issue is not just how to weigh unity against diversity but also how to tackle what is a complex and dynamic system. A person’s physiology (which itself combines universals and specifics), the family environment, the multiple cultural contexts, and indeed the circumstances of the day all interact, with continual and mutual feedback, and interact differently on different occasions.26 All that said, it is undeniable that cognitive scientists, while properly conscious of variation and alive to the pitfalls of generalization, typically pursue generality over specificity. This is a question not of intellectual conceptualization (how should cognition be conceived?) but of priorities (which dimension of cognition do I shine my torch on?): different disciplines set out different stalls. Where then does that leave the humanities? In general, the aim must surely be to experiment with humanities-­specific versions of the dynamic system of the particular and the universal, and to regard the universalizing perspectives not as a threat but as a way of sharpening the grasp on specificity. Being able to articulate what stays the same, or at least recognizably similar, not only provides one of the ingredients necessary for thinking about why we can relate to works that are almost 2,500 years old but also allows us to identify with more precision the nature and the

24  For an astute and interesting perspective on the issues in this section see Spolsky 2015, xxix–­xxxi. 25  The varying balance is described for a number of case studies in Lloyd 2007. For classical Greece, the exploration of ethics in the Republic and the Nicomachean Ethics are outstanding examples. 26  In an attempt to deal with this kind of complexity, psychologists have started to adopt dynamical systems theory (e.g. Hotton and Yoshimi 2010 and Buhrmann et al. 2013).

Introduction  11 scale of the changes. Cultural change and cultural specificity become more salient, and are more easily grasped, against the backdrop of continuity. It can be no surprise, therefore, that the cognitive humanities have risen enthusiastically to the challenge of thinking cognitively and historically at the same time. Richardson’s field map included as one of its areas ‘cognitive materialism and historicism’, which has if anything grown further since 2004, such that the  2015 Oxford Handbook of Cognitive Literary Studies made ‘cognitive ­historicism’ the opening section.27 Under the influence of the 4Es—­one of them ‘embeddedness’—attention to the situated nature of cognition, including cultural and historical contexts, has become axiomatic.28 This volume is altogether typical in that it contains several chapters that are concerned with specifics, cultural and otherwise, and relatively few if any that make claims to undiluted universality. Grethlein, for example, elaborates how, despite noteworthy overlaps in other respects, the importance of deception distinguishes fifth-­century bce Greek notions of fictionality from their twenty-­first-­century cognitive counterparts, and Corthals and Sluiter compare Neoptolemus’ predicament at the beginning of Sophocles’ Philoctetes with a famous mid-­twentieth-­century psychological ex­peri­ ment and its twenty-­first-­century cognitive reinterpretation, exploring cross-­ culturally recognizable cognitive patterns as well as their situatedness in their respective cultures. Cultural specificity, then, has a secure place in the cognitive humanities,29 but it is worth drawing attention also to a different form of specificity, one that I have only mentioned in passing so far but of which psychologists in particular are keenly aware: the variation among individuals within a single group or culture. Individual difference is the bread and butter of some forms of psychology (witness journals like Personality and Individual Differences and Journal of Individual Differences),30 but can be difficult to come to terms with in the study of culture. In Classics and elsewhere, the focus on context that provides the basis of cultural history and that has been so successful in counteracting unthinking trans­his­tor­ ical generalization also about literature, can obscure equally significant differences from person to person. The major reason no doubt is the nature of the evidence: the historical record gives us relatively little to go on when we want to 27  Zunshine 2015, 13–81. See also two edited volumes, Zunshine 2010 on ‘cognitive cultural studies’ and Bruhn and Wehrs 2014 on ‘cognition, literature, and history’. 28  See esp. the special issue on ‘situated cognition and the study of culture’, Morgan et al. 2017. 29  This is not to say that all cognitive humanities work pays specific attention to cultural factors. A primary focus on universality characterizes above all evolutionary approaches to literature (‘literary Darwinism’), e.g. Carroll 2004, Gottschall 2008, Boyd 2009, Carroll 2011. Those are not represented in this volume, and have a somewhat uneasy relationship with mainstream cognitive literary studies; see e.g. Richardson 2004, 12–14. 30  It is true that the study of individual difference is at home above all in social psychology (as  indeed the Journal of Individual Differences and Personality and Individual Difference are social psych­ology journals), but cognitive psychologists, too, focus sometimes on individual difference; for a discussion addressing some of the methodological issues see Seghier and Price 2018.

12  Felix Budelmann talk about, for example, individual variation among the audience at the Theatre of Dionysus—­almost the inverse of the issue faced by psychologists studying audience response, who can find it difficult to get beyond the overwhelming variation. Perhaps, one might suggest, individual variation within the literary texts, while hardly a compensation, may nevertheless provide fertile material for thinking about the place of individual variation in our account of cognition in antiquity. Greek tragedy, with its strongly drawn personalities, even if one sees them as growing out of types, would be a case in point. Individual variation, though not necessarily under that name, certainly comes to the fore in several chapters of this volume. Van Essen-­Fishman, for example, studies the individually different ways in which certain friends and foes recall Ajax, and Scodel looks at the rather different ways in which Antigone and Ismene explain the causes of their actions. I end this section with two illustrative images that sketch two possible (and indeed compatible) modes of thinking about the relationship between the universal and the specific in cognitive criticism, by no means the only such modes, but each of them significant and instructive. The first is that of the critic as doctor.31 The gold standard of evidence-­based medical research is the randomized controlled trial: randomization minimizes bias, and comparison with a control group allows certainty in concluding that the observed effects are to be attributed to the treatment that is being tested. (Cognitive and social psychology use very similar methods.) Randomized controlled trials establish how a treatment works for a particular type of patient, but this patient is of course a statistical construct rather than an individual, and the doctor who treats the individual will need to combine her statistical knowledge with her knowledge about the medical history, the circumstances, and perhaps the personality of the patient to ascertain the desirability and efficacy of a particular medical intervention. Her job is not a precise scientific undertaking but a necessarily messy weighing-­up of possibilities that will often involve a good deal of judgement. This is the kind of messiness and need for judgement that confronts the cognitive literary critic who tries to determine the relevance of a particular finding from the cognitive sciences for a particular character in a Greek tragedy. There is no easy way of ‘applying’ cognitive science to Greek tragedy, but as for the doctor treating the patient, this does not mean that it is impossible or that we should not try. The second illustrative image, borrowed from Terence Cave, is that of the ‘cognitive archive’.32 In so far as literary works prompt, and reflect on, acts of cognition, Cave argues, the literary archive is also a ‘cognitive archive’ that provides cognitive case histories—‘histories of pathologies and deficits, off-­scale instances,

31  I owe this analogy to my fellow editor. 32  Cave 2017. This article arguably is the most searching discussion of the relationship between universality and history in cognitive literary studies to date, and I am not here providing a full account. The quotations are from p. 243 and the abstract.

Introduction  13 one-­offs, delusive texts, all produced by the human mind, all lending themselves to interpretation as symptoms or traces, all indicating possibilities and constraints across the whole cognitive spectrum.’ Literature yields to the cognitive critic a wealth of different, individually situated, instances of imaginative thinking about cognition, which collectively provide material for tracing ‘a history of cognition and reflection on cognition’. Studying literature as a cognitive archive, as a storehouse of moments of reflection on cognition, involves a form of literary in­ter­pret­ ation that is conscious of cognitive universals while also attending to the way these universals are contingently shaped by each text. For all its universal properties, cognition will always manifest itself in historical specificity, and as such invites historical study. The answer to the question in the section title, then, is ‘both, of course!’—the truism with which I began—­but how this is so varies from text to text and in­ter­ pret­ation to interpretation. All work in the humanities, including all literary interpretation, grapples with permutations of the dialectic between the unique and the general (the thwarting of generic expectations, the development of au­thor­ial style, metrical variation, etc.). Engagement with cognitive approaches further expands this range of permutations, and with it the scope for creativity. The concern that the universalizing impetus of cognitive science is bound to render work in the cognitive humanities reductive is unfounded, but the issues raised by this anxiety are hardly trivial, and it would be wrong to pretend that there is no tension here. The task for the cognitive critic is to make the tension a creative one.

1.3.3  The ‘what’ and the ‘how’: does cognitive criticism generate new readings? There is little, perhaps, that is more intimately familiar to us than cognition. It is constant acts of cognition, after all, that give us our experience of self and world. We know viscerally (to use a bodily metaphor) how it feels to wonder, expect, touch, hate, or see. This is why, despite the dramatic scientific advances over the past two decades, cognitive studies more often than not give us ‘merely’ a better grasp of something that at least in nucleo already feels familiar. This is not to say, of course, that cognitive science does not produce striking discoveries and shifts in our understanding, especially in areas where our intuitive sense of the op­er­ ation of the mind—­our ‘folk psychology’—is systematically mistaken. (The gradual dismantling of the deeply ingrained body-­mind dualism is perhaps the most obvious example, which has in turn fuelled self-­consciously provocative philosophical thought experiments, such as Andy Clark’s notion that we should think of the mind not only as embodied but as extending to the tools that support our cognitive activity). Nonetheless, and in contrast for example to poststructuralism, cognitive studies rarely lose touch with our experiential understanding of how we

14  Felix Budelmann relate to our environment—­they elaborate, highlight, explain, and indeed correct, but they usually are, somehow, compatible with the instincts that we derive from day-­to-­day experience. It is this rootedness in our experience of self and world that makes cognitive criticism gravitate away from against-­the-­grain readings or interpretations that uncover hidden meanings, and that has given rise to a long-­standing debate, taken up in this volume by Carroll (pp. 64–5), over whether new readings can ever be an aim of cognitive literary criticism. Should cognitive critics rather keep to questions of how meaning is achieved and stay clear of what has been the core activity of the literary critic since New Criticism, the uncovering and elaboration of what a text means? Not just those sceptical of cognitive literary studies, but cognitive literary scholars themselves take different views.33 I here try to articulate a defence of cognitive criticism’s ability to generate readings, a defence, however, that comes with qualifications. Before getting onto the ‘what’, we should note that the ‘how’ is indeed undeniably a strength of cognitive literary criticism; several chapters in this volume attend to mechanisms, processes, and means. A major contribution made in this respect by the various cognitive disciplines, no less important than empirical findings, is a rich conceptual vocabulary, such as ‘joint attention’ and ‘conceptual blending’ in Duncan’s and Gołąb’s discussions of modes of fictionality and spectatorship, or ‘image schema’ in Carroll’s analysis of metaphor. In some cases, such concepts re-­frame and thus re-­validate beleaguered notions. ‘Theory of Mind’, which is central to Murnaghan’s and Van Emde Boas’ chapters, has for a while now helped to reinvigorate the study of character by shifting the focus from the ontology of the character to the mind of the spectator who reads the character (a shift, however, that, as Van Emde Boas’ discussion draws out, is conceptually less straightforward than has sometimes been made out).34 Less widely assimilated in the humanities, but no less effective, is ‘attribution’, marshalled here by Scodel to discuss motivation and causation.

33 Good entry points into this debate, apart from Carroll’s chapter, are Caracciolo 2016a and Willemsen et al. 2018. An example of the sceptical position is this quotation from Henrik Skov Nielsen in Kukkonen and Nielsen 2018, 477: ‘The conclusions arrived at by means of these protocols for in­ter­ pret­ation [= cognitive narratology] often seem from an unnatural [viz. narratology] point of view unsurprising and unspectacular (indeed, bordering on the trivial) because they often run the risk of stating what is immediately clear to any even remotely competent reader’. Within Classics see the remarks of Sharrock 2018, 26. 34  In Classics, see Scodel 2009, Budelmann and Easterling 2010, Scodel 2012, Sluiter et al. 2013, and Scodel 2015, and the partly sceptical position of Grethlein 2015, but note the earlier work of, among others, Palmer 2004, Zunshine 2006, and Herman 2008. The terminology, and with it the conceptualization of the phenomenon itself, are debated, and ‘Theory of Mind’ is in this Introduction used simply as a convenient shorthand. Alternatives include ‘social cognition’, ‘mindreading’, ‘mentalizing’, and ‘intentional reasoning’. For an interesting exploration of the conceptualizations from a literary point of view, see Chesters 2014.

Introduction  15 Cognitive literary studies, then, certainly offer something other than new readings, and it would be a mistake to judge them only by their success or otherwise in contributing readings. However—­and this is a second response to the ‘no cognitive hermeneutics’ challenge—­it is simply not the case that cognitive critics have to drop the ‘what’ as they explore the ‘how’. Not only Carroll, who develops an interpretation of Pelasgus’ metaphors specifically to show that cognitive literary studies have as much to say about the ‘what’ as the ‘how’, but in fact the majority of the chapters in this volume, build their discussion around readings, readings that are often closely entwined with questions of ‘how’, but readings nevertheless. In several of these chapters, the critical move that opens up the text is the focus on the minds of the characters—­ a text-­ immanent study of cognition—­ but Grethlein’s metapoetic reading of the Paedagogus’ false messenger speech shows that there are other possibilities. Those readings vary between the ‘cognitive-­light’ and the ‘cognitive-­heavy’, and they differ also in their chosen attitude towards the cognitive material: most look for a fit, but Noel’s and Van Emde Boas’ chapters show that just as much intellectual energy can be generated by an exploration of the gaps. Behind these differences sits the dual quality of literature as both like and unlike life, which allows critics to choose their stance, alternatively leaning more on the match or the mismatch between cognition in everyday life and cognition in literature. Both are real, and both can engender fresh readings.35 Ultimately, I suggest, the most pertinent response to questions about the feasibility of cognitive readings is that much depends on what we mean by ‘readings’. Carroll in his chapter is surely right to argue that the distinction between the ‘how’ and the ‘what’ is fuzzy at best. Not only that analysis of how a text creates meaning will often lead to, in fact require, careful thought about what meaning is created, but there is a specifically cognitive dimension to this. In cognitive literary criticism investigation of the ‘how’ need not be simply a matter of stylistics or even poetics but can soon draw one into fundamental questions of human cognition. The ‘how’ of cognitive literary studies is always, at one level, a question about how humans register and interpret their environment, and here issues of meaning are never far off. (When Scodel, for example, writes about the way Antigone and Ismene explain their own and each other’s actions, is she engaged in a study in ‘how’ or in ‘what’?). Cognitive literary criticism, then, doesn’t and shouldn’t abandon the project of producing readings. Rather, what deserves attention, and what will, one hopes, continue to give rise to fresh thinking in the field, is the nature of those readings. In this respect cognitive criticism is in fact part of broader trends in recent literary theory, where the widely shared sense that hermeneutics narrowly conceived are not enough has led to experimentation with alternatives. Above all perhaps, there is the ever-­growing interest in the ethics of reading,

35  For the importance of giving sufficient attention to the mismatch, see e.g. Kukkonen 2019, ch. 1.

16  Felix Budelmann a development that started with feminist and postcolonial criticism and has radiated more widely since, and there is also the dissatisfaction with what Paul Ricoeur called the ‘hermeneutics of suspicion’—readings that seek to expose hidden ideologies in a text—­which has prompted, inter alia, experimentation with ‘surface reading’.36 Cognitive literary criticism may usefully be understood as contributing to this search for new forms of responding to texts, modes of response that are invested in generating readings but at the same time committed to exploring what is in fact meant by a ‘reading’.37

1.3.4  Two-­way traffic: can cognitive criticism give back to cognitive science? Claims to bi-­directionality are frequent in the cognitive humanities: many of those working in the field stress that the humanities, literary criticism included, have something to give back, though there is less agreement on the precise nature of the putative contribution.38 I shall briefly sketch two broad areas where (like others) I believe that a case can be made, one concerning the humanities’ modes of thinking, the other their material. Habits of thought in the humanities differ greatly from those in the sciences, and scholars of literature should seek ways of bringing to bear those characteristic habits also when engaging with the cognitive sciences. A convenient way of pinpointing these differences is by focusing on the fundamental scientific method of reduction. Not just the natural sciences but also experimental psychology routinely breaks issues down into components that are small enough to permit the design of manageable experiments, and the overwhelming majority of scientific articles discuss very specific phenomena. The humanities, for which ‘reductionism’ is anathema (cf. p. 13 above), have a rather different vantage point. One issue here is integration and a broad perspective. An (admittedly ambitious) example would be consciousness. Many psychologists and neuroscientists consider the study of consciousness impossible (‘the hard problem’), but for those in the humanities—­philosophers obviously, but also classicists, and indeed scholars of

36  Ethical criticism: I am thinking for example of Suzanne Keen’s work on empathy, starting with Keen 2007, but ethical criticism comprises a wide range of practices. For two rather different contributions see Phelan 2007 and Macé 2013, and from the perspective of cognitive literary studies Easterlin 2019. Against ‘suspicion’: Sedgwick 2003, Felski 2015, Moi 2017. Surface reading: the foundational article is Best and Marcus 2009. 37  With a view to classical literature specifically, one might add that ancient notions of in­ter­pret­ ation themselves offer models that are radically different from those made mainstream by twentieth-­ century criticism. Classicists are well-­placed to adopt a distinctive vantage point here. See further (also on the approaches listed in the previous note) Billings and Budelmann forthcoming. 38  For a volume foregrounding bi-­directionality, see Burke and Troscianko 2017; the subtitle is Dialogues between Literature and Cognition.

Introduction  17 literature—­it is a much more natural and useful level at which to operate than the amygdala. The humanities will not solve scientific problems on behalf of science and manage to work out the physiology of consciousness, but they can think usefully about fitting together the pieces and explore holistic perspectives. A further issue is the humanities’ habit of interrogation and interest in complexity that sits behind the aversion to reduction. Reducing an issue to its bare bones in order to make it suitable for experimentation involves strategically pushing aside complexity. For the experimental scientist or psychologist, this complexity is noise that has to be filtered out to make progress, whereas for the humanities the noise is often where the interest lies. An example of where attention to noise can be productive is what in the jargon is called ‘ecological validity’: the controlled lab setting or carefully constructed questionnaire is a necessary but in many cases rather imperfect replication of life, and scholars in the humanities, to whom labs and questionnaires are usually alien, are naturally sensitive to what gets lost in translation, irrespective of whether they are themselves involved in the design of an empirical study or are responding to published research. For literary critics this sensitivity will be greatest where texts are concerned—­most of the texts used in empirical studies are infinitely simpler and shorter than the tragedies that are the subject of this volume39—but almost any psychological questionnaire is bound to prompt niggling questions from a scholar in the humanities. Some such questions can be beside the point, but others have the potential to lead to a better understanding of what a result means and doesn’t mean. My second broad point returns to Cave’s ‘cognitive archive’, and the notion that the humanities ‘own’ vital evidence for human cognition. Art, language, music, and (in this volume) literature are products of cognition. They have an important place quantitatively—­we spend considerable time reading, listening to music, and so on—­but above all qualitatively. Reading or watching a Greek tragedy involves highly complex and varied forms of mental processing, as we voluntarily enter a world that is not real, yet employ many of the cognitive faculties we use to negotiate our day-­to-­day environment. What is more, the plays themselves insistently explore—‘experiment with’—their characters’ and audiences’ cognitive cap­abil­ ities and limitations, and in doing so they have proved successful in engaging audiences and readers in strikingly different settings. Greek tragedy thus constitutes prime material for researching human cognition, both historically (Cave’s archive) and contemporaneously, material that has on the whole been neglected by cognitive science, in part no doubt exactly because it is so complex. A properly satisfactory account of human cognition cannot exclude cognition in and of

39  Relatedly, see Willems and Jacobs 2016, an article by an experimental linguist and an ex­peri­mental psychologist arguing (inter alia) that literary works offer greater ecological validity than typical ­lab-­based stimuli.

18  Felix Budelmann literature, and an account of cognition in and of literature cannot exclude the humanities. This second point is particularly important. Currently, most interdisciplinary work between cognitive science and the humanities is initiated on the humanities side, as indeed most calls for bi-­directionality originate in the humanities, often with only limited resonance in the sciences. It is to be hoped that this will change in due course, but in the meantime it is inherently beneficial for classicists to understand their task not only as drawing on material from psychology or neuro­ sci­ence, or indeed philosophy, to see new things in old texts—­immensely pro­ duct­ive as that can be in its own right—­but also as tackling questions of cognition through the study of literature. The thing to emphasize is that they will achieve this very much as classicists and literary critics. Rather than trying to mimic the sciences, importing their methods and demands, classicists should see themselves as using their own materials (such as Greek tragedy) and their own methods (such as literary criticism) to engage in a conversation about a topic of shared interest, cognition. For this, they of course have to acquire a degree of expertise in the cognitive sciences, but above all, emphatically, they have to remain classicists.40

1.4  The chapters See above, pp. 5–7, on the overarching structure of the volume. Evert van Emde Boas’s chapter is the first of two concerned with Theory of Mind. Basing his argument on recent developments in psychology and the phil­ oso­phy of mind, he challenges the notion (frequently expressed in cognitive literary studies—­ and indeed in earlier applications to Greek tragedy), that the interpretation of literary characters is straightforwardly analogous to everyday mindreading. Instead, he argues, character interpretation typically relies on more specialized cognitive resources. Van Emde Boas also suggests that a narrow application of mindreading can overlook crucial features of literary characters that have long been of interest within literary criticism (particularly under the header of ‘realism’), and argues that an eclectic cognitive approach is needed to capture literary characterization properly. He concludes his chapter with a reading of Euripides’ Medea, a play which, he argues, both models and thematizes the un­usual cognitive processes involved in making sense of its protagonist. Sheila Murnaghan writes about the mind of Sophocles’ Ajax. She starts with the notorious challenges to interpretation posed by the Trugrede, and explores the 40  The conviction that literature, and literary methods, have something to contribute is a theme of Raphael Lyne’s excellent blog What literature knows about your brain and of Terence Cave’s work, e.g. (again) Cave 2017.

Introduction  19 Chorus’ and Tecmessa’s attempts to make sense of this mysterious speech. In particular, she highlights their different, understandable, yet variously deficient ways in which both attribute changes of mind to Ajax. Their responses to Ajax are too simple and too partial to do justice to the complexity of the speech, and assume linear temporal development where Ajax himself had spoken of circular change. Exposure to these flawed readings of Ajax, Murnaghan suggests at the end, may have ultimately contributed to a more expansive understanding on the part of the audience. Murnaghan’s conceptual basis in this exploration is Theory of Mind. Sophocles, she emphasizes, does not merely showcase Theory of Mind in its many different forms, but is fundamentally interested in exploring its limitations in a cognitively murky universe. Michael Carroll presents a close reading of the metaphors employed by Pelasgus in Aeschylus’ Suppliants as he responds to the Danaids’ request to grant them asylum. Using a range of cognitive tools, notably the notion of image schemas but also relevance theory and blending, Carroll reads the metaphors as not merely a form of literary expression but as also a mechanism through which Pelasgus evaluates the situation and considers what courses of action are open to him. Carroll explicitly confronts the question whether cognitive literary theory is capable of generating new readings, and argues that his own reading could not have been arrived at by means of conventional accounts of metaphor. Ruth Scodel uses the toolkit of attribution theory (in the version of Bertram Malle) to analyse the types of factors to which Sophocles’ Antigone and Ismene attribute their own and one another’s actions. She finds that Antigone sees herself as wholly rational while Ismene has doubts, a pattern that is replicated in the modern psychological record, that Antigone, even though fiercely critical of Ismene, does not accuse her of cowardice, and that Antigone’s low valuation of her own life is a crucial factor for her thinking. In passing, Scodel points out that attribution theory also opens the door to cross-­cultural comparison. For example, people in more individualistic cultures have proven more likely to underestimate situational factors. Lucy Van Essen-­Fishman traces the different ways in which different characters remember Ajax, both before and after his suicide, and shows that there is much to be learned about their respective views of the world from the manner in which they remember. She points out notable similarities with characteristics of memory observed in social and cognitive psychology. Sophocles’ characters resemble the subjects of recent empirical studies in that their memories are closely tied to their emotions as well as to their sense of identity, and in that they change with circumstances. Van Essen-­Fishman’s aim here is not so much to emphasize universality as to use current knowledge of memory to make us alert to certain aspects of the ancient text. Anne-­Sophie Noel engages Lambros Malafouris’ notion of the ‘cognitive life of things’. The idea of a reciprocal relationship between person and thing, she suggests,

20  Felix Budelmann is both fruitful and in need of modification for reading Greek tragedy. Tragic characters repeatedly invest lifeless objects with a mind (Noel’s case studies are the house in the Oresteia, the recognition tokens in Choephori, and the bow in Hercules Furens). However, Noel argues, this investment is a conscious form of make-­believe: the characters know that they are engaging in a fiction. Noel ends therefore by arguing for the relevance of work on anthropomorphism in developmental and social psychology. For the Watchman, Electra, and Heracles, to endow lifeless objects with a mind is a form of managing solitary and desperate situ­ ations rather than indication of a metaphysical belief about the cognitive powers of things. Hanna Gołąb draws on Fauconnier and Turner’s conceptual blending theory to discuss the audience’s engagement with the fictional worlds of Greek drama. Building on work in performance studies by Bruce McConachie and Amy Cook, she emphasizes that reality and fiction form a dynamic relationship, and that the audience’s role in determining this relationship is an active one. She examines ancient testimonia to distinguish the types of reality-­fiction blend that characterize tragedy and comedy, respectively. Jonas Grethlein reads the false-­messenger speech of Sophocles’ Electra in conjunction with Gorgias’ fragment B 23 DK. Both texts combine the aesthetics of illusion with the ethics of deception. Gorgias does so for rhetorical effect, whereas Sophocles’ play prompts the question whether the two can in fact be neatly sep­ar­ ated when it draws the audience into the speech while leaving them in no doubt that the story is false. The role of cognitive studies in this argument is twofold. It first provides the tools for examining the immersive character of the Paedagogus’ speech: Grethlein shows that the speech responds well to an enactive analysis. Secondly, Grethlein points to the limits of universality: on the one hand, there are interesting points of contact between Gorgias’ notion of aesthetic illusion and today’s understanding of the processes involved; on the other, the persistent emphasis on deception is a specifically ancient Greek phenomenon. A. C. Duncan uses the concept of joint attention (shared engagement with the same object, such as watching a football match in a crowd of people or reading a book with a child) to conceptualize the phenomenological distinction between watching tragic action in performance and visualizing the same action when reading the script. Duncan develops his argument by means of case studies of what he calls ‘sight invitations’, textual cues that direct the viewers’ visual attention. Sight invitations variously align or separate the internal and external audience, and are variously clear or ambiguous. As a result, joint attention itself becomes a ­varied phenomenon, but all examples, including the most problematic ones, are unified in demonstrating its great affective powers and its capacity for involving audiences in the plays. Seth L. Schein’s chapter tackles genre, distinguishing two approaches that are well-­established yet may usefully be revisited with a cognitive perspective. One is

Introduction  21 essentially categorizing: Eleanor Rosch and George Lakoff provide cognitive tools for addressing the notorious fuzziness of genre categories. The second—­Schein’s chief interest—­studies genre for the way it raises, fulfils, and subverts ex­pect­ ations. Reading Sophocles’ Philoctetes and Euripides’ Alcestis and Orestes as his chief examples, Schein discusses a range of different ways in which genre ex­pect­ ations are thwarted, and interpretative questions raised as a result. Schein ends by asserting the complementarity of humanistic and scientific viewpoints. Bob Corthals and Ineke Sluiter study the situatedness of (meta)cognition by comparing two scenarios of human ethical behaviour under extreme pressure in Sophocles’ Philoctetes and Stanley Milgram’s experiments in ‘obedience to authority’. Each of them advertises its own cognitive model: the opposition between natural disposition and deceitful rhetoric in the Philoctetes, and the very notion of ‘obedience to authority’ in Milgram. Milgram’s experiments have been reinterpreted in explicitly cognitive terms by Herbert Clark in terms of cooperation and social coordination. Corthals and Sluiter argue that the three cognitive models for what is ultimately a similar scenario derive their argumentative and illuminating power from their differential situated salience: the cultural and historical contexts come with their own pressing questions, which prioritize different types of answer. All form part of the long history of reflection on cognition, which began centuries before the ‘cognitive turn’ and is likely to continue for centuries beyond it.

Acknowledgements I am grateful to my fellow editor as well as to Katharine Earnshaw and Evert van Emde Boas for helpful comments on earlier versions.

PART I

R E A DING MINDS

2

Mindreading, Character, and Realism The Case of Medea Evert van Emde Boas

2.1 Introduction The editors of this volume have done more than most to further the enterprise of applying sociocognitive approaches to Greek tragedy. Two key collaborative publications—‘Reading Minds in Greek Tragedy’ (Budelmann and Easterling 2010) and ‘In Medea’s Head: Reading the Mind of a Murderous Mother’ (Sluiter et al. 2013)1—have established the concept of mindreading2 in our methodological arsenal.3 This chapter seeks to build on those two articles: I, too, will take insights from the cognitive sciences and put them to work in making sense of how audiences understand characters on the tragic stage, and how those characters are portrayed as understanding each other. Yet in a few respects my approach will differ from that of Budelmann and Easterling and that of Sluiter and her colleagues. The first pertains to how they position their work with respect to existing scholarship on characterization in tragedy. Budelmann and Easterling, for instance, argue, ‘[w]ithout detracting from the important insights prompted by discussions of tragic character’, that ‘certain features of the plays may today be captured more convincingly by . . . looking at the plays not as presenting us with characters but as modelling the dynamics of what we shall call “reading minds” ’.4 Similarly, if somewhat more equivocally, Sluiter et al. see Theory of Mind as an ‘alternative discourse’ to that of more traditional approaches to characterization, whose elements ‘are, of course, 1  The translation of this title from the original Dutch, and of any passages cited below, is mine. 2  A note on terminology: while I prefer ‘mindreading’ over the somewhat unfortunate ‘Theory of Mind’ (for the terminological issues involved, see e.g. Apperly 2011, 3–4; Hutto 2011, 278–80), any individual use of these terms (or others such as ‘social cognition’, ‘intersubjectivity’, etc.) below should not be taken as evidence of a theoretical persuasion. 3  Earlier works ‘flirting’ with the application of mindreading to Greek tragedy are Ruffell 2008 and Scodel 2009. More recently within classics (but not on tragedy), see Scodel 2012; 2015, Grethlein 2015 (suggesting, like the present chapter, limits to the applicability of mindreading), Budelmann 2018, Minchin 2019, Cairns 2021, Currie forthcoming. 4  Budelmann and Easterling 2010, 290.

Evert van Emde Boas, Mindreading, Character, and Realism: The Case of Medea In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0002

26  Evert van Emde Boas still relevant’.5 Such caveats notwithstanding, the proposed shift of critical focus, ditching the language of ‘character’ for that of ‘reading minds’, comes in my view with the risk of throwing several literary-­critical babies out with the bathwater. It also somewhat limits the scope of the possible contribution that cognitive approaches may make to the study of characterization in tragedy (and literature more generally). More on all this below. The second way in which my approach will diverge from that of my two ­models has to do with how we can apply the concept of mindreading itself, given the ongoing debate about it in the cognitive sciences and the philosophy of mind.6 The key question here is how much our experience of character interpretation in tragedy is like, or unlike, the processes involved in our basic everyday inter­actions with other people. Addressing that question will take up the first of three sections that follow (2.2). In the next (2.3), I will attend to a somewhat different conceptualization of the like-­or-­unlike-­real-­life question, namely that of realism, particularly as that concept has been used in assessments of the title character of Euripides’ Medea.7 My discussion of realism will serve as the springboard for the formulation of an overall cognitive approach to literary characterization. In my final section (2.4), these strands will come together in a fuller assessment of Euripides’ heroine through a cognitive lens.

2.2  Reading real minds and/or minds on stage The pendulum keeps on swinging. The 20th-­century debate about literary characterization can be seen in terms of ‘dehumanizing’ and ‘humanizing’ movements,8 first emphasizing the artificiality and textualness of literary characters (in reaction to older criticism which had naïvely approached literary characters exactly as if they were actual persons with lives extending beyond the page or stage), and then gradually coming to terms with the troublesome fact that most readers and audiences most of the time have a ‘sense that fictional characters are uncannily similar to real people’.9 Yet in this century, work in the growing field of ‘unnatural

5  Sluiter et al. 2013, 12–13. 6  To be fair, neither Budelmann and Easterling nor Sluiter et al. go into the debate within the cognitive sciences at length, as their focus lies elsewhere. 7  The same figure took centre stage in Sluiter et al. 2013: in this respect too, my chapter may be seen to build on and respond to that article. More recently, see also Cairns 2021 (and the other contributions in the special issue of Greece and Rome of which Cairns’ article forms part, particularly that of Allan). 8  The terms are those of Culpeper 2001, 6–9. Such movements can be traced in literary criticism generally but also specifically in work on Greek tragedy. 9  Martin 1986, 120.

MINDREADING, Character, and Realism: the Case of Medea  27 narratology’ has once again argued for a conception of literary character—­or, at least, some literary characters—­that is strongly unlike real life.10 The cognitive twist on this debate frames it as a question of whether or not the same psychological mechanisms are at work when readers and audiences make sense of literary characters as in our everyday interactions. The answer to this question from the corner of cognitive literary scholars has been a remarkably (and suspiciously) unified ‘yes’, and often (more suspiciously still) that ‘yes’ is taken for granted rather than argued for. Perhaps the most elaborate explicit defence of the affirmative answer is provided by David Herman, one of the leading figures of cognitive narratology, who devotes his introduction to an edited volume to countering what he calls the ‘Exceptionality Thesis’ (‘the claim that readers’ experiences of fictional minds are different in kind from their experiences of the minds they encounter outside the domain of narrative fiction’).11 As for work on Greek tragedy, the initial offerings of Budelmann and Easterling and Sluiter and her co-­authors both appear to side with Herman, and emphasize similarity: [F]undamental to the functioning of drama (and indeed literature more widely) is its constant manipulation of the same mind-­reading skills that spectators and readers also use all the time in real life. (Budelmann and Easterling 2010, 291–2; my italics.) We can ascribe emotions, thoughts and intentions to others . . . We do this all the time. People, according to cognitive scientists, have a ‘Theory of Mind’—really an unfortunate term, but what is meant by it is that we continually and often automatically form hypotheses about what others think, feel, and mean. . . . We can come at Greek tragedies with these ideas in mind, since characters are walking invitations to do mindreading . . .  (Sluiter et al. 2013, 12; my italics.)

Yet there is hardly a consensus in the cognitive sciences about the nature of these mindreading skills, and where one stands in that debate has fundamental implications for how snug we deem the fit between everyday mindreading processes and the interpretative processes involved in making sense of an Oedipus or a Medea. I will argue, in fact, that current trends in cognitive science suggest that the exercise of reading literary minds is one that calls predominantly on in­ter­ pret­ative processes that in our day-­to-­day interactions we use not ‘all the time’ or 10  Some of the key relevant texts in unnatural narratology (something of a counter-­movement to the ‘natural narratology’ of Fludernik 1996) are Alber et al. 2010, Alber and Heinze 2011, Richardson et al. 2013. Although there is a focus in this work on postmodernist texts, adherents also emphasize ‘the inherent unnaturalness of conventional forms . . . in traditional works of realism’ (Alber et al. 2010, 131). 11  Herman 2011, 8.

28  Evert van Emde Boas ‘continually and often automatically’, but in fact relatively exceptionally and with significant cognitive effort.12 I take my cue here from recent attempts by cognitive scientists and philo­ sophers to find a way out of the long-­standing impasse between proponents of ‘theory theory’ and those of ‘simulation theory’.13 According to theory theory, humans ascribe thoughts, feelings, and intentions to each other through a process of inference from observable actions, based on a set of commonsensical beliefs and expectations—­ a folk-­ psychological ‘theory’—about how people tend to behave. Simulation theory, on the other hand, holds that rather than attributing mind states to others on the basis of inference, we use our own mind as effectively a simulator of other minds: we run through the thoughts and feelings of a target in our own mind (although we do this ‘off-­line’, decoupling the relevant mind states from their usual role in governing our own behaviour), and use them to form predictions about the resulting behaviour and further mind states of that target. Two independent recent trends have changed the outlines of this great debate. First, cognitive scientists have increasingly argued for hybrid models, and emphasize that mindreading should be seen as a complex and multi-­faceted phe­nom­ enon which draws on a wide range of cognitive faculties. Thus, Ian Apperly, who explicitly sets his work against the ‘TT–­ST’ debate, offers a ‘two-­systems’ account (observe that each of his two systems is itself complex): Simple ascriptions of perception, knowledge and belief are achieved via cognitively efficient ‘low-­level’ processing modules . . . In contrast, more complex and flexible, ‘high-­level’, mindreading makes use of the same general knowledge and inferential processes available for any other reasoning; it is more cognitively effortful . . .  (Apperly 2011, 108)

Apperly also suggests ‘that the role of mindreading in everyday social cognition is perhaps more limited than is often claimed’, and that many instances of social cognition may depend instead on such factors as ‘interactive alignment, social scripts, and knowledge of normative conditions for everyday interaction’.14 12  This has bearing on an argument frequently made in evolutionary studies of literature, namely that creating and consuming fictional narratives are evolutionarily beneficial because they allow humans to practise Theory-­of-­Mind skills (e.g. Zunshine 2006). If reading fiction equates to mindreading practice, it is practice of a very specialized set of mindreading processes. For the latest psychological and neuroscientific evidence on the link between fiction reading and Theory-­of-­Mind ability, see Tamir et al. 2016, Van Kuijk et al. 2018. 13  For helpful discussion of the debate between theory theory and simulation theory, see Apperly 2011, 176–9, Samson 2013. A good overview of the current state of psychological research is Apperly 2012. The implications of the debate for literary criticism are drawn out by e.g. Palmer 2004, 143–7, Zunshine 2006, Vermeule 2010, 35–48, Herman 2011, and, taking fuller account of recent contributions by philosophers of mind, Chesters 2014, Van Duijn 2016, 37–47. 14  Apperly 2011, 117. Significantly, Apperly here sees mindreading (proper) as only one element of social cognition (others might treat these terms as synonymous; for such terminological issues see also n. 2).

MINDREADING, Character, and Realism: the Case of Medea  29 Separately, philosophers of mind, most notably Shaun Gallagher and Daniel Hutto, have argued that what is involved in what they call ‘intersubjectivity’ is typically neither theory nor simulation—­indeed no kind of mentalistic ascription at all. Gallagher’s ‘interaction theory’, which is aligned with embodied and en­act­ ive theories of cognition, rejects what he sees as a the fatal flaw of the two main theories, namely ‘the Cartesian idea that other minds are hidden away and inaccessible’.15 Instead, [i]n most . . . situations of social interaction, we have a direct perceptual understanding of another person’s intentions because their intentions are explicitly expressed in their embodied actions and their expressive behaviors. This understanding does not require us to postulate or infer a belief or a desire hidden away in the other person’s mind. What we might reflectively or abstractly call their belief or desire is expressed directly in their actions and behaviors. (Gallagher and Hutto 2008, 20–1.)

A related point made in interaction theory is that humans in everyday social settings generally do not adopt a third-­person, observational stance to form hypotheses about others’ mind states. ‘Rather, our everyday encounters with others tend to be second-­person and interactive’.16 The qualifications in ‘most situations’, ‘generally do not adopt’, ‘tend to be second-­person’, are significant. Interaction theory does not deny that, ‘in circumstances instituted by . . . complex social practices and normative formations’, we regularly need ‘something more’ than basic perception and embodied interaction (Gallagher 2020, 173–4). Gallagher frames this ‘something more’ not in terms of theory or simulation, but in terms of communicative actions17 and (following Hutto’s earlier work on a ‘Narrative Practice Hypothesis’) narrative competence—­we understand others by fitting their behaviour into narrative patterns that we have internalized since childhood.18 For all their differences, what these newer approaches share is an emphasis on the complexity of social cognition, and on the idea that not all forms of it are alike. Both camps—­proponents of hybrid models and interaction theorists—­ suggest that most of our everyday social interactions involve fairly automatic, deep-­rooted processes, while higher-­level processes (the two strands differ in how these are conceived) take over only under certain, cognitively more challenging

15  Gallagher 2009, 292. The initial formulation of the theory is Gallagher 2001; certain aspects are updated in Gallagher 2005, ch. 5, Gallagher and Hutto 2008, and most recently the complete synthesis in Gallagher 2020 (esp. Part II). The theory finds its roots in classic continental phenomenology, especially the work of Husserl, Scheler, and Merleau-­Ponty: see, e.g. Gallagher 2001, 91–7. 16  Gallagher and Hutto 2008, 19. 17  Gallagher 2020, 155–60. 18  Gallagher and Hutto 2008, Gallagher 2020, 160–9. Hutto’s theory is presented most fully in Hutto 2008. He himself sketches some implications for literary criticism in Hutto 2011.

30  Evert van Emde Boas circumstances.19 And with this we come back to the question of similarity versus difference: it is reasonable to assume, I posit, that processing literature is an exercise that calls with greater frequency and greater intensity on cognitive resources that are of the more complex kind. The point, then, would be that comprehending literary characters, while not relying on mechanisms that we never use otherwise, nevertheless also does not reflect the automatic-­and-­all-­the-­time processes of social cognition that we use most frequently in our day-­to-­day lives. Two points of clarification and nuance are needed here. First, what I am talking about is a difference of degree (or, perhaps, balance) rather than one of kind. Making sense of other people in real life, too, is often a complex business, and it does call on us regularly enough to use the more advanced kinds of explanatory processes sketched above. In turn, literature (and drama, film, etc.) may often enough offer us moments where a character’s thoughts and intentions are more or less transparently accessible, especially if it provides sufficient physical cues for the embodied system to pick up on.20 Still—­recognizing the peril in any gen­er­al­ iza­tion about ‘what literature does’—we can assume that the interpretation of literary characters tends towards the complex side of the scale. In fact, the extent to which characters require interpretation that is non-­automatic, difficult, even li­able to go astray, is often thematized in literary works themselves: the reading of Medea I offer below hinges on this notion. Secondly, the fact that real-­life social cognition is itself not a single, all-­ encompassing mental ability has consequences that touch on the very foundations of the literature-­ versus-­ life distinction. Whether one frames the more complex mechanisms at work as ‘the same general knowledge and inferential processes available for any other reasoning’ (Apperly) or as mainly a matter of narrative competence (Gallagher and Hutto), it is clear that we are dealing with inferential processes based on concepts that are to no small extent culturally determined. As Marco Caracciolo has put it (2016b, 18): If we think of folk psychology as a toolset, beliefs and desires are relatively universal tools. Even more universal, it seems, is our capacity to grasp other people’s 19  Incidentally, as both strands point out, it is far from clear how these different-­level processes relate to the tasks at the heart of many early experimental studies (which are in turn often cited, but with their implications regularly left underexplored, in cognitive literary studies). Part of Gallagher’s critique, for instance, is that the kinds of experiments psychologists use to investigate Theory of Mind bear little resemblance to everyday experience (e.g. Gallagher 2005, 216–24, Gallagher 2020, 168–9). For discussion of the (often problematic) ways in which the results of psychological experiments have been linked to the processing of literature, see Van Duijn 2016, chs. 1 and 3; 2018. 20  Drama occupies a special position on this score: it is evidently more ‘embodied’ than purely written literature in that it presents actual bodies whose expressive actions can be ‘read’ by the audience; on the other hand, the stance that audience members must by necessity take in interpreting these cues is that of a third-­person observer, at a remove (literally) from the interactions happening on the stage. This in itself is a point of significant distinction from our everyday encounters with others, which ‘tend to be second-­person and interactive’ (Gallagher and Hutto, cited above).

MINDREADING, Character, and Realism: the Case of Medea  31 intentions and emotions as they are expressed by bodily movements and facial expressions . . . But surely human psychology is much more complicated than this: when interpreting other people’s behavior we can employ more sophisticated constructs . . . —theories of mental causation and psychological mo­tiv­ ation, for instance, or cultural models of intersubjectivity (romantic love, the mother-­son relationship, etc.). There is no doubt that these more advanced tools for understanding mind are provided by culture.

The fact that these advanced instruments come from culture chips away at the distinction between literature (as a determiner of culture) and life. Indeed, the upshot of Hutto’s Narrative Practice Hypothesis is that the stories we tell to each other, particularly to our children, are determinative of how we interpret each other in real life (provided that more basic, embodied methods of sensemaking cannot do the trick).21 One need only attach a name to the son in Caracciolo’s example of the mother-­ son relationship to see how this cycle can work: Oedipus—­or rather, Freud’s awkward reading of that character—­for a long time informed mainstream psychological thought about mother-­son relationships, and that line of thought to this day informs how some would make sense of certain characters in literature—­even including Oedipus.

2.3  Realism and character: the case of Medea The conclusions of my previous section pull in interestingly different directions: on the one hand I have argued that literature can be assumed to make a greater appeal to the more advanced instruments in our folk-­psychological tool set than most of our daily social interactions do; on the other hand, I observed that the cultural dependency of those advanced instruments begins to blur the distinction between life and literature. In this section I want to draw out some of the consequences of these tensions for the like-­or-­unlike-­real-­life debate as it has been held in more traditional literary criticism, particularly under the header of ‘realism’, and especially as it pertains to characterization in Euripides.22 The notion that the figures populating Euripides’ plays are somehow just like real people has a particularly long pedigree—­ dating back at least to the Aristotelian anecdote according to which Sophocles claimed (presumably derisively) that Euripides created people ‘as they are’.23 Yet Euripidean characters also 21  For the view that storytelling scaffolds our folk psychology, see also e.g. Herman 2009 and (from a slightly different perspective) Van Duijn 2018. See also Gallagher 2020, 163, on narratives, including ‘cultural narratives’ (n. 2) ‘leak[ing] into our everyday engagements’. 22  I develop here some of the ideas about realism in Euripidean characterization first set out in Van Emde Boas 2018. Euripidean realism more generally is well discussed in Lloyd 2020. 23 Arist. Poet. 1460b34.

32  Evert van Emde Boas figured prominently in the 20th-­century wave of scholarship that held that the realistic portrayal of character or psychology was of no concern at all to the Greek tragedians. One sophisticated exponent of that trend, John Gould, uses Medea as the key example in his Euripides entry in the Oxford Classical Dictionary:24 . . . there are strands of ‘realism’ in Euripides’ writing for the theatre: for example, Medea’s presentation of herself as mistrusted ‘foreigner’ and oppressed and exploited ‘woman’ (Med. 214–58) and her subsequent slow, tortured progress to infanticide . . . But these are strands only in an extremely fragmented whole. For it is arguable that a vision of human experience as inherently fragmented . . . forms the very heart of Euripidean sensibility. If we go back to Medea and read it attentively, we shall find that the Medea we have encountered . . . exists, within the world of the play, alongside other Medeas: before the passage mentioned, she has been heard off-­stage, . . . articulate only in universal cursing and damnation . . . immediately after it, she is transformed into a subtle adversary who patently and easily outwits her most powerful enemy. Subsequently she becomes successively brilliant orator, pathetic victim, devious manipulator, exultant (and uncanny) avenger, tormented mother until her final metamorphosis . . . into the demonic figure who . . . closes the play . . .  (Gould 2012)

In a similar but more trenchant fashion, George Gellie complained that ‘[a]ll these Medeas cannot fail to get in each other’s way’; that Euripides makes no attempt ‘to let those processes [i.e. Medea’s switches] emerge from a unitary personality structure’, because he is ‘less concerned than we are that a psychologically coherent human being should direct the course of his play’; and that the reason for this is that ‘myth and real life, or even realistic stage-­life, work so differently from each other’; in the end, Gellie concludes, ‘Medea just escapes us’ (Gellie 1988, 16–18, 22). I cite from these articles at length because they are suggestive of the limits, at which I hinted at the start of this chapter, of approaching literary character solely in terms of mindreading (at least as narrowly conceived). When Gould describes Medea as an example of the kind of fragmentation which lies at ‘the very heart of Euripidean sensibility’, and when Gellie emphasizes the contrast between myth and real life, such formulations remind us that literary characters may be seen as something other than ‘people-­like things’—that is, formed by (and formative of) other aspects than their supposed psychological endowments, backstories, and so  forth. Many theoretical models have attempted to capture these different

24  It does some injustice to Gould’s work to assimilate it so simply to ‘Tychoism’ (for the term, see e.g. Goldhill 1986, 171), and of course much work in that vein (beginning with Tycho von Wilamowitz’ 1917 book itself) is not so crudely dualistic as it has sometimes been made out to be. Gould expounded his views on characterization more fully in Gould 1978.

MINDREADING, Character, and Realism: the Case of Medea  33 constitutive dimensions of literary character; an elegant one is the tripartite model of James Phelan (1989, 2–3), who distinguishes between a ‘mimetic’ component (capturing the ways in which the interpretation of characters is analogous to that of actual persons), a ‘synthetic’ component (capturing the ways in which characters are artificial constructs, the product of textual cues, narrative patterns, preconditioned mythical roles, and so forth—­in the case of drama we should add theatrical aspects), and finally a ‘thematic’ component (capturing the ways in which characters can be taken as the embodiment of particular larger themes or ideas, or the expression of a particular world view—­Euripidean sensibilities included). Regardless of whether one agrees with Gould and Gellie about Medea, few would argue that aspects such as myth or artificiality are wholly unimportant in assessing Euripidean characters. And this leads into one obvious criticism that can be made of some of the existing approaches to literature that have used mindreading as their central avenue of approach, namely that they ignore Phelan’s synthetic and thematic components and are receptive only to the mimetic. Thus, Brian McHale warns that cognitive narratology (as currently pursued) lacks an ‘attentiveness to literary convention’ (2012, 123), and Maria Mäkelä argues that ‘by reducing fictional minds to exempla of actual human cognition we miss the essential dynamics between verbal art and real-­life experientiality’ (2013, 130). There is certainly something to these critiques. But the risk for such critics, in turn, is that they fall into the same trap that (I would argue) Gellie and Gould fell into, which is to see the different dimensions of literary character as mutually exclusive.25 The solution, I think, is not to dismiss the cognitive model, but to opt for a richer version of it. I would make three points here, which together amount to something of a ‘programme’ for a cognitive approach to literary character and characterization. First: our approach should attempt to canvass all of Phelan’s dimensions of literary character. Capturing not just the mimetic but also the conventional and artificial aspects of characterization in a cognitive model is perfectly possible,26 as long as we recognize that such aspects are not somehow inherent qualities of the characters but that they belong to readerly interpretation. The inherent focus of cognitive approaches on reception (there is a natural affinity with reader-­response work) allows us to clarify that what we are talking about when referring to 25  When Gould and Gellie ‘throw in the towel’ (as Sluiter et al. aptly put it, 2013, 7–8, 19) on the attempt to make sense of Medea at a personal level, they demonstrate an approach which takes the synthetic as coming at the expense of the mimetic. I am, of course, not the first to argue against such rigid divisions: for Greek literature, still hugely important are the essays in Pelling 1990; see also De Temmerman and Van Emde Boas 2018a, 11–19. 26  See recently also Polvinen and Sklar 2019 for an attempt to frame Phelan’s model in cognitive terms; Polvinen and Sklar take issue, as this chapter does, with Herman’s outright rejection of the ‘Exceptionality Thesis’ (above, p. 27).

34  Evert van Emde Boas mimetic, synthetic, and thematic ‘components’ of literary character is really different interpretative strategies on the part of readers and audiences, strategies which can vary from audience member to audience member and which can, for individual audience members, co-­exist and interact. Secondly, our approach should be sensitive to the interactions between the various aspects of literary character. If we accept that mimetic, synthetic, and thematic aspects are all interpretative strategies, we still seem to be left with some strategies (mimetic ones) which are more, and others (synthetic and thematic ones) which are less like our everyday experiences with others. But here it is worth taking seriously the conclusions of my previous section. The complex nature of social cognition means that the differences between mimetic, synthetic, and thematic aspects are, while not irrelevant, also not as clear-­cut as they may  seem to begin with. We use all kinds of culturally determined—­arguably ‘thematic’—models in our interpretation of actual people, too: we may think back here to Caracciolo’s examples of romantic love and the mother-­son relationship. And as we have already seen, such concepts may be informed by literature as much as by real life: there is multi-­directional traffic between the levels. Thirdly, we should allow for what I might call cognitive eclecticism: specifically, we should see the science of mindreading as only one pillar of our approach. The scientific debate about mindreading, as very narrowly conceived, could even be seen as something of a red herring for literary studies, in that some of the ex­plana­ tory processes involved in making sense of literary characters have not necessarily been at its heart.27 But this does not mean that cognitive science cannot offer other insights on what is at work in those explanatory processes. There are well-­ developed models of (social) cognition that allow us to describe them with some precision, and some of these models have already been used to good effect in studies of literary characterization.28 One important one is attribution theory, discussed by Scodel in this volume.29 Another is a more general theory of how we process new information by forming mental models, combining the top-­down activation of pre-­existing schemas and scripts (connected information stored as  packages in memory) with the bottom-­up piecemeal integration of new information.30 This theory is particularly useful because it captures several 27  See n. 19 above. 28  Good surveys include Eder et al. 2010 (also their thematic bibliography in the same volume); Herman 2013, §3.2, (f) and (g), Jannidis 2013. An earlier integrative framework combining various psychological models is that of Culpeper 2001, 2002. 29  Also see e.g. Culpeper 2001, Palmer 2007. 30  For this model as it pertains to literary characterization, see e.g. Culpeper 2001, 26–38; Schneider 2001; it is applied to Greek literature in De Temmerman and Van Emde Boas 2018b, 16–17; script theory is used independently by Minchin 2011. There is an ongoing debate in cognitive science about whether models such as this one are reconcilable with enactive/embodied models of cognition like that of Gallagher, described above (both Gallagher and Hutto adhere to a version known as ‘radical’ enactivism; see e.g. Hutto and Myin 2013, Gallagher 2017); for a moderate proposal, pertaining particularly to language comprehension, see e.g. Zwaan 2014.

MINDREADING, Character, and Realism: the Case of Medea  35 fundamental aspects of character interpretation: first, that it is a dynamic process, in which audience members constantly integrate new information about characters into the image of them that they have already formed;31 and second, that in forming and adjusting character models audience members make use (just as they do in real life) of a whole host of character-­related schemas (social categories, prototypes and stereotypes, etc.), possibly including predominantly literary ones (mythical and narrative roles, character types such as ‘the wicked stepmother’ or ‘the wise adviser’, etc). Max van Duijn has convincingly argued, in fact, that the application of such schemas makes the task of mindreading literary characters tractable to begin with.32 In sum, these theories (and others) when taken together offer a rich framework for understanding what goes on in the interpretation of literary characters, in a way that any one of them individually may not. One other important contribution that such a pick-­and-­mix cognitive approach can make is that it can give greater precision to some of our standby terms and concepts, including such notoriously slippery ones as realism. We do not need cognitive approaches, of course, to observe that realism is a vague notion that can capture a wide range of interrelations between a literary text and ‘reality’.33 But cognitive approaches may help us to get a handle on some of the key distinguishing features of certain variants of it. The version of realism that Gould and Gellie argue is lacking in Medea, for instance, is one that deals primarily in notions of ‘coherence’, and such a version hinges on the dynamic modelling process described in my previous paragraph.34 For Gould and Gellie, the character schemas that go into forming the Medea model are too numerous and too diverse, and as a result the integrative modelling exercise falls flat.35 Another version of realism would focus more on ‘depth’, that is, the feeling that there is ‘something there’ (one gets the sense that Gould’s ‘strands of “realism” ’ are to be taken this way). Such a version seems to depend on the extent to which a text or play offers sufficient mimetic cues in the first place for our folk-­psychological mechanisms to latch onto. Another conception of realism pertains to the specific kinds of character schemas and categories that are activated, especially those of lower-­ class as opposed to higher-­class people: this is the ‘social’ (or even ‘democratic’) realism of characters who are more like ‘us’ than like the larger-­than-­life heroes 31  This is the ‘construction’ of Easterling 1990, revisited in Budelmann and Easterling 2010, 291. Piecemeal integration of information is also at the heart of the argument of Sluiter et al. 2013. 32  Van Duijn et al. 2015, Van Duijn 2016, ch. 2. Apperly, cited above (p. 28 with n. 14), also asserts the importance of social scripts in understanding others. 33  For some attempts to untangle realism as it pertains to (Euripidean) tragedy, see Goff 1999, Budelmann 2013, Lloyd 2020. I leave aside in this chapter the ‘cognitive realism’ of Troscianko 2014. 34  This is clear from Gellie’s opening paragraph (1988, 15), which frames the spectator’s activity as ‘collect[ing], as the play proceeds, the clues . . . that will help us to categorize the kind of person a stage-­ figure is meant to represent’ (note the idea of categorization). 35  On some accounts realistic characterization means idiosyncrasy, and comes close to what Forster called ‘roundness’. On such a view realism must strike a delicate balance, relying neither on too few character schemas (overschematization) nor on too many.

36  Evert van Emde Boas and kings with whom they contrast.36 Of course, all these notions of realism—­ various others could be added—­can and will sometimes overlap, but they may also diverge. Whether any of them are applicable to Medea is the question that I address presently.

2.4  Medea’s multifarious mind This final section represents a slight methodological pivot: rather than theorizing about how interpretation works, I will for the remainder of this chapter be ‘doing’ interpretation. There is a concurrent shift of focus away from the mindreading processes of the theatre audience towards those of the characters portrayed onstage, and with that shift, much of the methodological daylight between my own approach and that of Budelmann, Sluiter, and their respective co-­authors disappears. Indeed, the germ of my reading of Medea is largely present in Sluiter et al.’s rich analysis of the mindreading modelled onstage (by other characters of Medea and vice versa, and by Medea of her own mind). There is a crucial difference of emphasis, though: on my view, what the play models—­and indeed brings about for its audience—­is not so much an exercise in run-­of-­the-­mill, automatic mindreading processes, but above all the difficult cognitive struggles that arise when such interpretative efforts break down. I also suggest that looking at the text in this way offers a corrective to the vision of a fragmented multi-­Medea offered by, for instance, Gould and Gellie—­a reading which leaves ample room for a mimetic Medea, however enigmatic, alongside their purely synthetic one.37 ‘If ’, to use Gould’s phrase, ‘we go back to Medea and read it attentively’, looking closely at those moments where other characters attempt to assess what is going on in Medea’s mind, and those moments where she does so herself, we will find that Medea escapes not just us, but the other characters too. Medea is different from one moment to the next, but in part because she means to be, and in part also because she is subject to genuinely complex emotional and psychological impulses. Her intentions and motivations are difficult to read because she goes to great lengths to hide them, and because they are actually shifting and sometimes conflicting. All this makes Medea a formidable challenge for the mindreader, whether intra- or extradramatic. But, crucially, this is a motivated challenge: the reasons for it are there to find in the play itself, and they are explicitly reflected on by the characters, not least Medea herself. 36  I have argued elsewhere that this kind of realism is operative in the opening scenes of Medea, particularly in the discussion between the Tutor and the Nurse (Van Emde Boas 2018, 358–9). 37  On this point, too, my argument latches on to points made in Sluiter et al. 2013. However, to my mind that piece fails to make, at least explicitly, a crucial final step in the argument, namely that the extent to which the intractability of Medea’s mind is both modelled and explicitly reflected on in the play itself serves, ironically, to vitiate a Gould-­and-­Gellie-­style reading.

MINDREADING, Character, and Realism: the Case of Medea  37 With the revealing exception of Jason, who confidently (mis)attributes beliefs and desires to Medea in the agôn scene,38 and who in the deception scene accepts Medea’s false account of her change of mind with disastrous naïveté, other characters consistently find it difficult to determine what she is thinking. This is true right from the start of the play: the Nurse in her prologue speech is the first to worry about Medea’s intentions, which she cannot pin down with precision, although she relates them ominously to the children:39 She hates her children and takes no joy from seeing them. And I am afraid that she may plan some new untoward thing. στυγεῖ δὲ παῖδας οὐδʼ ὁρῶσʼ εὐφραίνεται. δέδοικα δʼ αὐτὴν μή τι βουλεύσῃ νέον· (36–7)

Sluiter et al. rightly point out the importance of the embodied gesture of Medea’s gaze for the Nurse’s interpretative efforts. They comment (2013, 13): ‘The Nurse has no other cue for this explicit mindread than Medea’s gaze: she devises a hypothesis about what is going on in Medea’s head.’ But does she, really? If the Nurse is explicit about anything it is that she is unclear about Medea’s precise intentions: all she can get at is the sense that there is something inauspicious (τι . . . νέον) going on, worthy of fear (δέδοικα).40 Such mindreading difficulties are portrayed again and again in the words of the Nurse herself and others in the play. In her subsequent anapaestic exchange with Medea, the Nurse repeats her fears for the children (117–18), and connects to them a prescient generalization about the temperamental emotions of the high-­ and-­mighty (‘the characters of our rulers alter their passions harshly’, δεινὰ τυράννων λήματα . . . χαλεπῶς ὀργὰς μεταβάλλουσιν, 119–21).41 Later, in the

38  E.g. at 555, 568, 569–71, 588–90, 621. 39  The question of interpolation in 38–43 has bearing on this (for discussion, see Mastronarde 2002 and Mossman 2011 ad loc.). If the lines are allowed to stand the Nurse at least fills in alternative possibilities for Medea’s plans. But to my mind the case for deleting all of 38–43 is strong. 40  One of the anonymous readers suggests that we should take the Nurse’s imprecision as eu­phem­ism: she avoids any explicit mention of death ‘in the fear that mention could become the “cause” of the event’. This would have implications for our interpretation of the parallel between the Nurse’s fearful uncertainty and that of Creon, discussed below. To my mind, however, it is an important part of Euripides’ manipulation of audience emotion that the threat to the children remains at least somewhat vague at this point of the play, so that it can still come as ‘a shock when Medea announces she will murder them at 792’ (Mossman 2011 ad 36). The question of whether the original audience was or was not already familiar with versions of the myth in which Medea kills the children, as well as related questions about possible intertextuality with Neophron’s Medea, need not be seen as too important in assessing this suspense curve: Euripides’ strategy works either way. 41  Some interpreters take χαλεπῶς . . . μεταβάλλουσιν rather as ‘change [i.e. let go of] their anger with difficulty’ (so e.g. Mastronarde 2002 ad loc.), in which case the Nurse’s gnomic statement would refer to obstinacy rather than volatility. In my view the translation offered here is preferable in view of

38  Evert van Emde Boas Aegeus scene, there is another neat example of an act of embodied understanding which breaks down, in the Athenian king’s question to Medea why her face and complexion are ‘wasted away’ (689: τί γὰρ σὸν ὄμμα χρώς τε συνέτηχ’ ὅδε;): Aegeus cannot make sense of Medea’s expressive behaviour so he cannot read her mind, and he does not even guess at it: in order to reliably attribute thoughts and intentions to her, he has to ask her directly.42 Similarly, the Tutor fails to comprehend Medea’s verbal and expressive cues in reaction to his report that the children have delivered the robes (1008–14), a reaction which is not ‘consonant’ (ξυνῳδά, 1008) with what he believes to be good news. And so on.43 The most direct parallel for the Nurse’s fearful uncertainty—­including almost exact verbal similarities—­is found in the Creon scene. The king cites his fear as grounds for mistrusting Medea’s seemingly conciliatory tone in asking to be allowed to stay in Corinth:44 (Medea:) I’ll keep quiet, even though I’ve been wronged: I’ve been bested by those who are stronger than I am.—(Creon:) What you say is comforting to hear, but I have a fear that you are plotting an evil scheme against me in your mind. I trust you so much the less than before. For a sharp-­tempered woman—­ and just so a man—­is easier to guard against than a clever one who keeps quiet. MH.               καὶ γὰρ ἠδικημένοι σιγησόμεσθα, κρεισσόνων νικώμενοι. ΚΡ. λέγεις ἀκοῦσαι μαλθάκʼ, ἀλλʼ ἔσω φρενῶν ὀρρωδία μοι μή τι βουλεύῃς κακόν, τοσῷδε δʼ ἧσσον ἢ πάρος πέποιθά σοι· γυνὴ γὰρ ὀξύθυμος, ὡς δʼ αὔτως ἀνήρ, ῥᾴων φυλάσσειν ἢ σιωπηλὸς σοφή. (314–20)

Creon’s justification is revealing: Medea is a γυνὴ . . . σιωπηλὸς σοφή, a woman who is both dangerously clever and who keeps her thoughts to herself. Sluiter et al. comment that Creon, in his inability to access Medea’s thoughts, is ‘a kind of Gellie avant la lettre’ (2013, 14). I wholly agree; yet what deserves greater em­phasis is that this holds for almost everyone else in the play, too; that the play itself amply motivates such difficulties of comprehension (as we will see presently); and that it what follows: ‘For it is better to be accustomed to living on terms of equality’ (τὸ γὰρ εἰθίσθαι ζῆν ἐπ’ ἴσοισιν κρεῖσσον, 123–4), i.e. within moderating limits. 42  This is a good example of what Hutto calls ‘the horse’s mouth principle’ (2008, 20; 2011, 280). 43  Other moments/characters that could be discussed in detail: e.g. Jason at 929, the Messenger at 1129–31, the Chorus at 1265–7. 44  The fear clauses in 37 and 317 are strikingly similar, and call attention to the differences (Creon’s more explicit κακόν for the Nurse’s νέον, and if the emendation is accepted, his more immediate present subjunctive βουλεύῃς (‘are planning’) for her aorist subjunctive βουλεύσῃ (‘will plan’)).

MINDREADING, Character, and Realism: the Case of Medea  39 is this explicit foregrounding of mindreading troubles which, ironically, makes readings such as those of Gellie and Gould fundamentally problematic. If Gellie had accorded greater interpretative weight to the difficulties of his avant-­la-­lettre counterpart(s), he might himself have been less troubled by Medea’s multiplicity. Creon, by his own admission, cannot fully trust his own mindreading faculties in approaching Medea, because he fears deception. He is, of course, quickly proven right. In the first of several moments in which Medea herself signals that her outward behaviour and her actual intentions do not align, she reveals to the Chorus how she has tricked the king: Do you think that I would ever have fawned on this man if I weren’t gaining some advantage or plotting some scheme? I would neither have spoken to him nor touched him with my hands. δοκεῖς γὰρ ἄν με τόνδε θωπεῦσαί ποτε εἰ μή τι κερδαίνουσαν ἢ τεχνωμένην; οὐδʼ ἂν προσεῖπον οὐδʼ ἂν ἡψάμην χεροῖν. (368–70)

Sluiter et al. effectively show (2013, 14–15) that the way in which Medea frames this admission underscores her own formidable mindreading abilities: it shows how skilfully she played Creon and how well she anticipates the Chorus’ (mistaken) interpretation of her earlier behaviour. Significant here, I would add, is Medea’s use of ‘flatter’, ‘soothe’ (θωπεύω), which gives an ominous spin to Creon’s earlier description of Medea’s words as ‘comfortable’, ‘soothing’ (μαλθακά). As it turns out, Medea’s deception hinged on a carefully curated manner of speaking. This is important: the play in this way explicitly rationalizes the shifts of register which so troubled Gellie, Gould, and others. Medea’s ‘multi-­vocal quality’, as Mossman saw (2011, 44), rather than being the mere result of Euripidean formalism and fragmentation, belongs to her ‘supreme skill at persuasion, at being all things to all men (and women)’. Creon’s adjective μαλθακός returns several hundred lines later, as Medea details to the Chorus the outlines of her upcoming δόλος: Now I will tell you all of my plans: hear my words which give no pleasure. I will send one of my servants to Jason and ask him to come to see me. And when he comes I will speak soft words to him, saying that I agree with him . . . ἤδη δὲ πάντα τἀμά σοι βουλεύματα λέξω· δέχου δὲ μὴ πρὸς ἡδονὴν λόγους. πέμψασʼ ἐμῶν τινʼ οἰκετῶν Ἰάσονα

40  Evert van Emde Boas ἐς ὄψιν ἐλθεῖν τὴν ἐμὴν αἰτήσομαι. μολόντι δʼ αὐτῷ μαλθακοὺς λέξω λόγους, ὡς καὶ δοκεῖ μοι ταὐτὰ . . .  (772–7)

Again, Medea’s speech makes clear that her deception hinges on adjusting her manner of speaking to suit her interlocutor.45 And again, she leaves no doubt that this is all by design: these switches of tone belong to her ‘plans’ (βουλεύματα; we have seen the key βουλευ- root used earlier by both the Nurse and Creon). And yet Medea’s βουλεύματα are not the only force driving her behaviour: they compete with other motivations and impulses, as we well know from Medea’s ‘Great Monologue’ (1019–80). I do not propose to go into that speech (or its text­ ual problems) at great length;46 what is important about it for my purposes is that it brings out into the open Medea’s internal conflicts, and moreover that it shows her sophisticated awareness of those conflicts—­Medea as an accomplished reader of her own mind. As scholars have seen, Medea’s centrepiece is amply prepared for by earlier moments in the play: her self-­exhortation to action, after being granted her extra day by Creon (401–4), first shows us a woman in dialogue with herself, urging herself to do terrible deeds (and thus obviously aware of the impulses working against them); when she first mentions the filicide (791–6), it is again clear that she feels the opposing forces in herself against which she must steel herself, ‘daring’ or ‘enduring’ (τλᾶσ’ 796) her most heinous crimes; Medea narrates a similar self-­dialogue—­albeit a fake one—­to Jason in her deceptive speech at 873–81; and such a dialogue makes another brief reprise after the great speech, when she departs the stage at 1236–50. Even the dramatic shift in the prologue, from the agonized woman whose cries we hear from inside the house to the composed speaker who comes out to meet the women of the chorus—­the shift which so troubled Gellie, Gould, and others—­can plausibly be seen in this light, as a sign of Medea’s assertion of self-­control over conflicting emotions and impulses.47 The great speech itself is a lengthy exhibition of Medea’s inner turmoil, which leads her to entertain the possibility of letting go of her βουλεύματα (1044, 1048), and in which—­however one reads the famous last lines 1079–80—those βουλεύματα stand in some kind of tension with her ‘passionate impulse’, ‘temper’ (θυμός). 45  We ought not to be surprised that μαλθακοὶ λόγοι means something very different in the re­spect­ ive cases of Creon and of Jason. Her persuasion of Creon depends on a speech (292–315) whose rhetoric has ‘a certain frigidity’ (Mossman 2011, 44 n. 161) and on an appeal to his pity for the children (344–7); her speech to Jason (869–905) is much more colourful, and precisely calibrated to appeal to the views that he expressed in the agôn scene. For good analysis of this speech, particularly her remarkable use of self-­quotation at 873–81, see Mossman 2011 ad loc. and Sluiter et al. 2013, 15–16. 46 For discussion and bibliography I refer to the commentaries of Mossman and Mastronarde (including his appendix), and Rutherford 2012, 315–22. 47  So e.g. Seidensticker 2008, 342, Mossman 2011, 45.

MINDREADING, Character, and Realism: the Case of Medea  41 More generally we may note the abundance of psychological ter­min­ology Medea uses in describing her own mental and emotional states: among the factors at play are her ‘heart’ (καρδία, 1042), ‘mind’ (φρήν, 1052), and her θυμός (1057, 1079); all the while she ‘experiences’ (πάσχω, 1049), ‘wants’ (βούλομαι, 1049, 1069), and ‘lacks willpower’ (οὔκετ’ εἰμὶ . . . οἵα τε, 1076–7).48 In short, the scene makes it difficult to ascribe a coherent set of mind states to Medea, for the simple reason that her mind is in actual turmoil. We might even want to call this ‘fragmentation’, but it is fragmentation of a very different kind from that meant by Gould. From the outset, then, the audience is confronted with, on the one hand, various characters who express their inability to access Medea’s mind, combined with her own disclosures to the Chorus that she is being deceitful; on the other hand, repeated expositions by Medea herself of the various emotional impulses which pull her in different directions. Together, Medea’s craftiness and her shifting mo­tiv­ations make it almost impossible to pin down her thoughts and intentions with precision at any point. This is true for the external audience as much as for the characters; but unlike the characters, the audience at least get a complete sense of why reading Medea’s mind should be so difficult. Still, that knowledge hardly makes Medea’s mind more penetrable. It is no surprise in this light that so many critical chestnuts of the play remain open questions to this day: to what extent is she being deceitful when speaking to Aegeus? At what point has she decided to kill the children? Is she motivated by sex or not? When Mastronarde, having noted that ‘[t]here are . . . many factors involved in Medea’s motivation’, suggests that ‘no simple interpretation of her behaviour or character is to be privileged’ (2002, 20), that is the expression of a healthy critical posture to adopt when reading between the lines of dramatic texts. In this particular play, however, it is a matter of reading the lines themselves as much as reading between them. Medea is a prime example of a literary work that asks us to do folk-­psychological work at a level of complexity that we will rarely encounter in our everyday lives (little surprise—­we don’t meet filicides every day). And with this, we come back one last time to the notion of realism. How psychologically realistic is Medea? Unsurprisingly, given the conclusion of my middle section, the answer depends on one’s conception of realism. The exercise of reading her mind is certainly not ‘just like real life’, or at least I suspect it isn’t for most of us (or, for that matter, for most members of the original fifth-­century audience). But, as Sluiter et al. rightly point out (2013, 18–19)—it is perhaps their core argument—­most of us also do not stop trying. The alternative, the Medea whose mind ‘just escapes us’, who so distorts our folk-­psychological processes that we have to give up and accept that she is nothing but a fragmentary artefact,

48  For the importance of such ‘composite mind’ terminology in Euripidean characterization, see Thumiger 2007a, 65–74.

42  Evert van Emde Boas seems unattractive not least because the play itself so consistently calls attention to Medea’s mental states. There has been a great deal of interest recently in ‘strange’, ‘unnatural’, or ‘deviant’ characters in contemporary fiction, ranging from autistic first-­person novelistic narrators to serial killers in films. In his study of some such characters, Marco Caracciolo (already cited above) helpfully traces the area where realism and strangeness can overlap: readers can still judge as realistic ‘characters who seem to elude, problematize, or even threaten their folk-­psychological skills’. The at­tend­ ant feelings of strangeness may, in turn, even lead them to revise their own folk psychology: ‘fictional and real minds are in constant dialogue via readers’ interpretive strategies—­and feelings of strangeness . . . play a key role in opening and sustaining this dialogue’ (2016b, 20–21). Medea fits this mould. She so threatens our comprehension that some interpreters have called her ‘demonic’. Yet even demons can be real, and meeting them in books, films, and plays can broaden our horizons. That, as Greek tragedy so often does, is what Euripides’ play achieves.

Acknowledgements I am grateful to the editors and three anonymous readers for their invaluable comments. For helping me to get to grips with realism, I am indebted to presenters and audiences at the ‘Realism’ seminar series in Oxford in 2015, and to an audience in Reading. I also wish to acknowledge productive discussions with two graduate students in Oxford, Charlotte Franssen and Oscar Harrington-­Shaw, who read earlier drafts of this chapter.

3

Reading the Mind of Ajax Sheila Murnaghan

3.1 Introduction In a suggestive and influential essay entitled ‘Reading Minds in Greek Tragedy’, Felix Budelmann and Pat Easterling open a new critical avenue for students of Greek tragedy by drawing on work in psychology, philosophy of mind, and neuro­sci­ence that is concerned with what is often called ‘mindreading’. This body of research by cognitive psychologists such as Simon Baron-­Cohen, Dan Sperber, and Bertram Malle, and cognitively inclined literary critics such as Alan Palmer and Lisa Zunshine, focuses on the ‘Theories of Mind’, or ‘folk psychologies’, through which all human beings perform the fundamental, indispensable task of figuring out what other people are thinking. Budelmann and Easterling propose that paying attention to how the tragedians represent mindreading might be a fruitful alternative to the more familiar, but vexed and contested quest to capture their portrayals of character. Several decades of debate within tragic criticism have shown that character is an elusive phenomenon and arguably more per­tin­ent to modern than to ancient literature. They aim instead ‘to investigate the benefits of looking at the plays not as presenting us with characters but as modelling the dynamics of . . . “reading minds” ’ (290), an approach that can illuminate the analogous experiences of figures within plays and of the audiences who have to make sense of what is enacted before them. After offering exemplary analyses of episodes from Aeschylus’ Agamemnon and Sophocles’ Antigone, they call for further investigations along similar lines, mentioning scenes of ‘deception or madness’ (303) as especially promising material. While Budelmann and Easterling offer the concept of mindreading primarily as a new tool for interpreting tragedy, tragedy also has the capacity to clarify the particular utility of mindreading as one concept within the range of terms and models through which the capacity to envision the thoughts, intentions, and emotions of others is designated in current debates. Mindreading indicates a particularly cerebral and self-­conscious way of understanding others and is situated, along with the closely related concept of ‘Theory of Mind’, at one end of a spectrum that also includes more embodied and pre-­reflective forms of social cognition. Tragic plots involve just the kinds of baffling, unforeseen, and disruptive

Sheila Murnaghan, Reading the Mind of Ajax In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0003

44  Sheila Murnaghan events that demand the most deliberate forms of interpretation.1 Tragedy epit­om­ izes the way that literary works can, as Evert van Emde Boas puts it in his chapter in this volume, ‘[ask] us to do folk-­psychological work at a level of complexity that we will rarely encounter in our everyday lives’ (41), often with the effect of expanding our cognitive horizons.2 Van Emde Boas develops this important observation through a reading of Euripides’ Medea, in which (building on Sluiter et al. 2013) he analyses Euripides’ portrayal of effortful mindreading on the part of various characters who have to contend with a protagonist, Medea, who is exceptionally hard to interpret. This portrayal is sharpened and made more instructive for the play’s audience by those characters’ self-­conscious reflections on their difficulties and by various indications, coming both from them and from her, of what makes Medea so opaque. She is herself deliberately deceptive, and her state of mind shifts. She is subject to conflicting emotions that are correlated with different intended acts: love for her children that makes it impossible to kill them and a desire for revenge that makes it imperative that she do so. The interpretative challenges faced by the characters are motivated in ways they themselves have access to (though not always soon enough to avert disaster) and can themselves articulate. This chapter concerns another tragedy, Sophocles’ Ajax, that offers an equally illuminating but differently conceived account of mindreading under especially challenging and consequential conditions. The characters in this play are caught up in less self-­conscious attempts to make sense of an extraordinary mind than in Medea, and the playwright does not afford them or the play’s audience the kind of access to that mind that Euripides allows for Medea. Like Euripides, Sophocles provides a nuanced, closely observed account of human beings exercising their  mindreading capacities, but within a more capacious cosmic vision in which mindreading, as a merely human capacity, has significant limits. The play sheds light on mindreading as an indispensable human resource, in ways that are examined in some detail below, but it also uses instances of proficient but inadequate mindreading to indicate a superhuman reality that is beyond ordinary cognition.

1  Greek tragedy in particular reflects a general preoccupation with deliberation that was pervasive in the culture that produced it (Hall 2009, 74–9) and a performance context in which the non-­verbal clues on which people often rely for interpreting others were attenuated, since the actors were masked and viewed by audiences from a considerable distance. 2  As Timothy Chesters puts it, in the challenging situations presented in certain literary texts, ‘the attribution of, and conjecturing over, concealed mind-­states has . . . become necessary because the normal course of things has gone awry’ (2014, 71). For complex mindreading as ‘a skill that is needed when the context requires deviation from a default’, see Van Duijn 2016, 13; for the idea that particular contexts demand a resort to ‘Theory of Mind’ rather than less reflective but more efficient cognitive capacities, see Hutto 2008, 7, Apperly and Butterfill 2009, Apperly 2011, 108–42, Van Duijn 2016, 44–7. On the question of terminology, see further Van Emde Boas’s chapter.

Reading the mind of Ajax  45

3.2  The challenge of Ajax’ ‘deception speech’ The challenges to mindreading presented by the protagonist of Sophocles’ Ajax are most pronounced in relation to one of those scenes of deception of which Budelmann and Easterling especially invite investigation: the so-­called ‘deception speech’, or Trugrede (646–92).3 This speech is notoriously difficult to interpret in  terms of conventional signs of character such as coherent motivation or ­consistent identifying traits, and it is not even clear that it is intentionally deceptive. Ajax emerges from his tent, into which he has exited with the ­evident intention of killing himself, and appears to his auditors, his war-­captive partner Tecmessa, and the loyal followers who form the chorus, to renounce that intention. He speaks of his plans in terms that seem to indicate a purification ritual rather than a suicide, and he correlates those plans with yielding to the gods and revering the sons of Atreus, the leaders whom he hates and whom he had planned to escape and thwart by killing himself. But his declarations are so ambiguously worded and his intentions so opaque that the speech can be also read in the opposite way, as a statement that Ajax still intends to kill himself rather than that he does not. For most (although not all) interpreters, Ajax’ actual intention when he gives the speech can be determined retrospectively on the basis of the fact that he does go ahead and kill himself.4 But the recognition that Ajax’ intention to die remains constant even as he presents himself as having changed creates further challenges, revealing the inadequacy of any simple equation between intention and state of mind and complicating one of the most serviceable functions of mindreading, which is to predict what someone will do next.5 And it leaves unresolved the question of what Ajax’ intentions are in relation to the speech itself. If he intends it to have the impression on its auditors that it does, then he must be deliberately deceiving them. But for many scholars, going back to F. G. Welcker in the nineteenth century, it is simply ‘out of character’ for a traditional hero like Ajax to be deceptive. If that consideration is put aside, whether because it is not so clear that heroes are incapable of being deceptive, or whether because Ajax has in fact been 3  For lucid overviews of the difficulties raised by the speech and the principal solutions that have been proposed, with bibliography, see Hesk 2003, 74–95, Lardinois 2006, 213–14; for detailed commentary, see Finglass 2011, 328–41. 4  It is harder to gauge the likely responses of original audience members, who would have expected Ajax’ suicide as part of the traditional legend but would have also been prepared for an individual playwright to introduce surprising innovations. A group of modern spectators who were presented with the speech after being told Ajax would go on to kill himself tended to choose the interpretation that attributes the least changeability to Ajax, namely that he was being deliberately deceptive (Budelmann et al. 2016, 105–9). 5  As John Gibert puts it, ‘. . . the indeterminacy of his intentions before and during the speech can lead us to make an unusual distinction, which is encouraged and rewarded, between “mind” and “intentions” ’ (1995, 120). On predicting action as one of the practical uses of mindreading, see Baron-­Cohen 1995, 21.

46  Sheila Murnaghan acting ‘out of character’ throughout the entire play, it remains hard to see how deceiving his auditors, and especially Tecmessa, can be reconciled with the opening words of the speech, in which Ajax volunteers an account of his own state of mind. All things that are obscure are brought forth by long and uncountable time, and hidden again when they have been revealed. Nothing is beyond expectation; the dread oath and the unflinching purpose can be overcome. For I too, who was mightily steadfast like iron when it has been dipped, have been feminised in my mouth by this woman. I feel pity at leaving her a widow among my enemies, and my child an orphan. ἅπανθ᾽ ὁ μακρὸς κἀναρίθμητος χρόνος φύει τ᾽ ἄδηλα καὶ φανέντα κρύπτεται· κοὐκ ἔστ᾽ ἄελπτον οὐδέν, ἀλλ᾽ ἁλίσκεται χὠ δεινὸς ὅρκος χαἰ περισκελεῖς φρένες. κἀγὼ γάρ, ὃς τὰ δείν᾽ ἐκαρτέρουν τότε βαφῇ σίδηρος ὣς, ἐθηλύνθην στόμα πρὸς τῆσδε τῆς γυναικός· οἰκτίρω δέ νιν χήραν παρ᾽ ἐχθροῖς παῖδά τ᾽ ὀρφανὸν λιπεῖν. (646–53)6

With a strong emphasis on change as an inescapable if unexpected fact of life, Ajax announces that he now feels a new sympathy for Tecmessa. Deceiving Tecmessa seems, then, like a strange way to respond to new feelings of sympathy: his deception of her could only be temporary and would be followed by even greater pain. But new sympathy for Tecmessa is equally incompatible with the idea that Ajax is not trying to deceive her and thinks of himself as announcing an unaltered intention to kill himself, since she has been intent on persuading him not to do that, especially by trying to awaken his sympathy for her. As Easterling puts it, ‘There can be no answers to the question of what he intends vis-­à-­vis the people who hear him’ (1984, 5). When Ajax seems to equate his intended action with yielding to the gods and doing reverence to the sons of Atreus, this has struck a substantial number of critics, among them Bernard Knox (1979b, 141), R. P.  Winnington-­Ingram (1980, 49), Christopher Gill (1996, 211), and Jon Hesk (2003, 80–1), as so inconsistent with his heroically proud character and his unwavering determination to die that he cannot really mean it and must be speaking sarcastically. This view is grounded especially in the fact that Ajax in line 667 speaks of yielding to the gods with a term more appropriate to temporal leaders (εἴκειν) and revering the sons of Atreus with a term more appropriate to gods (σέβειν). Such unconventional usage 6  Quotations in both Greek and English are from Finglass’s edition, unless otherwise noted.

Reading the mind of Ajax  47 demands interpretation, and the detection of sarcasm is one possible way of accommodating the gap between Ajax’ words and what he might be expected to say. It is also a solution that highlights the interpretative impasse that gives rise to it, drawing attention to a point at which a text cannot be comfortably processed in a more straightforward way. Finally, Ajax voices a perspective on his coming death that departs from or­din­ ary modes of knowing and understanding, couched in various forms of indeterminate or superhuman expression. He uses traditional gnomic expressions in ways that summon up their multiple and contradictory meanings (Lardinois 2006); he speaks of the future in the visionary terms of a divinely inspired prophet (Scodel 1984, 23); his language takes on the otherworldly significance of words spoken in the course of mystic initiation (Seaford 1994a, 395–7, 1994b, 282–8). It is not uncommon in tragedy for speakers to convey more than they as humans can be expected to know, but those speakers are usually marked as exceptional, whether as actual prophets with special access to the divine or as members of choruses whose identity as characters within the drama is supplemented by their role as vehicles for broader communal traditions and forms of wisdom. In other cases, speakers may say more than they realize because they are victims of dramatic irony, whose limited understanding prevents them from recognizing their true situation as well as the audience does; but Ajax’ words are confusing precisely because their meanings are hard to delimit and the audience has too inconclusive a sense of his self-­understanding to measure it against his circumstances. The expansiveness of Ajax’ vision most likely reflects the special significance of his death as a prelude to cult heroism, but that significance is hinted at rather than spelled out in the play. It is not clear how well Ajax understands his own future in ritual.7 The softening towards Tecmessa that Ajax claims has occurred during his time offstage, something to which we have no direct access, has brought with it a new way of speaking: as he describes it, he has been made female with respect to his speech, ἐθηλύνθην στόμα (651). As a result, the suicide we are yet to see is described in different terms and endowed with a different meaning from the suicide we had expected to hear had happened offstage, but the relationship between that new meaning and Ajax’ thinking is impossible to pin down. There is more being said in this speech than can be correlated with an ordinary sense of a person’s state of mind, and its pervasive ambiguity frustrates interpretation; for the

7  In this respect, he differs from Oedipus in OC, who expresses a stronger sense of himself as marked for a special death and afterlife, whether or not he was in fact the object of a comparable cult in Sophocles’ day. On the different degrees to which Sophocles accentuates the protagonist’s possible heroization in Aj. and OC, see Henrichs 1993, 176–8. On Ajax’ ability to voice a super-­human, quasi-­ choral vision as a function not of his character but of his position of extraordinary isolation, see Buxton 2006, 21–3.

48  Sheila Murnaghan audience member or critic, the rhetoric of the speech can only ‘block . . . a move from words to mind’ (Goldhill 1986, 190).

3.3  Internal readers of Ajax’ mind Instead of trying to bring this speech into line with any workable concept of character, it may be more profitable to consider how its internal auditors interpret it as they deploy the mindreading techniques that are part of their basic equipment as human beings. We are offered several brief but telling interpretations by the Chorus and then Tecmessa in the unsettled stretch of time between when the speech is delivered and when its meaning is, at least in part, settled by Ajax’ suicide. Inevitably, these characters’ readings of this over-­determined speech are partial, both in the sense that they are selective and in the sense that they are filtered through the interpreters’ own concerns. Responding to both what Ajax says and what he seems to say, they offer a set of variations on the idea of a ‘change of mind’, a series of formulations that show what a serviceable, even indispensable concept ‘change of mind’ can be, but also how dubious and slippery. The first of these comes in the famously misguided euphoric choral ode that follows immediately after the speech (693–718). In the strophe, the Chorus cry out that they are impelled to dance and sing by overwhelming joy, which they describe in erotic terms, ἔφριξ’ ἔρωτι, ‘I thrill with longing’ (693, my translation); in the antistrophe, they articulate their reasons, which are inflected by their own strong desires and ecstatic state of mind. Ares has taken dreadful pain away from my eyes. Io, io, now once again, now, Zeus, can the bright light of happy days shine on the swift ships that sped across the sea, now that Ajax has gone back to forgetting his troubles and fulfilled the rites of the gods with all their sacrifices, doing them reverence in the greatest good order. Great Time causes all things to decay. And I would declare that there is nothing you can say is impossible, now that Ajax has unexpectedly been led to repent of his anger against the Atridae and his mighty quarrel. ἔλυσεν αἰνὸν ἄχος ἀπ᾽ ὀμμάτων Ἄρης. ἰὼ ἰώ, νῦν αὖ, νῦν, ὦ Ζεῦ, πάρα λευκὸν εὐάμερον πελάσαι φάος θοᾶν ὠκυάλων νεῶν, ὅτ᾽ Αἴας λαθίπονος πάλιν, θεῶν δ᾽ αὖ πάνθυτα θέσμι᾽ ἐξήνυσ᾽ εὐνομίᾳ σέβων μεγίστᾳ. πάνθ᾽ ὁ μέγας χρόνος μαραίνει·

Reading the mind of Ajax  49 κοὐδὲν ἀναύδητον φατίσαιμ᾽ ἄν, εὖτέ γ᾽ ἐξ ἀέλπτων Αἴας μετανεγνώσθη θυμοῦ τ᾽ Ἀτρείδαις μεγάλων τε νεικέων. (706–18)

The Chorus’ interpretation is rooted in a close, selective reading of Ajax’ words, as their near-­quotation of the beginning of his speech indicates; they echo his opening gnomê about the power of time to efface everything that exists, and also his assertion that the fact that he has somehow changed is proof that even the most unexpected things can happen.8 But they attach those sentiments to the fulfilment of their own desire; they make them apply to the particular change of plan that they want for him and for themselves, although it is not at all clear that this is what they refer to in Ajax’ speech, where they are offered only to explain his altered mode of speaking. The change of plan the Chorus impute to Ajax is taken to imply a broader change of feeling or attitude: they assume he is no longer angry at the sons of Atreus. This is a plausible inference from his statements that he will learn to revere them and that as leaders they must be yielded to, but it is powerfully contradicted at the point of his suicide by the bitter curses he utters against them (835–44). In keeping with the lyric mode of the passage, the Chorus seem to lose track of linear time. The change that they understand Ajax to have undergone is not a forward progression into a new state but a return to a previous condition. Getting a little ahead of themselves in time, since they assume that Ajax has already carried out a set of purifying rituals, they also position him back in the past, declaring that he ‘has gone back to forgetting his troubles’ (λαθίπονος πάλιν). This idea of returning to a past state may be present as well in the otherwise unattested verb that the Chorus use for Ajax’s change of mind, μετανεγνώσθη, which Stanford in his commentary glosses as ‘has had his mind changed back again’ (1963, ad 717–­18). The passive voice of μετανεγνώσθη is also remarkable, especially since the closely related and more fully attested verb μεταγιγνώσκω (as opposed to μεταναγιγνώσκω, which occurs only here) never appears in the passive. Changing his mind is apparently not something that Ajax does but rather something that is done to him. Ajax himself also uses the passive voice to report on his perceived internal change when he announces that he has been feminized, ἐθηλύνθην (651).9 According to Bernard Knox, for whom ‘[t]he mainspring of Sophoclean tragedy is the hero’s stubborn refusal to change’, these formulations show what an alien, unchosen experience this is: ‘the idea is expressed by phrases which present a 8  Such close verbal echoes are typical of statements in which people apply a gnomic expression to a particular situation (Lardinois 2006, 216). 9  Finglass 2011, 351. On the use of the aorist passive in Greek to express non-­volitional changes of mental state, which expanded during the classical period, see Allan 2003, 64–76, 158–60.

50  Sheila Murnaghan change of mind not as a personal decision but as something imposed from ­outside or else in pejorative metaphors which suggest it is a thing to be avoided, above all by heroes’ (1979c, 232–4). Knox may be too quick to dismiss the possibility that Ajax could retain the same plan of action while undergoing a significant inner change, but his emphasis on outside agency is helpful.10 Despite their single-­ minded, self-­oriented, and erroneous conviction about Ajax’ specific intention, the Chorus do pick up something of the deception speech’s sense of unfathomable external forces at work along with its mystical collapsing of time.11 But when the messenger sent by Teucer from the Achaean camp arrives and the dramatic mode shifts from song and dance to speech and action, the Chorus reiterate their conviction that Ajax has adopted a new plan in quite different terms, which put Ajax very much in charge of the trajectory of his own mind. The messenger asks if Ajax is at home, and they respond: He is not inside, but has departed just now, after yoking new counsels to new actions. οὐκ ἔνδον, ἀλλὰ φροῦδος ἀρτίως, νέας βουλὰς νέοισιν ἐγκαταζεύξας τρόποις. (735–6)

Through an image of internal yoking that evokes self-­mastery, Ajax is said to have made new plans that accord with new tropoi—­a term which is variously glossed as ‘mood’ (Stanford), ‘temper’ (Garvie), ‘disposition’ (Jebb), ‘habitus’ (Kamerbeek), or ‘state of mind’ (Gibert 1995, 127). This formulation once again foregrounds the idea that changed plans are a manifestation of a broader change in attitude or affect, but the Chorus now project a quite different narrative from the one they suggested before: not a return to the past but purposeful forward movement in a new direction, a mental progression that is acted out in Ajax’ physical movement away from the hut (reflected in Patrick Finglass’s translation of νέοισιν . . . τρόποις as ‘new actions’, quoted above). When the messenger counters that Ajax’ de­part­ ure from the hut might not be a cause for celebration if Calchas has prophesied correctly, the Chorus is unmoved, voicing again the view that Ajax has relinquished his earlier anger: ‘But he is gone, I tell you, after turning to more prof­it­ able thoughts, to make an end of his anger against the gods’, ἀλλ᾽ οἴχεταί τοι, πρὸς τὸ κέρδιον τραπεὶς | γνώμης, θεοῖσιν ὡς καταλλαχθῇ χόλου (743–4). The messenger flatly dismisses this last statement as ‘full of great foolishness’, μωρίας πολλῆς πλέα (745) and both of the Chorus’ optimistic assertions contain darker undertones of which they are unaware; forms of νέος often connote what is 10  On Knox’s tendency to overestimate the inflexibility of the ‘heroic temper’ in this and other contexts, see Gibert 1995, 135, 255–62. 11  As Jebb notes, this use of the passive ‘does not mean that the Chorus suppose Ajax to have been converted by themselves or Tecmessa: the cause is left indefinite’ (1896, ad 717).

Reading the mind of Ajax  51 not only new but also bad, and the Chorus’ emphatic οἴχεται is also ‘unconsciously ominous’ (Jebb). It is only after the messenger’s extended report on the words of Calchas that the Chorus are finally convinced that they have indeed foolishly misread Ajax, and they summon Tecmessa to hear this news. This then precipitates Tecmessa’s own interpretation of Ajax’ intentions when giving the speech, which differs from the Chorus’ several versions because it is informed by her dawning suspicion that Ajax had not actually changed his mind about the suicide. And yet, it still includes the idea that Ajax has somehow changed his mind: ‘For I recognize that I have been cheated of the man [or: deceived by the man], and cast out of the favour in which I used to be held’, ἔγνωκα γὰρ δὴ φωτὸς ἠπατημένη | καὶ τῆς παλαιᾶς χάριτος ἐκβεβλημένη (807–8). The fact that Tecmessa has new and better information about Ajax’ unaltered intention to kill himself does not make her interpretation of his state of mind more definitive or less partial than those voiced by the Chorus. Although she may seem in the first of these two lines to rule on the question of whether Ajax was being intentionally deceptive during his speech, she is actually in no position to do that. She knows that she misinterpreted his words when she concluded that he no longer planned to kill himself, but it is not clear whether she believes she was deliberately misled; depending on how the genitive φωτός is construed, she could be saying she was deceived by Ajax or, more likely, that she was deceived about him.12 In either case, she is simply making an inference on the basis of her own disappointed expectations. In the next line, Tecmessa does read out of Ajax’ newly clarified intentions a change of attitude or feeling on his part towards her: she has been cast out of the χάρις (kharis)—the reciprocal relationship of affection and mutual benefit—­that she enjoyed in the past. This is an understandable but highly dubious reconstruction of Ajax’ state of mind, both in terms of the feelings she ascribes to him and in its envisioning of a locatable moment of change. It is contradicted by his own compelling words at the beginning of the deception speech about having changed so that he feels more, not less sympathy for Tecmessa. And it is very hard to know when this change of heart—­this casting from kharis—­should be imagined to have occurred. For Tecmessa, it is presumably identified with Ajax’ evident failure—­ despite what he says about his pity for her in the deception speech—­to be moved by her pointed appeal to kharis during their earlier agôn (522–4). But his immediate rejection of that appeal involved the affirmation of an existing resolution, coupled with a strong assertion that she should not expect to change his ­ 12  Jebb, following a scholiast who paraphrases her words as καταφρονήσας ἠπατήσέ με, takes φωτός as genitive of the agent as if with ὑπό, but this usage is not well paralleled. See Moorhouse 1982, 75–6, who argues for a genitive of separation such as occurs with ψευδόμαι, and the thorough discussion along the same lines at Gibert 1995, 124–5n32. This interpretation is also followed in their notes on line 807 and their translations by both Garvie (‘deceived in the man’) and Finglass (‘cheated of the man’).

52  Sheila Murnaghan character: ‘Your thoughts seem foolish to me, if you imagine that you can educate my character now’, μῶρά μοι δοκεῖς φρονεῖν, | εἰ τοὐμὸν ἦθος ἄρτι παιδεύειν νοεῖς (594–5). And it now seems clear to her that those heartening later words from which she took the impression that he had changed his mind after all must have been deceptive. Tecmessa is operating with familiar assumptions about the relationship between someone’s state of mind and his actions that make it hard for her to imagine that Ajax could feel sincere pity for her and still not do as she has asked.13 The former kharis that Tecmessa is alluding to is never on display at any point in the play or in her reminiscences of the event that immediately preceded it. If there really was a change, it would have had to have taken place before the play began. Tecmessa is evidently extrapolating from her own distinct moment of recognition that Ajax was not intending what she wanted him to be intending during the deception speech to construct a corresponding moment of rejection on his part. A discrete event occurring in her mind becomes on her interpretation a discrete event occurring in his, both expressed in the perfect tense. All of these passages show both the Chorus and Tecmessa engaging in proficient acts of mindreading. They attribute states of mind to Ajax on the basis of their selective uptake of cues in his speech and his actions, interpreting them in relation to their own expectations and their own hopes and fears. In this process of attribution, they rely particularly on the useful and flexible concept of a ‘change of mind’, working with a broad assumption that minds change from one definable state to another, which is a cherished principle of folk psychology and an effective expedient for reducing cognitive dissonance. This concept of ‘change of mind’ is so serviceable for integrating mixed signals and contradictory pieces of information, so useful a way of putting two things together, that it is embraced even when it strains against the mindreader’s sense of what can possibly be expected, as in the case of the Chorus, or even when it brings with it the perception of having been rejected and deceived by someone loved and depended on, as in the case of Tecmessa. From the perspective of a modern cognitive scientist, these characters are demonstrating a basic human skill, a mode of understanding and negotiating the world that is essential to successful functioning in interactions with others, something that people who have difficulty functioning, notably those with autism, lack. From the perspective of a reader or spectator of Sophocles’ play, these acts of mindreading are competent and psychologically plausible but also, of course, misreadings, or in the words of the messenger, ‘full of great foolishness’ (μωρίας πολλῆς πλέα, 745). Whatever is happening in Ajax’ mind when he gives that 13  And yet that counterintuitive but plausible position is expressly articulated by Ajax’ Homeric prototype Hector in his response to Andromache at Il. 6.440–6. On the comparison, see Easterling 1984, 6.

Reading the mind of Ajax  53 speech, none of these reconstructions can be said to capture it adequately or ac­cur­ate­ly. He has undergone some sort of change, as he himself announces, but not in the ways that any of his interpreters imagine: he has not decided not to kill himself; he has not given up his anger at the sons of Atreus; and he has not turned against Tecmessa. The idea of a ‘change of mind’, which is essential to their formulations, turns out to be muddled and inconsistent: a change of mind can be actively brought about or passively undergone; it represents a turning point in a forward-­progressing linear narrative, but can at the same time indicate a return to the past (as in the idea of having ‘gone back to forgetting his troubles’ (λαθίπονος πάλιν, 711), or may be hard to tie to any particular moment (as in the case of Ajax supposedly casting Tecmessa out of his favour). These readings do not add up to a coherent picture either of what is in Ajax’ mind or of how minds work in general.

3.4  Human minds in a divinely charged universe In his dramatic portrayals of human interactions, Sophocles bears witness to what modern cognitive science has systematically demonstrated: that mindreading is a definitive human activity. But for him that means revealing the inevitable limitations of mindreading as well as charting the many and various mechanisms by which it functions. Cognitively inclined critics have certainly observed that literary texts often record the pitfalls and failures of human attempts at attribution, and misinterpretation can be understood as itself representing the competent exercise of interpretative skills.14 But unlike both the subjects of contemporary psychological study and the characters of modern novels, Sophocles’ characters inhabit a universe dominated by baffling external forces, whose principles and purposes are opaque. The influence of powerful divine beings complicates the circumstances under which human beings try to understand each other, in ways that often lead to increased suffering and that are often pointed up for audiences through irony. Before the audience of Ajax is confronted with the ‘deception speech’, they witness, in the play’s opening episode, an unusually direct and explicit depiction of the gap between human perception and divine purposes, and of the obstacles to successful mindreading created by divine intervention. There Odysseus, a mortal known for superior intelligence, struggles to understand what is happening with Ajax as he tries to discover both his whereabouts and his state of mind through efforts that are repeatedly described in the language of tracking (5–8, 18–20, 31–3, 36). According to one theory, the capacity for mindreading first evolved because of its role in animal tracking, which requires awareness of the movements 14  Zunshine 2006, 22–5. As one cognitive scientist puts it, there is a ‘well-­attested tendency for egocentric errors during mindreading’ (Apperly 2011, 131).

54  Sheila Murnaghan and intentions of both human and non-­human individuals (Carruthers 2000, 272–3), but Odysseus’ tracking skills are insufficient for comprehending Ajax. They are briskly preempted by Athena, who easily explains what Odysseus has not been able to discern. As soon as she confirms that it was indeed Ajax who attacked the army’s livestock, Odysseus asks a question sparked by his own inability to read Ajax’ mind: ‘And to what purpose did he thus set in motion his incomprehensible act of violence [literally: dart out his irrational hand]?’ (καὶ πρὸς τί δυσλόγιστον ὧδ᾽ ᾖξεν χέρα;, 40). For Odysseus, Ajax’ action is duslogiston, impossible to square with ordinary reasoning, and so impossible to interpret. Athena responds by describing a mind equally shaped by inner and outer forces; Ajax’ assault on the livestock has been motivated both by his own anger over the awarding of Achilles’ armour and by the mistaken belief, which Athena herself has vindictively imposed on him, that the army’s animals are the Achaean leaders. Explaining how she kept him from killing the leaders, Athena stresses her own agency: ‘It was I who held him back from this irresistible joy, by casting in­toler­ able fantasies on his eyes’, ἐγὼ σφ’ ἀπείργω, δυσφόρους ἐπ’ ὀμμασι | γνώμας βαλούσα, τῆς ἀνηκέστου χαρᾶς (51–2). And as she revels in what she has accomplished, she makes a point of the fact that she has turned Ajax into someone who no longer matches his apparently established character: Do you see, Odysseus, how great is the power of the gods? Who, I ask you, could have been found more farsighted than this man, or better at doing what was right? ὁρᾷς, Ὀδυσσεῦ, τὴν θεῶν ἰσχὺν ὅση; τούτου τίς ἄν σοι τἀνδρὸς ἢ προνούστερος ἢ δρᾶν ἀμείνων ηὑρέθη τὰ καίρια; (118–20)

With her overt declarations of her own agency, Athena presents herself in terms that align with the elements of folk psychology that some cognitive scientists of religion have identified as the basis of religious belief. Theorists such as Stewart Guthrie, Scott Atran, and Pascal Boyer have linked the rise of religion to an evolutionarily beneficial tendency to attribute agency where it may not exist, which is applied to nonhuman beings.15 But here, as in those episodes in which humans read one another’s minds, Sophocles both shows the cogency and exposes the limits of ordinary cognitive procedures. It is clear from what Athena tells Odysseus that Ajax’ mind is not entirely his own, but the actual boundary between a sane, authentic version of himself and the effects of divinely imposed madness is highly elusive and much-­debated. Many of his thoughts and actions, including his rage at the award of Achilles’ armour, his desire for suicide, and his 15  Guthrie 1993, Barrett 2000, Boyer 2001, Atran 2002.

Reading the mind of Ajax  55 rashly over-­confident words when he left Salamis for Troy, and not just the mis­ iden­ti­fi­ca­tion of the livestock that Athena explicitly claims as her handwork, can be interpreted as signs of abnormality or irrationality, as they are by the various characters who try to understand him, especially Tecmessa and the Chorus. It is not only in relation to the deception speech that critics have struggled to understand the mind of Ajax.16 Athena’s direct appearance at the opening of Ajax is an unusual occurrence; Sophocles’ characters usually have to manage on their own, without the kind of authoritative information she provides, to make sense of circumstances shaped by divine powers. As they do so, they can only articulate their insights in human language, using concepts drawn from human experience. The clearest, most accessible formulations of divine purpose that Sophoclean characters are able to offer foreground the phenomenon of change, the fact that nothing stays the same over time.

3.5  Shifting mental states and the permanence of change The idea that divine influence on human lives is manifested by change, which pervades Sophocles’ plays, is registered with wonderful simplicity by Ismene in Oedipus at Colonus when she arrives to announce new oracles about Oedipus and sums up his circumstances by saying ‘Now the gods are raising you up;  before they destroyed you’ (νῦν γὰρ θεοί σ᾽ ὀρθοῦσι, πρόσθε δ᾽ ὤλλυσαν, 394). Awareness of change as a fundamental principle of the universe is expressed by characters who possess particular wisdom or a particular access to sacred mysteries, like Oedipus himself in the same play, who explains this to Theseus. Dearest child of Aegeus, to the gods alone old age comes not, nor death, but all-­powerful time confounds all other things. Earth’s strength decays, the body’s strength decays, trust dies, distrustfulness springs up, and the same spirit never stays constant between men who are friends or from city to city. (trans. Blundell) ὦ φίλτατ᾽ Αἰγέως παῖ, μόνοις οὐ γίγνεται θεοῖσι γῆρας οὐδὲ κατθανεῖν ποτε. τὰ δ᾽ ἄλλα συγχεῖ πάνθ᾽ ὁ παγκρατὴς χρόνος.

16  On the difficulty of delimiting Ajax’ madness, see Winnington-­Ingram 1980, 32–42, Goldhill 1986, 185–93, and Hesk 2003, 136–41.

56  Sheila Murnaghan φθίνει μὲν ἰσχὺς γῆς, φθίνει δὲ σώματος, θνῄσκει δὲ πίστις, βλαστάνει δ᾽ ἀπιστία, καὶ πνεῦμα ταὐτὸν οὔποτ᾽ οὔτ᾽ ἐν ἀνδράσιν φίλοις βέβηκεν οὔτε πρὸς πόλιν πόλει. (607–13)

As Oedipus’ opening words make clear, there is a paradox at work here. The operation of the gods may be apprehended by humans through change, as Ismene suggests, but they themselves are changeless; they alone are free from age and death. Change is a feature of human experience, which necessarily unfolds over time, and the idea of change is a human explanatory tool, an incommensurate, approximate way of getting at an unchanging reality. As a result, the most il­lu­ min­at­ing expressions in Sophocles are often paradoxes, formulations asserting that something is also its opposite, as in Ajax’ avowedly gnomic declaration in the deception speech that the gifts of enemies are not gifts: ‘But the saying of men is true: the gifts of men are evil gifts [literally: ‘non-­gifts’] and bring no benefits’ (ἀλλ᾽ ἔστ᾽ ἀληθὴς ἡ βροτῶν παροιμία, | ἐχθρῶν ἄδωρα δῶρα κοὐκ ὀνήσιμα, 664–5).17 The most visionary accounts of change are cyclical, as also in Ajax’ deception speech, in which he aligns his plans with a series of recurrent natural alterations: winter and summer, night and day, calm and storm, sleep and waking. For even the terrible and the most powerful yield to prerogatives. Winters in which one treads on snow make way for summer, rich in crops. Night’s dread sphere makes way for white-­horsed day to blaze forth its light. The blowing of terrible winds lulls to sleep the moaning sea. Like the rest, all-­powerful Sleep releases what he binds, and does not hold forever what he seizes. καὶ γὰρ τὰ δεινὰ καὶ τὰ καρτερώτατα τιμαῖς ὑπείκει· τοῦτο μὲν νιφοστιβεῖς χειμῶνες ἐκχωροῦσιν εὐκάρπῳ θέρει· ἐξίσταται δὲ νυκτὸς αἰανὴς κύκλος τῇ λευκοπώλῳ φέγγος ἡμέρᾳ φλέγειν· δεινῶν τ᾽ ἄημα πνευμάτων ἐκοίμισε στένοντα πόντον· ἐν δ᾽ ὁ παγκρατὴς Ὓπνος λύει πεδήσας, οὐδ᾽ ἀεὶ λαβὼν ἔχει. (669–76)

Cyclical change as a concept strains against a more familiar idea of change as a linear succession of discrete, incompatible, unforeseeable new conditions, both because cyclical change is constant and permanent and because any prevailing state predictably implies the emergence of its opposite; every step forward is also a return to the past. What Ajax is invoking in his speech is a kind of change that is 17  On the unity of opposites as an important point of contact between Sophocles’ world view and the mystically inflected thought of Heraclitus, see Kamerbeek 1948, Seaford 1994b, 282.

Reading the mind of Ajax  57 simultaneously a form of non-­change. As he himself points out, this invariable principle rules out any concept of human character as permanently settled such as he himself had invoked when he instructed Tecmessa not to hope to school his êthos. In terms of his own life as a human being, the cyclical change of the natural world provides a poignant contrast with the permanent change from life to death to which he, as he looks ahead to his suicide, assimilates it. But, at the same time, it also offers a paradigm for thinking about the change he is about to undergo as something more than an irreversible step into darkness, oblivion, and extinction, as something that transcends an ordinary human passage from life to death.18 As mentioned already, there is more being expressed in Ajax’ deception speech than we normally think of as being in a person’s mind at any particular moment. The speech channels a superhuman perspective that sometimes comes to heroic individuals at the point of death, as in the cases of Patroclus and Hector in the Iliad and Oedipus in Oedipus at Colonus, and that is one reason the speech does not map easily onto any idea of coherent character. When his auditors interpret that speech as a window on the mind of a character who has throughout the play been presented in contradictory terms, we see them relying on an essential, partial, and merely human cognitive resource: an interpretative schema that takes the form of change that goes in only one direction. This occurs even in the Chorus’ ecstatic lyrics that follow immediately on the speech and that, however misguided the belief that inspires them, seem to capture something of its vision. Even though they are paraphrasing Ajax, the Chorus completely miss what he is saying about the cyclical character of time. They evoke a single, one-­directional form of change, saying simply that time extinguishes all things: πάνθ᾽ ὁ μέγας χρόνος μαραίνει· (714). This limitation to their understanding is pointed up by the fact that our manuscripts include an interpolation that seeks to correct it, by adding in ‘and kindles them’ (τε καὶ φλέγει).19 This Chorus’ simpler, more straightforwardly human idea of change is the basis for an attribution that corresponds to their desires and works well for them in the moment, but is not fully adequate to the mind they are trying to read.

3.6  Conclusion: reading minds in Athenian tragedy The promise of cognitive approaches to literary works is bound up with the expectation that the episodes of interpretation depicted within texts will il­lu­min­ate the thought processes of the audiences who receive those texts. As Lisa 18  On the particular prominence of mutability as a theme in Aj., see March 1993, 22–4, and Cairns 2006. 19  These additional words are contained in all manuscripts and the Suda, but omitted from the extract quoted by Stobaeus (1.97.18) and deleted by most modern editors. There are no words in the strophe to provide the necessary metrical correspondence, so those who retain the phrase necessarily assume a lacuna at 701. On the Chorus’ ‘repression of [Ajax’] complex sense of change’ as dramatizing the way a search for intention results in misreading, see Goldhill 1986, 191–2.

58  Sheila Murnaghan Zunshine stresses, the cognitive experiences stimulated by literature are in­ev­it­ably shaped by historical conditions and answer to changing cultural agendas (2006, 153–5). Her own investigations of modern fiction foreground the pleasure and profit of exercising the essential human skill of mindreading as the source of literature’s value for contemporary readers. Another student of recent fiction, Mark Bracher (2013), finds a political purpose in literature’s depictions of people misreading one another’s minds, arguing that certain novels are designed to expose the deeply entrenched, reflexive misattributions that stand in the way of social change. Moving back in time to the early modern period, N.  R.  Helms (2012) investigates spectators’ possible responses to Orsino in Shakespeare’s Twelfth Night, a character whose presentation, like that of Ajax, involves perplexing elements of ambiguity and inconsistency. Helms argues that, in that context, the need for mindreading prompts spectators to use their cognitive faculties, especially those that involve conceptual blending, in ways that lead to better appreciation of that complexity of character that is more clearly a constituent of modern than of ancient drama; they become more attuned to ‘the power of shifting emotions’ and ‘the conflicted nature of human action’ (134). In the historical context of fifth-­century Athens, Sophocles was composing in a genre that tested humans’ capacity to understand and manage their own circumstances in a world shaped by divinities who were not only supremely powerful but also opaque. His sympathetic portraits of Tecmessa and the Chorus misreading the mind of Ajax can be seen as one response to the challenge of conveying a universe in which the knowable and the unknowable are intertwined.20 Under the momentary, mind-­concentrating pressures of extreme circumstances, in which they fear Ajax’ death as the loss of their only protector, these characters read Ajax’ speech through the conviction that their own fortunes depend on him changing his mind about his intended suicide. For the Chorus, his apparent reversal inspires an ecstatic experience of release from care; for Tecmessa, his continued resolution spells cruel indifference to her plight and to the obligations that spring from their reciprocal relationship. In one of the most suggestive critical discussions of the deception speech, Oliver Taplin (1979) proposes that from a longer perspective Ajax’ death may, however, be more beneficial to his dependents than they are able to envision, since the restoration of his honour that comes after it will save them from disaster. Ajax’ awareness, however imprecise, that the principle of inevitable change over time promises this future benefit for them is registered in the ambiguous language that they misread. Sophocles’ fifth-­century audience was certainly in a position to take a longer view of Ajax’ painful disgrace and suicide. They could see well beyond the ending of the play and measure the seeming catastrophe of Ajax’ death against their 20  For an account of Sophoclean language as variously deployed to give spectators a simultaneous experience of knowing and not knowing, see Budelmann 2000, especially 9–13.

Reading the mind of Ajax  59 knowledge of his potent and highly honoured role in the religious and political life of Athens, something that is only hinted at in Sophocles’ text. Their exposure to these other characters’ compelling but manifestly short-­sighted readings of Ajax’ challenging speech may have prompted a more expansive understanding of a death whose significance transcended typical human experiences of finality and loss. And it may have helped them to comprehend his enigmatic speech as the vehicle of a paradoxical, otherworldly vision that defies ordinary interpretation, one in which it is possible to recognize change as permanence, yielding to others as self-­assertion, self-­destruction as salvation, and deceptive words as expressions of truth.

Acknowledgements Versions of this chapter were delivered at the ‘Minds on Stage’ conference in Leiden in April 2016 and at a conference on Classics and Cognitive Science at NYU in October 2016. I am indebted to many audience members on both occasions for stimulating responses and questions and to the editors of this volume for their acute comments on an earlier draft.

4

Space for Deliberation Image Schemas, Metaphorical Reasoning, and the Dilemma of Pelasgus Michael Carroll

4.1 Introduction At the beginning of their landmark 1980 work Metaphors We Live By, George Lakoff and Mark Johnson draw a sharp distinction between their own position and the majority view (as they characterize it) that metaphor is a ‘device of the poetic imagination and the rhetorical flourish’ and ‘a matter of words rather than thought and action’.1 Their proposal, on the contrary, is that metaphor is, first and foremost, a conceptual tool that humans rely on to think about and make sense of the world. Idiomatic metaphors that generally escape our notice, such as ‘I’m at a crossroads in life’, reflect deeper sets of conceptual ‘mappings’ between domains of experience, in this case the domains of journeying and of life.2 In other words, we speak about life as if it were a journey because that is how we very often think about it: our knowledge of the source domain (i.e. the vehicle)—in the case of journeys, our knowledge that travellers tend to follow paths towards destinations, may meet obstacles, and so on—­has an influence on what kinds of inferences we draw as we reason about the target domain (i.e. the tenor). A crucial further step in Metaphors We Live By is the claim that such conceptual metaphors—­Lakoff and Johnson’s term for a set of mappings between two domains—­have their basis in human embodied experience. The constantly repeated experience at a young age of having to move from one location to another to achieve a particular goal leads to the formation of the basic or ‘primary’ metaphor PURPOSES ARE DESTINATIONS, and it is primary metaphors such as this that underlie more specific conceptual metaphors such as LIFE IS A JOURNEY.3

1  Lakoff and Johnson 1980, 1. 2  This example is drawn from Lakoff and Turner 1989, ch. 2. The related conceptual metaphor LOVE IS A JOURNEY features prominently in Metaphors We Live By. 3  For a helpfully clear discussion of primary metaphors and their formation, see Dancygier and Sweetser 2014, 21ff. I follow the convention of using capital letters when referring to concepts. Michael Carroll, Space for Deliberation: Image Schemas, Metaphorical Reasoning, and the Dilemma of Pelasgus In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0004

SPACE FOR DELIBERATION  61 It was not long before Lakoff turned his attention to literature, arguing, together with Mark Turner, that even the most sophisticated poetic metaphors can frequently be shown to depend on metaphors such as those identified in Metaphors We Live By.4 Since then many scholars have analysed literary texts in the light of the growing body of research on the relationship between figurative language, thought, and embodied experience.5 But even if it is worthwhile to know that works of literature rely on activating many of the same modes of thought that we use in day-­to-­day life, our sceptical literary scholar might still wonder whether these developments tell us much about all that sets metaphor in literature apart. Given that critics have long been aware that particular stylistic choices may have rich semantic implications and have developed a range of methods for analysing such effects, how pressing is the need for literature departments to familiarize themselves with this theoretical literature? What can cognitive linguistics (the field that emerged from the pioneering work of scholars like Lakoff and Johnson) tell us that we do not already know, even if we are likely to express ourselves in very different language?6 This is an important question, and the onus is on anyone who holds, as I do, that cognitive linguistics has much to add to our understanding of the workings of literary metaphor to explain exactly why more familiar approaches are in­cap­ able of yielding the same insights. The fourth section of this chapter will be devoted to a close reading of the metaphors used by the Argive king Pelasgus in Aeschylus’ Suppliants as he comes under increasing pressure from the Danaids—­ the group of sisters who make up the play’s Chorus—­to grant their request for asylum. These metaphors have already received a certain amount of critical attention, but I intend to show that, once we understand them as opening a window into the king’s thought process, the result is a richer and more interesting interpretation of the supplication scene as a whole. My aim, however, is not simply to offer a reading that overlaps with the theme of this volume. Rather, I shall be concerned from the start to address the worry that any insights that emerge from this and similar readings could be expressed without the encumbrance of extensive theoretical discussion and specialized terminology, and significant space will accordingly be set aside for clarifying in what respects my argument depends on the cognitive linguistic account of metaphor. I shall begin in the second section by laying out as starkly as I can the challenge facing those of us who believe that cognitive linguistics can be of value to the literary critic. Examining one suggestive metaphor used by the Danaids in the supplication scene will give me the opportunity to underline the interpretative power of the 4  Lakoff and Turner 1989. 5  For overviews of this body of research, see e.g. Freeman 2007, Semino and Steen 2008. A recent study of metaphor in Homer informed by Lakoff and Johnson’s framework is Zanker 2019 (the introduction contains references to other studies within classics that adopt such an approach). 6  For discussions of this worry, see e.g. Semino and Steen 2008, Fludernik 2011.

62  Michael Carroll methods already at the disposal of literary scholars. The question I am interested in is closely connected to a larger debate within cognitive literary studies on whether the tools of cognitive science have the potential to open up new literary interpretations, and in the second part of the second section I shall briefly consider some of the issues raised by that debate. The aim of the third section is to outline those tenets of the cognitive ­linguistic understanding that I shall be relying on in my analysis of Pelasgus’ metaphors. The key assumption underlying the familiar mode of analysis ­represented by my close reading in the second section is that the meaning of a word or expression can be separated into its denotation and the connotations it also happens to have. The account of metaphor developed within ‘relevance theory’ in recent decades is designed to avoid some of the problems associated with the denotation–­connotation distinction, but the picture that emerges nevertheless has much in common with the ‘mainstream’ view according to which a poetic metaphor typ­ic­al­ly presents us with a range of implications that need to be teased out. The cognitive linguistic understanding, by contrast, as I hope the reading in the fourth section will make clear, is the only model that does justice to what is experientially immediate in Pelasgus’ metaphors: it is the spatial and force-­ dynamic parameters of the pictures he elaborates, I shall argue, which give us the sense that we are following his thoughts and indeed in a position to reason alongside him. By tracing the development of the conceptualization of mental and physical space in Pelasgus’ metaphors—­and I shall be depending on the the­or­et­ ic­al notion of the ‘image schema’ here—­we find that a sudden shift in that conceptualization towards the end of the scene dramatizes a moment of epiphany which directly motivates Pelasgus’ decision to advocate the Danaids’ cause to the Argive people. While Pelasgus’ metaphors in this climactic passage have not previously been interpreted as reflecting such a marked change of understanding, I hope to show that the king’s epiphany brings to light paradoxes of a thoroughly Aeschylean character. The fifth section, finally, will contain reflections on what an analysis in terms of image schemas can reveal about the relationship between metaphor and dramatic character.

4.2  Literary criticism and cognitive poetics A glance at the scholarship on Aeschylus certainly does not leave one with the impression that the densely metaphorical language of the tragedies can be studied in isolation from the ideas that they explore; indeed, one of the most productive ways of reading Aeschylus to emerge in the last hundred years has seen scholars examine the thematic significance of the networks of recurring imagery which

SPACE FOR DELIBERATION  63 are such a striking feature of the plays.7 And though the details of individual metaphors have received comparatively little attention,8 that in itself does not suggest that the tools already at the disposal of classicists are unequal to the task, as an initial example from Aeschylus’ Suppliants may help to demonstrate. It is only when Pelasgus is satisfied that, despite their foreign appearance, the Danaids do indeed have a genealogical connection with Argos through their ancestor Io that he turns to the question of why they are occupying the shrine by the sea where he and his men have found them. In a tense stichomythic exchange (Supp. 335–47) he soon discovers that they are seeking asylum in an effort to escape the forceful advances of their cousins, the Aegyptids. Immediately realizing that granting their request would lead to war with the Aegyptids, the king expresses doubts about the justice of their cause, and the Danaids’ response is to appeal to the special force which the act of supplication acquires in a sacred setting:9 Respect the stern of the city, garlanded in this way. αἰδοῦ σὺ πρύμναν πόλεος ὧδ᾽ ἐστεμμένην. (345)

The Danaids’ metaphor evokes the ritual of garlanding which in Athens preceded the departure of the vessel that set out on the annual sacred mission to Delos.10 Few classicists nowadays would be happy to describe the metaphor here as merely a decorative way of referring to the shrine and the suppliant-­boughs which the Danaids have strewn on the altar. In the first place, the stern is the most im­port­ ant part of the ship,11 the place where the key decisions are made, and the Danaids want religious considerations, and in particular the sanctity of suppliants, to prevail over all other types of consideration. A further implication is that Pelasgus is himself the steersman and thus in a position to direct the course taken by the ship of state.12 But the evocation of the specific scenario of a ship setting off on a sacred mission adds further resonances, or dissonances if we prefer: in the case of the sacred mission to Delos, the priest of Apollo who carried out the garlanding was acting on behalf of the citizen body, while the Danaids have come from a distant land, and, rather than marking the beginning of a peaceful journey to another polis, their plea if successful is certain to embroil Argos in a war. 7 For an outline of this general approach and references to the most important studies, see Rutherford 2012, ch. 4. 8 One exception is Sansone 1975, which focuses on metaphors used to describe intellectual activity. 9  Here and in what follows I cite Sommerstein’s Loeb text (which is mostly consistent with West’s Teubner edition in the relevant passages). Translations are my own. 10  Friis Johansen and Whittle 1980, ad loc. On the garlanding ritual, see Rutherford 2013, 174. 11  Cf. Sommerstein 2019, ad loc. 12  On the imagery of the ship of state in early Greek literature (and the political significance of the figure of the helmsman in the archaic period), see Brock 2013, ch. 4.

64  Michael Carroll These are the sorts of ironies that critics tend to be very comfortable exploring, and the reference to the institution of the sacred mission also gives scholars with historicist leanings the opportunity to consider what this allusion might reveal about the play’s engagement with social and political issues of immediate concern to Aeschylus’ audience.13 A cognitive linguist could add that the ship of state metaphor depends on the same basic conceptualization of time in terms of space which underlies idiomatic expressions such as ‘I’m strolling through life’, with the ruler having the responsibility to ensure that the city follows the best possible path into the future, but the mapping between space and time was already implicit in the analysis given above and it is unclear what expressing the point in this way adds to our understanding of the meaning of this particular metaphor. The response of many scholars within cognitive literary studies to this objection is simply to deny that their primary aim is to produce new readings, and in support of this stance an appeal is often made to the distinction between her­men­ eut­ics and poetics.14 If the former, in Jonathan Culler’s words, ‘starts with forms and seeks to interpret them, to tell us what they really mean’, the latter ‘starts with attested meanings or effects and asks how they are achieved’.15 Thus according to Craig Hamilton, for example, literary scholars working within the cognitive paradigm ‘find it more fruitful to do cognitive poetics so as to engage with the epis­ tem­ol­ogy of our interpretative practices rather than add yet another interpretation of a text to the MLA bibliography.’16 The most convincing case for the value of literature departments can be made by showing that ‘the student of literature is really a student of the human mind.’17 There is no space to discuss this line of response in any detail here, but it might be wondered whether it is really possible to draw such a neat distinction between poetics and hermeneutics.18 If, by telling us more about how the human mind makes sense of literature, cognitive science can bring to light aspects of the fictional world that are there to be responded to even if our current tools are not up to the task of capturing them, is the potential not there for such insights to form the basis of new interpretations? Terence Cave is surely right to suggest that ‘[f]or mainstream literary specialists and readers . . . what is likely to matter most in the end is the openings offered by cognitively inflected reading’,19 and critics who believe that cognitive approaches can offer new perspectives on the meaning of literary works ought not to shrink from stating this emphatically. I am convinced that cognitive linguistics can enrich our understanding of the functioning of Greek tragic metaphor in its dramatic context for a number of ­reasons, but in what follows I wish to concentrate on just one of them: the 13  A recent historicist study of the play focusing on the theme of immigration is Bakewell 2013. 14  For an overview of this debate, see Vandaele and Brône 2009, 1–8. 15  Culler 1997, 61 (cited in Vandaele and Brône 2009, 3). 16  Hamilton 2002, 2. 17  Hamilton 2003, 64. 18  Cf. Stockwell 2006. 19  Cave 2016, 31.

SPACE FOR DELIBERATION  65 possibility raised by cognitive linguistics that metaphor can provide a form of access to the minds of the characters onstage and, in particular, to the activity of reasoning in which those figures are engaged.20 Although the surviving corpus of Greek tragedy is characterized by a rich and creative use of metaphor, as yet few scholars of Greek tragedy have analysed the language of the plays from this perspective.21 In Shakespeare studies the situation is rather different,22 and a number of scholars have argued that the astonishingly complex language which we again and again find Shakespearean characters using at moments of dramatic significance offers a precious glimpse of the train of thought underlying their words.23 It is not generally made clear, however, whether cognitive linguistics (and cognitive science more generally) has merely inspired the authors to adopt this perspective or whether such a position could not be adequately defended by more traditional means. Furthermore, much of this work takes insufficient account of the em­bodied, experiential dimension of metaphor as it emerges in the cognitive linguistic account,24 and it is precisely this aspect of Pelasgus’ metaphors on which my reading of the supplication scene will depend.

4.3  Two models of metaphor comprehension According to what I have characterized as the mainstream view of metaphor within literary studies, it is the connotations or associations introduced by a choice of metaphor that stimulate the type of rich interpretative engagement that interests the literary critic.25 Thus at the start of his book-length study of metaphor, Denis Donoghue can claim in reference to an example from T. S. Eliot’s The Waste Land that ‘[t]he point of the metaphor is to bring different associations, more dramatic connotations into the reader’s mind’,26 and we saw in the last section that the associations introduced by the Danaids’ nautical metaphor could form the

20  In this respect, metaphor can be seen as one of the means by which the plays both encourage and challenge the ‘mindreading’ skills of the spectators; on mindreading and Greek tragedy, see Budelmann and Easterling 2010, and the other two chapters in Part I of this volume. 21  But see e.g. Cairns 2017a. 22 Indeed Shakespeare has (unsurprisingly) received more attention from cognitively minded scholars than any other single author. For an overview of some of this literature, see Lyne 2011, 37ff. 23 Lyne 2011 is the most extensive study to adopt a cognitive perspective on the relationship between figurative language and dramatic character in Shakespeare. See also Read 2013. Davis 2007 explores the experience of ‘Shakespearean thinking’, but is not particularly interested in what this reveals about the character speaking the words. 24  Lyne 2011, 35 admits that the link between metaphor and embodied experience is not his primary concern. 25  Silk 1974, an influential study of early Greek poetic imagery, explicitly adheres to this model of meaning. A recent study that uses the language of associations and connotations throughout (and is directly influenced by Silk) is Matzner 2016. 26  Donoghue 2014, 2.

66  Michael Carroll basis of a familiar form of close reading that seeks to identify the broader thematic significance of a particular stylistic feature. While the vagueness and flexibility of the notion of a connotation is a large part of its appeal for the literary critic, a similar understanding of poetic metaphor also underlies the much more rigorous account put forward by Dan Sperber and Deirdre Wilson as part of their broader theory of communication known as ‘relevance theory’, and a brief overview of its key tenets will help to bring into focus those aspects of the cognitive linguistic theory on which my reading in the next section will depend.27 According to Sperber and Wilson, both hyperbolic and metaphorical utterances are understood through more or less the same interpretative process. Take the statement ‘The water is boiling’ as an example.28 If this is said by someone who has just stepped into a bath, only some information associated with the concept of boiling will be relevant, and Sperber and Wilson argue that a listener will automatically adjust the lexically encoded concept BOILING to an ad hoc concept BOILING* which supports those context-­specific implications alone (that the water is too hot for bathing, feels unpleasant on the skin, etc.). On the other hand, if a person shouts the same words in the middle of a storm at sea (a usage that would generally be considered metaphorical), the relevant information associated with BOILING will concern the water’s appearance and perhaps also the speaker’s perception of the storm as dangerous, but the result is again the formation of an ad hoc concept (BOILING**). From this perspective, then, metaphorical language emerges simply as a variety of ‘loose talk’ alongside approximate and hyperbolic utterances, and like them it is interpreted through a process of inference rather than by association of ideas.29 Sperber and Wilson refer to the implications which a speaker intends to communicate as ‘implicatures’, and they suggest that speakers (and writers) often choose to make use of metaphor in order to be able to convey a wide range of ‘weak’ implicatures. Such an effect is particularly characteristic of poetic language: ‘the wider the range of potential implicatures and the greater the hearer’s responsibility for constructing them, the more poetic the effect, the more creative the metaphor.’30 An important feature of this account, as Robyn Carston observes, is that it is ‘resolutely propositional’:31 each of the weak implicatures a metaphor may happen to convey can be expressed in the form of a proposition (‘The water is bubbling’, ‘The water is foaming’, etc.). In the case of extended metaphors, it seems somewhat implausible that the unfolding scenario is understood through the formation of one ad hoc concept after another, and Carston’s proposal is that in such cases a second, more reflective mode of processing comes into play: with the help of mental imaging, the scenario as a whole is ‘mentally held and 27  The key work in relevance theory is Sperber and Wilson 1995. 28  This example is drawn from Carston 2010, 303–5. 29  Sperber and Wilson 2008, 84–5. 30  Sperber and Wilson 1995, 236. 31  Carston 2010, 313.

SPACE FOR DELIBERATION  67 submitted for further reflective inferential processing’.32 The important point to note is that here too the resulting interpretation consists of many weak implicatures, and Carston in fact argues that this second mode calls for slower and more deliberate inferences than the process of ad hoc concept formation.33 At first glance an utterance such as ‘Robert is a bulldozer’ appears to pose no special challenge to the model I have just outlined; relevance theorists will argue that we again form an ad hoc concept, this time encompassing people who share certain relevant qualities with bulldozers. The problem is that the relevance-­ theory account is unable to explain why qualities such as forcefulness and unstoppability can have both a physical and a psychological application.34 Boiling water and agitated, storm-­tossed waves share certain objective similarities, but what information associated with the lexicalized concept BULLDOZER is also directly applicable to a person’s character? It seems, rather, that the physical notions of forcefulness, unstoppability, and so on, must already have been given a metaphorical interpretation before they can form the basis of an ad hoc concept whose denotation extends both to a person with a particular set of characteristics and to a type of machine.35 Cognitive linguists have a ready-­made answer to this problem: all of us constantly think about more abstract domains of experience such as personality, emotion, time, and morality in physical terms, and ‘image schemas’—mental representations that ‘arise directly from our interaction with and observation of the world’36—are what allow us to do so. Basic examples of such representations include PATH, CONTAINER, FORCE, and BALANCE, and the standard view within cognitive linguistics is that these are abstract conceptual structures that derive from our many specific experiences of having a certain spatial position with respect to our environment and of being affected by forces that may require us to exert a counterforce.37 The key hypothesis, then, is that these representations not only give structure and coherence to our perception and movement but also have a significant influence on how we think and reason about the world. I mentioned in the introduction that primary metaphors such as PURPOSES ARE DESTINATIONS underlie more specific conceptual metaphors such as LIFE IS A JOURNEY and that these primary metaphors are the result of experiential correlations that we encounter from a young age. Cognitive linguists claim that correlations between image schemas (such as SOURCE-­ PATH-­ GOAL in the case of PURPOSES ARE DESTINATIONS) and other aspects of experience are very often what give rise to 32  Carston 2010, 309. 33  Carston 2010, 310. 34  See e.g. Tendahl and Gibbs 2008, 1838–40, Ritchie 2013, 61–3 (a summary of his earlier articles on the problem). 35  Ritchie 2013, 61 refers to this as the ‘problem of circularity’. 36  Evans and Green 2006, 176. 37  This formulation is based on Dancygier and Sweetser 2014, 23.

68  Michael Carroll such primary metaphors.38 It is because we are already used to thinking of confrontations between people—­which may of course involve no actual movement or physical contact—­ in terms of image schemas such as MOMENTUM, SOURCE-­PATH-­GOAL, and OBSTACLE that we are immediately able to make sense of an utterance such as ‘Robert is a bulldozer’. A significant amount of empirical evidence has accumulated in recent decades which suggests that when we imagine a particular action or hear it described, we simulate (mostly without realizing it) the sensorimotor experiences associated with that action.39 And what has emerged is that metaphorical language rooted in image schemas such as SOURCE-­ PATH-­ GOAL is also understood through embodied simulations, even when the events or properties in question (such as Robert’s ability to bulldoze his way through any objection) are not physically ­possible. In the light of this evidence, the psycholinguist Raymond Gibbs has ­suggested that image schemas are not in fact abstract representations stored in long-­term memory, but are created in the moment as we construct simulations of embodied experience and thus serve as ‘temporary linkages between sensory experience and short-­ lived conceptualizations of both concrete events and abstract ideas.’40 What matters for our purposes is that, on this view, metaphorical meaning is neither purely propositional nor a matter of reflecting on an image in the manner of a detached bystander, but rather depends on our experiencing on some level what it would be like to engage in the actions associated with the image schema in question. And because our familiarity with such experiences means that we are automatically aware of the options open to us as we move along a path, say, or resist a force being exerted on us, even the most elaborate and stylized extended metaphors do not convey those possibilities and constraints as vague implications to be teased out: instead there is an immediacy and coherence to how we experience their spatial and force-­dynamic parameters. Cognitive linguists argue that metaphors very often require us to construct extremely complex simulations built on such image-­schematic foundations, and within cognitive linguistics Gilles Fauconnier and Mark Turner’s conceptual blending theory is currently the framework best suited to uncovering the cognitive mechanisms which allow conceptual structures drawn from diverse domains of experience to be integrated.41 One of blending theory’s particular strengths is that it draws attention to the fantastic properties in many metaphorical scenarios 38  Dancygier and Sweetser 2014, 24ff. 39  See e.g. Gibbs and Colston 1995; Gibbs 2006; Gibbs and Matlock 2008. On the contribution of embodied simulation to the experience of literature in particular, see Gibbs 2017. 40  Gibbs 2006, 113. 41  The theory is presented most comprehensively in Fauconnier and Turner 2002. For a clear overview, see Evans and Green 2006, ch. 12. Hanna Gołąb (this volume) makes much more extensive use of blending theory. Budelmann and LeVen 2014 argue for the benefits of analysing the poetry of Timotheus in light of the notion of emergent structure.

SPACE FOR DELIBERATION  69 (known in the literature as their ‘emergent structure’) which are the direct result of this process of integration. To tell someone that they are ‘digging their own grave’, for example, evokes the strange scene of a person unknowingly digging a grave that they will subsequently have to lie in. Fauconnier and Turner are able to offer a detailed account of the processes at work here, such as the common reconfiguration of the cause-­effect relationship (the digging of a grave is usually the effect of death rather than its cause),42 and they persuasively argue that such processes too have their origin in the patterns and correlations of embodied experience.43 It seems to me that blending theory is capable of greatly enriching our understanding of literary metaphor, and I hope that using it to explicate the emergent structure in Pelasgus’ extended metaphors will give some sense of its explanatory force. My primary interest in the next section, however, will be in the basic image-­ schematic structure of those metaphors, and the reason for this is that—­if the cognitive linguistic account is well-­founded—­this structure by itself offers a valuable insight into how Pelasgus is conceptualizing the situation he has suddenly found himself in. Both the mainstream view and the relevance-­theory account stress the capacity of poetic metaphor to pose a challenge to the reader by opening up a range of interpretative paths that might potentially be followed, but my reading of Pelasgus’ metaphors will be concerned with those aspects of his meaning that are processed immediately and thus offer us the sensation of having special access to the king’s train of thought as he deliberates before us. And what I  hope to demonstrate is that tracing this train of thought has the potential to make a significant difference to how we understand the dramatic trajectory of the scene as a whole.

4.4  Metaphor and the mind of Pelasgus The Danaids’ plea for asylum appears to leave Pelasgus with a straight choice between two unappealing alternatives comparable to the dilemmas of other Aeschylean figures like Agamemnon and Eteocles.44 As he declares in one of the short speeches he delivers in response to impassioned stanzas sung by the Chorus, he is at a loss (ἀμηχανῶ, Supp. 379) as to whether to help them or disregard their prayers, act or not act (δρᾶσαι τε μὴ δρᾶσαι τε, 380). Complicating all of this, however, is the question of who is to be responsible for the final decision. Soon after the passage of stichomythia in which the Danaids urge Pelasgus to respect

42  For this example, see Fauconnier and Turner 2002, 131–4. 43  Cf. Fauconnier and Turner 2002, ch. 5. 44  Important discussions of the supplication scene include Kitto 1961, 8ff., Lesky 1966, Burian 1974. See also Garvie 2006, 130ff., Papadopoulou 2011, ch. 6.

70  Michael Carroll the stern of the city, he tells them that it is not for him to decide the matter given that the safety of all of Argos is threatened (368–9). Some lines later he appears to indicate that he does have the authority to decide there and then (the scholarly consensus is that οὐδέ περ κρατῶν in line 399 means ‘even though I have the power’),45 but from the moment he announces his intention to entrust the decision to the Argives he never suggests that he has reversed this position, and when towards the end of the scene he comes down on the Danaids’ side, the result is simply a promise to support their cause in the assembly. It thus emerges that four distinct options were open to Pelasgus: to accept or reject the Chorus’ supplication himself, leave the decision entirely up to the people, or make up his own mind and then argue the case before the Argives. Critics have not always been alive to the delicacy with which the two strands of dilemma and political authority are fitted together,46 but, as we shall see, one benefit of this additional complexity is the subtlety with which it allows the king’s thoughts to develop. The pattern of alternating stanzas and short speeches is eventually broken when Pelasgus delivers a speech of eleven lines that begins with the first of three extended maritime metaphors in this final part of the scene.47 The king has already referred to the intractability of the dilemma and his own perplexity, but now his attention turns to the mental processes called for in such a situation: Certainly there is need of a deep, saving thought, for a sharp-­sighted eye not excessively dulled by wine to reach, like a diver, right down to the bottom . . . δεῖ τοι βαθείας φροντίδος σωτηρίου, δίκην κολυμβητῆρος εἰς βύθον μολεῖν δεδορκὸς ὄμμα μηδ’ ἄγαν ᾠνωμένον . . .  (407–9)

This is a metaphorical picture of remarkable complexity, and the idea of emergent structure proves extremely useful as a way of accounting for its bizarre features. The reference to wine calls to mind the blurring that objects undergo when viewed through water and which is analogous to the blurring that results from drunkenness.48 In this case, the eye’s sobriety is manifested in its capacity to 45  For arguments in favour of understanding the participle as concessive, see Papadopoulou 2011, 139n.20. 46  Cf. Burian 1974, 5. 47  Tarkow 1970 is unable to shed much light on the significance of the sequence of metaphors because his analysis remains at the level of their shared maritime setting. For other discussions of the metaphors (though not as a sequence), see e.g. Van Nes 1963, 40–1, 87–91, 163–4; Sansone 1975, 22–3, 26–7 (first two metaphors only). I shall consider the relationship between these extended metaphors and the Danaids’ ‘stern of the city’ metaphor towards the end of this section. 48  There is no need to relate the eye’s potential drunkenness to the sphere of diving in particular (even if, as Sommerstein 2019, ad 408–­9 notes, ‘the effect of alcohol on vision would be particularly acute ... under water’). One of the benefits of blending theory is that, unlike the traditional model of metaphor, it explicitly allows for the possibility that multiple source (i.e. vehicle) domains may contribute to a single image.

SPACE FOR DELIBERATION  71 counteract this property of the water by moving down through it. A change of location is often the only way to bring something invisible into view, and Pelasgus’ hope is that the eye’s dive beneath the surface will allow a solution to present itself. What is so strange about this eye, however, is its ability to move in­de­pend­ ent­ly and of its own accord,49 an emergent property which draws attention to an important feature of the deliberative process from Pelasgus’ perspective. Thought, unlike perception, allows a person to ‘see’ a possible outcome or opportunity without having to move closer to it in reality—­the eye of cognition can be detached from its owner’s present circumstances—­and Pelasgus is unwilling to commit to a particular course of action until he has left time for his deliberations to come across fresh considerations.50 The cognitive linguistic account suggests that, despite the complexity of this picture, a listener will immediately recreate its spatial parameters and grasp their significance. In image-­schematic terms, Pelasgus represents his cognitive faculties as free to move within a space that is ultimately bounded (only the vertical axis is relevant here), and this is because Pelasgus knows that a finite number of options will be available to him once he has had sufficient time to think. If we accept the relevance-­theory model, these spatial implications can only be understood as weak implicatures that contribute to the vagueness of meaning supposedly characteristic of poetic metaphor (and understanding them as connotations leads to much the same conclusion). According to the reading I am developing, by contrast, it is precisely because of the immediacy with which such spatial implications are conveyed that the sequence of maritime metaphors in this part of the play can shed light on the assumptions underlying Pelasgus’ deliberations. Four stanzas now follow from the Chorus as the pressure they are exerting gains in intensity, and at the start of Pelasgus’ next speech we discover that his deliberations have been continuing all the while:51 Indeed, I have considered, and my reflections have run aground here: provoking a great war against either this side or that is wholly unavoidable. The ship has been bolted tight, and remains at the shore, as it were, thanks to windlass cables; nowhere is there a route free from pain. καὶ δὴ πέφρασμαι, δεῦρο δ᾽ ἐξοκέλλεται· ἢ τοῖσιν ἢ τοῖς πόλεμον αἴρεσθαι μέγαν 49  For the eye as the actual diver descending—­rather than the diver’s eye, as e.g. Sommerstein 2019, ad 408–­9 understands it—­see e.g. Sansone 1975, 22ff. 50  Underlying Pelasgus’ choice of image here is the conceptual metaphor KNOWING IS SEEING, which again has its basis in embodied experience according to cognitive linguists; see Sweetser 1990, esp. ch. 2. 51  I follow Sommerstein 2008, 346 in accepting Friis Johansen’s tentative emendation πρὸς γῇ μένον instead of the manuscript reading προσηγμένον (though Sommerstein expresses scepticism about that solution in his recent commentary). If προσηγμένον could mean ‘having been hauled down to the sea’, then we would still have the basic picture of a ship moored and ready to sail.

72  Michael Carroll πᾶσ᾽ ἔστ᾽ ἀνάγκη· καὶ γεγόμφωται σκάφος, στρέβλαισι ναυτικαῖσιν ὣς πρὸς γῇ μένον, ἄνευ δὲ λύπης οὐδαμοῦ καταστροφή. (438–42)

In their last stanza the Danaids had urged Pelasgus to consider (φράσαι, 437) the fact that, whatever his decision, there would be consequences for his own children and house. Pelasgus assures them that he has been thinking through the various options and their consequences, and his thoughts have alighted on the same dilemma: war is inevitable, whether with the Aegyptids or the gods whose statues dominate the shrine. The single word ἐξοκέλλεται is sufficient to evoke the scenario of the conclusion of a sea journey, and it is significant that, as Friis Johansen and Whittle note, ‘ἐξοκέλλειν denotes enforced “running aground” as opposed to voluntary “landing” (κέλλειν)’.52 The emphatic πέφρασμαι suggests that Pelasgus initially felt in control of his deliberations, and because purposive action is typically conceptualized in terms of the embodied, image-­schematic experience of self-­propelled motion,53 we can immediately infer that Pelasgus was engaged in a mental sea journey that has come to a halt.54 Now that all the available options have been explored Pelasgus realizes that the horns of the dilemma represent the only possible outcomes, and while in the earlier metaphor this could be economically represented by the diver’s reaching the bottom without finding a solution, here a scenario is evoked according to whose emergent structure the ship of thought is forced to come to land once it is clear that the alternative routes which it has explored are not practicable in reality. Pelasgus then proceeds to elaborate a new scenario. The text is uncertain, but Friis Johansen and Whittle’s proposal that Pelasgus is describing the final two stages before a newly built ship sets sail for the first time is persuasive.55 The last bolts have been hammered into the hull and the ship has been hauled down to the sea, with only cables to hold it in place. The deliberative preliminaries are finished, Pelasgus appears to be saying, and it only remains to deliver a verdict on which of the two wars ought to be waged (whether by making an executive decision himself or by representing the Danaids’ cause); whichever course is chosen, the launch is to mark the start of an ill-­fated journey.56 In the space of a few lines, then, two separate sea journeys are evoked, and the idea seems to be that only once Pelasgus has mentally traversed the options open to him is the boat of action ready to be launched. What this means in image-­schematic terms is that, on the level of action, Pelasgus has remained 52  Friis Johansen and Whittle 1980, ad loc. 53  See e.g. Dancygier and Sweetser 2014, 44–5. 54  Pace Sansone 1975, 27, who understands the ship to have run aground after being ‘tempest-­ tossed on a sea of turmoil’. 55  Friis Johansen and Whittle 1980, ad 440–1. 56  Friis Johansen and Whittle 1980, ad 442 argue that καταστροφή ought to be understood as ‘direction’, in which case line 442 can be taken to mark a further development of the nautical metaphor.

SPACE FOR DELIBERATION  73 stationary—­whether at the surface of the sea in the earlier metaphor or in this case on the shore where the boat is being made ready—­while his thoughts have been in motion. As the speech continues Pelasgus becomes preoccupied by the thought that, because bloodshed cannot be undone, it must be avoided at all costs, and he evidently derives new fixity of purpose from this line of reasoning: Truly I have stepped aside from this quarrel: I wish to be ignorant rather than apprised of troubles. May all turn out well, contrary to what I judge likely. ἦ κάρτα νείκους τοῦδ᾽ ἐγὼ παροίχομαι· θέλω δ᾽ ἄιδρις μᾶλλον ἢ σοφὸς κακῶν εἶναι· γένοιτο δ᾽ εὖ παρὰ γνώμην ἐμήν. (452–4)

It seems to me that Pelasgus is doing more in these lines than simply ‘reaffirming as final his previous refusal to get personally involved’.57 When Pelasgus earlier announced his intention to involve the Argives in the decision-­making process, he did not make it clear what his own role in the process was to be. Now for the first time the king appears to decide on a course of action for himself, and it turns out to be the third of the options outlined above, a refusal to have any influence on the decision. The meaning of παροίχομαι (‘I have stepped aside’) has an image-­ schematic basis just as much as the earlier extended metaphors, and we can immediately grasp that, by moving to one side, Pelasgus is refusing to commit to either of the two routes into the future which, in the terms of the previous metaphor, the boat of action would have to take: his decision is to stay on the shore.58 The Danaids appear to sense this new resolve in Pelasgus, and in the passage of stichomythia that follows they threaten to commit suicide by hanging themselves from the statues in the shrine. So effective is the threat of religious pollution that by the midway point of his next speech Pelasgus has decided to represent the Danaids’ cause, but the opening lines show that this change of heart coincides with a revised understanding of the situation: Truly these circumstances are in many ways hard to wrestle with. A mass of troubles is rushing in my direction like a river; this is a bottomless sea of ruin, far from easily navigable, that I have ventured into, and nowhere is there a haven from troubles. †καὶ μὴν πολλαχῇ† γε δυσπάλαιστα πράγματα, κακῶν δὲ πλῆθος ποταμὸς ὣς ἐπέρχεται· ἄτης δ᾽ ἄβυσσον πέλαγος οὐ μάλ᾽ εὔπορον τόδ᾽ εἰσβέβηκα, κοὐδαμοῦ λιμὴν κακῶν. (468–71) 57  Friis ­Johansen and Whittle 1980, ad 452. 58  It is therefore misleading to suggest that this speech, like the previous, is ‘aporetic’ (Sommerstein 2019, 156), even if Pelasgus’ choice of the third option is short-­lived.

74  Michael Carroll The description of the circumstances as hard to wrestle with at first seems to suggest that the difficulty of choosing between the horns of the dilemma is continuing to preoccupy Pelasgus. We quickly discover, however, that his concern is now with the disastrous consequences flooding towards him whatever he does. If previously he believed that it was possible to step aside and avoid becoming involved in launching the response to the Danaids’ supplication, it is now clear that the troubles of which he wished to be ignorant are making directly for him. There follows what is in image-­schematic terms a remarkable discontinuity, made all the more striking by the closeness with which the two metaphorical scen­arios correspond: in both the disaster is represented as a body of water, but now instead of being a static onlooker Pelasgus is on board a moving ship.59 And what this combination of metaphors points to, I suggest, is the king’s new-­found understanding of the paradoxical nature of the situation. The Danaids’ threat has made it clear that he is directly involved whether he wishes to be or not: if his worry before was that making the wrong decision would cause civil unrest, he can now see that allowing the suppliants to kill themselves in the shrine would be just as much his responsibility. And while the flood metaphor is a perfectly valid way of conceptualizing his position as passive recipient of evils, it is not the only one available. Pelasgus’ intention was to withdraw from the dispute, but instead he has unwittingly arrived in a situation—­the emphatic active first person (εἰσβέβηκα) sharpens the paradox—­which has placed him not only on board a ship but, we can presume, at the helm. Again, the notion of emergent structure is helpful here: in a similar way to those who dig their own grave, Pelasgus was not aware that he had moved from dry land to the open sea. Yet despite the fact that on one level he is a victim of circumstance, as captain he must nevertheless take responsibility. As long as Pelasgus continued to think of himself as having the breathing space that comes with immobility, free to move around in thought without committing himself to any outcome, all of the four options outlined above seemed open to him. A helmsman at sea, by contrast, has no choice but to direct the ship along a particular course, and given the immediacy of the threat of pollution compared to the prospect of war, it is no surprise that this new understanding induces Pelasgus to side with the Danaids. That Pelasgus, as captain, has the power to direct the city’s course was of course implicit in the Danaids’ appeal for him to respect the stern of the city, and now with their threat of suicide they have succeeded in forcing their understanding upon him. Although there is still the debate in the assembly to come, if an experienced helmsman suggests a course of action it will be a brave crew who disagree, and in the next scene we hear that the Argives have voted unanimously to grant the Danaids asylum. The cognitive linguistic account of metaphor is the only one that can support the reading I have offered in this section. In contrast both to the mainstream and 59  On the ‘functional parallelism’ of the two clauses, see Friis Johansen and Whittle 1980, ad 469–­71.

SPACE FOR DELIBERATION  75 to the relevance-­theory approach, it suggests that there is an immediacy and coherence to how a listener will process the topography of these metaphors—­ their image-­schematic structure is experienced as a gestalt, cognitive linguists would say60—and this means that the listener is automatically aware of the possibilities of physical movement in the scenarios that the king evokes. And an audience’s instinct as they listen will be to understand Pelasgus to be experiencing the image-­schematic entailments of these scenarios with equal immediacy. Rather than translating his thoughts into metaphorical terms, Pelasgus is using metaphor as a means to think through the situation and the place of his deliberations in it. Thus the fact that we are able to think through the consequences of his choices of metaphor alongside him helps to narrow the gap separating us from the figure onstage. Just like Pelasgus, we grasp without needing to be consciously aware of it that a person on solid ground has the option of standing to one side. And it is thanks to the immediacy with which we sense the different entailments of the sea-­voyage picture (with immobility no longer an option) that we can appreciate that it represents a revised understanding of Pelasgus’ situation and share that moment of insight with him. .

4.5  Embodied experience and dramatic character The supplication scene in Suppliants offers by far the most protracted instance of decision-­making in Aeschylus. It has often been claimed, however, that Pelasgus lacks a distinctive character, merely exemplifying the type of ‘democratic’ king, and that Aeschylus is not concerned to give us a sense of the internal struggle that we might expect from a person in such circumstances.61 Rosenmeyer’s judgement is in this respect representative:62 Aeschylus prefers not to articulate the process of reflection. There is no reasoned speech in which Pelasgus explores the pros and cons in detail. . . . Choice is not dramatized; it is reported, just as other decisions are reported in Aeschylean drama. The agony of the decision making is hinted at, but largely displaced and in effect neutralized by imagery, silence, and the intrusion of the chorus.

Lying behind Rosenmeyer’s claim that Pelasgus’ imagery draws our attention away from the decision-­making process taking place within, it seems to me, is the conception of metaphor as a means of veiling one’s meaning, of vaguely hinting at

60  Evans and Green 2006, 185. 61  See e.g. Kitto 1961, 8, Seidensticker 2009, 220. For Lesky 1966, 80, by contrast, ‘Aeschylus elab­or­ates the psychological development of the characters more fully than his successors’. 62  Rosenmeyer 1982, 305.

76  Michael Carroll many things, which underlies both the mainstream position and the relevance-­ theory view of poetic metaphor. In the last section, on the other hand, Pelasgus’ metaphors were analysed not simply as stimulating the listener to think in new ways—­as both the mainstream and relevance-­theory accounts also maintain—­ but as directly reflecting the speaker’s own conceptualization of the topic at issue. It was this isomorphism between conceptualization and verbal expression that allowed us to trace the progress of Pelasgus’ thinking.63 One of the reasons I have chosen this scene as my case study is that Pelasgus’ metaphors do not merely dramatize his train of thought: in two cases his thoughts also form the subject matter of the metaphors, with the result that, as we picture his deliberations as moving down towards the seabed or across the main, the very notion of a thought process is brought into focus. It is interesting, however, that Pelasgus’ deliberations are not brought to an end, as he had hoped, by the discovery of a third, painless option. Rather than finding a different route into the future, the king acquires a fresh perspective on the context within which his own ­deliberations have been taking place. From the start, the Danaids have had the option of bringing pollution on Argos through suicide, and this new understanding makes Pelasgus realize that his mental wanderings have not been taking place on land, as it were, but at sea. My sense is that, as well as finding Pelasgus rather one-­dimensional as a character, critics have felt there to be a degree of awkwardness in the way the two themes of dilemma and political authority are combined in this scene. After all the build-­up, it can seem disappointing that Pelasgus agrees merely to represent the Danaids’ cause. But what I hope the train of thought we have been reconstructing brings to light is the intricacy with which these two strands are inter­ woven. It is not simply the consideration that bloodshed is irreversible that motivates the king’s preliminary decision to avoid coming down on either side: Pelasgus’ cognitive faculties have not succeeded in finding a way out of the dilemma, and the analytical reflexivity he manifests in the first two extended metaphors is consistent with his more general concern to establish where the limits of his authority ought to lie in such a crisis. As we have seen, the king’s self-­ consciousness culminates in the realization that circumstances have placed him in a position of responsibility that makes neutrality impossible, and this newly refined understanding of the relationship between ruler and city leads directly to the decision to side with the Danaids announced some lines later. 63  The suggestion that there is an experiential immediacy to some of what is conveyed by Pelasgus’ metaphors should not be taken to imply that the metaphors do not also repay sustained attention and reflection (as indeed we would expect with poetic language of such richness). For the analyst who decides to focus on the spatial parameters of a poetic metaphor or sequence of metaphors, moreover, the task of articulating their significance is far from straightforward, as the length of the discussion in the previous section illustrates. The relevance-­theory account, it seems to me, is particularly well placed to do justice to this careful, reflective mode of interpretative engagement. I am grateful to the editors for helping me to clarify this point.

SPACE FOR DELIBERATION  77 Even if little is revealed of Pelasgus’ ‘true’ character in this scene, then, the risk in focusing on the generic aspects of his portrayal is that we overlook the complex and dynamic cues that allow the audience to construct a consciousness behind the words.64 What we are faced with in the case of Pelasgus, I would argue, is an individual who, far from placidly fulfilling the role of consensus-­seeking ruler, is wrestling with the challenges of his position: that he wishes to avoid incurring the people’s blame perhaps signifies that his behaviour is partly motivated by self-­ interest, and when he expresses the desire to remain ignorant of troubles we may already wonder how realistic a stance this is for a ruler to adopt. Yet he also manifests a capacity for self-­analysis well suited to the position of democratically minded king: opening up one’s decision-­making process for inspection is one way of avoiding the aloofness characteristic of the tyrant. It is by juxtaposing two metaphors with contrasting image-­schematic entailments that Pelasgus articulates his eventual epiphany, and the paradoxes which this combination of metaphors allows him to touch on are of a sort that will be very familiar to readers of Aeschylus. Without realizing what was happening and without having acted improperly, the king has arrived in a situation where the nature of his position requires him to initiate a course of action likely to have grave consequences for himself, and according to most scholarly reconstructions of the rest of the trilogy, Pelasgus will in fact lose his life in the conflict with the Aegyptids.65 Tragic scholars are well aware both that, within a genre that already stands out for its figurative density, Aeschylus’ tragedies are especially rich in their use of metaphor, and that the plays explore issues of knowledge, agency, and justice with remarkable intensity of focus. The great value of the cognitive linguistic theory of metaphor in this connection is that it can offer fresh insights into the interdependence of these two dimensions of Aeschylean drama. I am confident that paying attention to the image-­schematic basis of Aeschylus’ metaphors, as well as untangling the embodied conceptual processes that contribute to their fantastic properties (something that I have only been able to do in passing here), will reveal the extent to which this language involves the audience in the same search for meaning and understanding in which the dramatis personae are engaged. That is not to say that the cognitive linguistic framework necessarily has all the answers, and a number of scholars have argued that blending theory, for example, has much to gain from engaging more closely with relevance theory.66 As I suggested in the first section, moreover, there may also prove to be many cases where the denotation–­connotation opposition is perfectly adequate to the job at hand. Finally, although my interest has largely been in the perspective of the literary critic, it is perhaps not entirely immodest to suggest that, given a sufficient sample size, close readings such as the one advanced in the fourth section might offer 64  Cf. Budelmann and Easterling 2010. 66  See e.g. Tendahl and Gibbs 2008.

65  See Garvie 2006, 198–9.

78  Michael Carroll metaphor theorists a fresh way to test the virtues and limitations of their models. For a relevance theorist who finds my interpretation of the supplication scene plausible, for example, one response might be to dispute my assertion that the spatial implications of Pelasgus’ metaphors could not be communicated as strong implicatures and then to flesh out the theory to show why this is so. If we can accept that the theoretical literature on figurative language is capable of stimulating greater sensitivity to the workings of literary metaphor, we ought not to discount the possibility that the relative persuasiveness of the interpretations inspired by particular theoretical models might serve as one source of evidence when assessing their merits.

PART II

C O GN IT IV E WOR K B Y C HA R AC T E R S

5

Attribution and Antigone Ruth Scodel

5.1 Introduction This chapter considers the exchanges between Antigone and Ismene in Antigone through attribution theory, the branch of social psychology that studies how ­people attribute causes to behaviours and outcomes, both their own and others’. Although attribution is closely related to Theory of Mind, as a field it belongs in social rather than cognitive psychology. Theory of Mind has had a large influence on literary studies in the last decade, but attribution theory less so, though it has a place in rhetoric.1 Yet literary characters do not just guess at what others think and feel, but make judgements about their motives. Traditional attribution theory studies how and why people locate causes. Do they identify external factors as causes or rather the dispositions or mental states of individuals? And with what respective frequencies? To some extent, attribution is universal, since there is a limited repertory of possible causes and reasons for action (especially when we categorize all potentially different kinds of super­nat­ ural interference as external causes). However, there are cultural differences in attributional practices. Cultures differ in how likely people are to provide disposi­ tional rather than situational explanations; individualist cultures are more prone to underestimate situational factors (Fundamental Attribution Error).2 However, when experimental subjects from different cultures are asked to make inferences specifically about dispositions on the basis of behaviour, there is no difference in their response.3 People tend to overestimate how fully behaviour corresponds to disposition—­this is Correspondence Bias. In tragedy, however, it  is not a mistake to expect behaviour to reveal dispositions, since the genre ­typ­ic­al­ly represents actions that reveal the nature of the characters. 1  For Theory of Mind, see Zunshine 2006; for attribution, Pollard-­Gott 1993, Palmer 2010 (group attributions), Smith 2015. Grethlein 2015 objects to the central role Zunshine and Palmer give Theory of Mind in fiction generally through a discussion of Heliodorus’ plot-­centred Aethiopica, but see the responses immediately following this article by Palmer, M. Fludernik, and M.-L. Ryan. Nünning 2014, 231–2 acknowledges that many fictions do not invite much Theory of Mind from their audiences, but have other cognitive benefits. See also Van Emde Boas in this volume. 2  E.g. Miller 1984. 3  Krull et al. 1999. However, Americans make more dispositional inferences than Japanese when the diagnostic evidence is weak: Miyamoto and Kitayama 2002.

Ruth Scodel, Attribution and Antigone In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0005

82  Ruth Scodel Attribution in tragedy is often explicit, since characters constantly present their own motives and their inferences about the motives of others. Attribution theory could therefore be applied to a text such as Sophocles’ Antigone. The investigation could be looking for the folk psychology of ancient Greeks, trying to analyse how Greek tendencies in attribution are like or unlike those of the scholar’s own cul­ ture. This chapter, however, will apply attribution theory to the exchanges between Antigone and Ismene primarily to gain insight into the text itself, although we have to remember that cultural difference may be at work. It is useful because some significant qualities of these exchanges become clearer when they are viewed from this perspective. First, Antigone presents herself as extremely rational, almost as a homo economicus. She claims that she has acted after weighing the advantages and disadvantages of possible actions and choosing the most advantageous.4 Other characters, however, see her as wildly irrational, and see this irrationality as dispositional. Second, Antigone’s interpretation of Ismene’s motives for refusing to join her is more complex than appears at first. Third, Antigone’s attributions depend on Antigone’s very low valuation of her own life in the first part of the play.

5.2  Malle’s attribution theory This chapter will use the particular attribution theory of Bertram Malle, whose work emphasizes how we first consider whether an action is intentional or not, and then seek reasons or causes, respectively.5 Malle rejects the model used in standard attribution theory, which relies on co-­ variation—­ that is, how and whether a behaviour changes as a possible cause changes.6 The co-­variation model is appropriate where the observer has information about the agent’s behav­ iour at other times and in other situations, and can evaluate how possible causes are correlated with their behavioural effects. Its classic form considers three vari­ ables, Consensus, Distinctiveness, and Consistency. If we ask why Jane fell asleep in a lecture, we will consider, as far as we have the information, whether others also dozed? (Consensus) Does Jane often fall asleep in lectures? (Distinctiveness) Does Jane always fall asleep when this speaker is lecturing, even at a different time of day? (Consistency).7 Combining these, we can decide that the speaker is boring or that the room was overheated (situational cause), or that Jane uniquely finds the speaker boring or is prone to falling asleep (internal cause). Malle, how­ ever, convincingly argues that the co-­variation model works in the lab better than it does in everyday life, since the experimenter provides the relevant information, and that more generally it works best for unintentional behaviour. 4  ‘economic man’: Hashimzade, Black, and Myles 2017; Thaler 2000. 5  Malle 2004. 6  See Kelley 1967. Where the information needed to evaluate co-­variation is lacking, Kelley 1972b proposed that observers use causal schemata, that is, their structures of belief about what causes are ­sufficient or necessary for a particular effect. 7  A good introduction to ‘traditional’ attribution theory is Försterling 2001.

Attribution and Antigone  83 In any case, co-­variation would have little to say about how Sophoclean charac­ ters understand each other’s motives, since tragedy is primarily interested in the dispositions of its characters. There is no doubt, for either characters or the external audience, that, despite the extraordinary situation, both Antigone’s and Ismene’s responses to Creon’s edict flow from their basic dispositions. Creon, indeed, says that his actions after assuming power will permit accurate attributions about his character (Ant. 175–7). Co-­variation therefore is not required—­the spectator will usually assume that significant actions reflect the character’s traits. Since drama lacks an omniscient narrator who can provide direct information about motivations, attribution is continual. Characters attribute motives to them­ selves and to others, and members of the audience make their own attributions. Spectators will also attribute motives to explain why characters make the attribu­ tions that they do. Using attribution theory can at least help clarify what interpreters do when they make pronouncements about characters’ motives. Budelmann, Maguire, and Teasdale have found that spectators of dramatic scenes in which characters’ motives are ambiguous are likelier to select a single motivation than they are when considering the motives of a real politician making a historically momen­ tous decision. It is also notable that different members of an audience will confi­ dently select different and incompatible interpretations of characters’ motives. For example, spectators of Clytaemnestra’s response to the report of Orestes’ death in Aeschylus’ Libation Bearers saw either genuine or faked grief, rather than a complex, mixed emotion.8 An earlier experiment showed that the slant of a par­ ticular performance had less influence than anyone would have expected—­that a performance that made Creon more sympathetic did not change audience responses.9 Both experiments invite the inference that individual preferences and preconceptions powerfully influence how a spectator makes attributions and evaluations in drama. Unless we posit a homogeneous Athenian audience, we should assume that the spectators of Antigone would weight different explicit attributions differently and would differ also in their own attributions about the motives for these attributions.10 In Malle’s system, the person making an attribution first decides whether behaviour is unintentional or intentional.11 Unintentional behaviour has causes, but intentional behaviour has reasons. If we have judged a behaviour uninten­ tional, we decide whether to locate the cause outside the person or in the person, 8  The experiment described in Budelmann, Maguire, and Teasdale 2016. This response is a form of ‘discounting’: where more than one cause of a behaviour is possible, ‘the role of a given cause in producing a given effect is discounted if other plausible causes are also present’ (Kelley 1972a, 8–14 (quota­tion p. 8). The participants presumably found the causes incompatible and so discounted one entirely. 9  Budelmann, Maguire, and Teasdale 2013. 10  So Sourvinou-­Inwood 1989 argues that the original audience would have seen Antigone as a ‘bad woman’, assuming that all Athenians would have made the same attributions. 11  Malle 2004, 5–27 (critique of earlier attribution theory), especially 15–19 on co-­variation, 87–90 on intentionality.

84  Ruth Scodel and then, if it is in the person, whether it represents a trait. We may say that someone failed an exam because the teacher was incompetent (external), or because a personal crisis made him unable to study (internal), or because he’s lazy (trait) or stupid (trait). If we are explaining an intentional act, we often define reasons as Beliefs, Desires, or Valuings. Beliefs and Desires are relatively easy to understand. It is worth noting that agents offer more Belief reasons than observers do. Actions based on Beliefs appear rational (although the beliefs themselves may lack any rational basis), and people wish to seem rational.  Also, observers find Beliefs harder to infer than Desires. Valuings are our affects about particular actions or experiences, and while they are often hard to distinguish from Beliefs or Desires, they are worth differentiating. Unlike Beliefs, they cannot be false, and unlike Desires, they can refer to the past (‘she told everybody how great the show was because she enjoyed it so much’). There is no difference between agent and observer attributions of Valuing.  The terms for Belief or Desire are often not explicit: If José has not already tried and failed to borrow money, there may be no real attributional difference between ‘Jane lent José the money because nobody else would’ and ‘Jane lent José the money because she believed that nobody else would’. Sometimes we assume that the agent’s reasons need no explanation, and the question that we ask is what made the action possible. In such a case, we may cite Enabling Factors. If someone asks why Antigone gave burial rites to her brother, a respondent might decide that the question was not ‘what were Antigone’s reasons for trying to bury her brother?’ but ‘how was it possible for Antigone to perform rites for her brother?’ Such a respondent might suggest an Enabling Factor, such as help from the gods. ‘Causal History of Reason’ is the most interesting category for tragedy. Here the explainer, instead of giving the actual reasons, goes to some information that pro­ vides background for the agent’s reasons. These are not the same as Enabling Factors, which do not lie behind the agent’s reasons. If someone asks why the chicken crossed the road, ‘there was no fence to stop her’ is an Enabling Factor, while ‘chickens are never satisfied where they are’ is a Causal History of Reason. A Causal History of Reason can be internal (‘Why did she run over to hold the door for the man in the wheelchair?’ ‘She’s a kind person’) or external (‘Why did she drink so much at that party?’ ‘It’s part of the culture of that office’). An internal Causal History may appear at the same time (‘And she is a conformist’). A par­ ticular reason may be implied, or the explainer may give both a Reason and a Causal History. ‘It’s part of the office culture’ may appear with ‘She didn’t want to stand out’ (Desire) or ‘She thought that it would hurt her standing in the office not to join in’ (Belief). Causal Histories are not conscious reasons. ‘She is a kind person’ is not quite the same as ‘She wants to be a kind person’. In trait Causal History of Reasons, moral judgements go beyond the action to the person. Sometimes we will provide both a reason and an Enabling Factor, or more than one reason—­as long as the various attributions do not contradict each other, we

Attribution and Antigone  85 can provide several. The choice of attribution, even when it is entirely sincere and not, for example, an excuse, may depend on the situation. So, in Antigone, there is no inconsistency in Antigone’s greater emphasis on divine law when arguing with Creon than when arguing with Ismene. Ismene, like Antigone, has personal ­reasons to want to bury Polynices, but when Antigone confronts Creon, she emphasizes those reasons that are universally applicable. This diagram represents in simplified form the logical structure of attribution, where each arrow indicates a stage in the process of forming and expressing an attribution. If the behaviour is unintentional, the observer cites a cause or causes, which may lie in the situation or the person. If the cause is located in the person, it may or not be a trait. If the observer decides that a behaviour is intentional, that observer then chooses to cite a Reason, an Enabling Factor, or a Causal History of Reason.12 People can and do, however, offer more than one, so the diagram illus­ trates only the hierarchy of attributions, and not the actual sequence of cognitive activity. Unintentional Cause Situation

Person Trait

Intentional

Enabling Factor

Situation

Person Belief Trait

Reason

Causal History of Reason

Desire Valuing

Situation

Person Trait

12  Malle does not allow for Enabling Factors for unintentional behaviour. Subsidiary causes look very much like Enabling Factors, but they can be seen simply as subsidiary causes, since they do not need to be distinguished from reasons. ‘Why did he drive off the road?’ could be answered ‘He is a bad driver and it’s a tricky curve’ (trait and situation).

86  Ruth Scodel In persuading others, we offer reasons to act a particular way, and someone who accepts a particular argument also frequently accepts the reason, which is then incorporated into a self-­attribution. Implicitly or explicitly, attributions are central to persuasion. In political contexts, they are often Causal Histories of Reason that groups share. In persuasive contexts, modern people often seek to pressure others by making negative attributions. In such a context, the speaker who cries ‘You’re just a coward!’ jumps over all the reasons both interlocutors might have for performing or not performing a particular action. Tragic charac­ ters regularly try to persuade themselves or others to perform or avoid particular actions by categorizing these actions as indicative of traits of character. Sometimes  a prospective attribution is completely explicit, but sometimes it is not, and it may or may not refer to the process of future attribution by others: there is often no sharp distinction, especially for Greeks, between ‘if you perform this action, you will be X’ and ‘if you perform this action, people will say that you are X’. Both types of attribution appear retrospectively as well, of course. In their prospective form, however, these arguments in tragedy often become tests or challenges. The rhetorical use of attributions is much more complicated than a psy­cho­ logic­al experiment can show. Impression management (not wanting to sound like a bigot, for example) is only the first factor that turns attributions into something other than honest attempts at evaluation. Speakers in a confrontation, as they fail to persuade their interlocutors and become angry, become more extreme in their attributions (a particular form of the tendency of angry speakers to hyperbole noted by Aristotle, Rhetoric 1413a). This is frequent, as we all know, in tragic stichomythia. At some point such attributions may not be ‘serious’ arguments about the reasons or dispositions of others. Though hyberbolic, these typically still reflect an actual attribution, but distort it.

5.3  Antigone: cause or reason? In Antigone, the basic distinction between intentional and unintentional behav­ iour is not unambiguous. Ismene (99), Creon (561–2), and the Chorus (603) all characterize Antigone as ἄνους, although Antigone repeatedly presents what appear to be strikingly rational arguments. The elders of the chorus imply that supernatural forces have interfered with Antigone’s mental condition (603).13 These characters thereby refuse to engage with Antigone’s reasons and deny her

13  The elders of the chorus also offer divine anger as a cause of Antigone’s fate at 856, and she acknowledges the earlier misfortunes of the family (857–66), but without specifying either an origin for divine resentment or a mechanism by which it could have operated. See Sewell-­ Rutter 2007, 115–19.

Attribution and Antigone  87 meaningful intentionality. Because Antigone’s arguments differ from each other, many modern interpreters have also treated Antigone’s stated reasons as ration­ alizations of intuitive action, although different but compatible arguments are appropriate for different interlocutors.14 Antigone herself considers the possibility that her suffering is not independent of her family’s history (857–66), but never allows that she has not acted for ­reasons. This conforms to the experimental results that agents seek to appear rational. By claiming that Antigone suffers from ἄνοια, other characters put her behaviour into the unintentional category. Their attributions are not all the same, however, since it is clear that Creon and Chorus believe that Antigone’s irrationality is a trait, but Ismene never implies that Antigone is irrational by nature or has always been irrational. Since Chorus and Creon believe that Antigone’s irrationality is inherited from her parents, it is not surprising that her sister does not share a belief that would make her also inherently inclined to irrationality.

5.4  Antigone’s and Ismene’s attributions Antigone is full of attributed reasons. Some are completely straightforward. For example, near the opening of the play, Antigone explains to Ismene why she has called her outside: I knew it well, and I sent for you to come out of the courtyard gates so that you alone would hear ἤδη καλῶς, καί σ’ ἐκτὸς αὐλείων πυλῶν τοῦδ’ οὕνεκ’ ἐξέπεμπον, ὡς μόνη κλύοις (18–19)

There is a Belief reason (Antigone believed that Ismene did not know) and a Desire reason (Antigone wanted to speak to Ismene privately). Things quickly become more complicated. It is worth noting that Antigone says nothing about Creon’s reasons for issuing his decree. She sarcastically refers to him as ‘the noble Creon’ (τὸν ἀγαθὸν Κρέοντα) at 31. This is probably an implicit trait Causal History of Reason, although we cannot know whether it is based only on this edict. Antigone implies not only that Creon is not ἀγαθός and that an ἀγαθός would not have issued the edict, but that he believes that he is 14  Hester 1971, 14 catalogues twelve earlier scholars who hold this view. An excellent defence of taking Antigone’s reasons seriously is Foley 2001, 181–3. Foley cites nobody who denies Antigone’s rationality from later than 1964—feminism presumably made interpreters aware that sexism lay behind the interpretative strategy.

88  Ruth Scodel ἀγαθός, and that this belief in his own nobility lies behind the edict (she is at least partly right about that). So attributions are not entirely simple. The attributions that Antigone and Ismene exchange deserve some closer analysis. Antigone concludes her speech about the edict with a prospective trait Causal History of Reason that is an unmis­ takable test aimed at Ismene: So, you see how it is, and you will soon show whether you are noble in your nature or bad though from good ancestry. οὕτως ἔχει σοι ταῦτα, καὶ δείξεις τάχα εἴτ’ εὐγενὴς πέφυκας εἴτ’ ἐσθλῶν κακή. (38–9)

Ismene does not immediately understand what she is being challenged to do. Even when we learn what Antigone wants, these very general moral terms do not satisfactorily indicate what specific traits Ismene’s decision will reveal. Εὐγενής in Sophocles is used of a varying set of traits associated with aristocratic origins, including honesty (not at issue here), political loyalty, and endurance. Here, though, I think the audience is likely to have assumed that courage would be the trait most at issue. An εὐγενής, as Ajax says, must live nobly or die nobly (Aj. 479–80). Antigone performs another implicit attribution at 45–6, when she responds to Ismene’s ‘Do you intend to bury him, though it is forbidden for the city?’ (ἦ γὰρ νοεῖς θάπτειν σφ’, ἀπόρρητον πόλει)15 with ‘My brother at any rate, and [I will also bury] yours, if you are not willing’ (τὸν γοῦν ἐμόν, καὶ τὸν σόν, ἢν σὺ μὴ θέλῃς, | ἀδελφόν).16 The peculiar formulation gives each sister a distinct obligation, and by stressing Ismene’s relationship to Polynices, and treating her refusal to help bury him as evidence of unwillingness, Antigone implies that Ismene does not value the relationship enough. Antigone began by evoking the family’s history: Do you know what of the evils deriving from Oedipus Zeus does not fulfil against us, who still live? For there is nothing either grievous or disastrous or shameful or dishonouring that I have not seen among your troubles and mine. ἆρ’ οἶσθ’ ὅ τι Ζεὺς τῶν ἀπ’ Οἰδίπου κακῶν— ὁποῖον οὐχὶ νῷν ἔτι ζώσαιν τελεῖ;

15  Or possibly ‘by the city’—while Antigone is likely to maintain a sharp distinction between Creon and the larger community, Ismene might not. 16  This translation follows Radt 1971 and Jebb 1900.

Attribution and Antigone  89 οὐδὲν γὰρ οὔτ’ ἀλγεινὸν οὔτ’ ἀτήριον17 οὔτ’ αἰσχρὸν οὔτ’ ἄτιμόν ἐσθ’, ὁποῖον οὐ τῶν σῶν τε κἀμῶν οὐκ ὄπωπ’ ἐγὼ κακῶν. (2–6)

Antigone regards her family as noble, yet she describes their misfortunes as bringing shame and dishonour. Implicitly, she is already preparing an argument that she will make explicitly to Creon. On her calculation, her life, given the family history, is not precious. Ismene’s argument at 49–68 recites family history for the opposite purpose.18 She concludes: And now we two who alone are left—­consider how most wretchedly we will perish, if in violence to the law we violate the vote or powers of rulers. νῦν δ’ αὖ μόνα δὴ νὼ λελειμμένα σκόπει ὅσῳ κάκιστ’ ὀλούμεθ’, εἰ νόμου βίᾳ ψῆφον τυράννων ἢ κράτη παρέξιμεν. (58–60)

Ismene has not recognized or not responded to Antigone’s subtext, so for her the likelihood that the sisters will die if they attempt to bury their brother is a reason not to make the attempt. That they are the only remaining members of the family (Labdacids) implies a further Valuing argument that implicitly answers Antigone’s valuing: the sisters care, or should care, about the continuation of the family line. There may be a further implication that the family history itself makes it unlikely that they would escape the bad outcome. Ismene’s categorization of what Antigone proposes implies also that the action is wrong, because she evidently regards Creon’s rule as legitimate. Violating his command would be acting against law or custom, νόμου βίᾳ, and would transgress not just his decision, ψῆφον, but his power (κράτη). The word νόμος offers an implicit Belief reason for Ismene’s refusal, but this seems to be a second-­order nomos. Ismene does not need to regard Creon’s decree as just in itself to grant it a kind of legitimacy, because she apparently believes that it is nomos to obey what the ruling power commands. A Belief reason becomes explicit when she argues that as women, they should not fight against men:

17  This is Brunck’s conjecture for the incomprehensible ἄτης ἄτερ. 18 So Winnington-­Ingram 1980, 128: ‘the awful past (1ff., 49ff.), which dictates submission to Ismene, only strengthens the obstinate resolve of Antigone’.

90  Ruth Scodel We must consider that, that we were born women, not intended to fight against men. ἀλλ’ ἐννοεῖν χρὴ τοῦτο μὲν γυναῖχ’ ὅτι ἔφυμεν, ὡς πρὸς ἄνδρας οὐ μαχουμένα· (61–2)

‘I refuse to do this because I believe that women are not suited to fighting with men’. Ismene does not need to mark her belief as an opinion, because she also believes that this understanding of human nature is a recognized truth. She is aware that Antigone may not share this belief, but she states it not as a belief, but as a fact. The sisters then disagree over whether not burying Polynices constitutes dis­ honouring the dead or the gods: So I, asking those beneath the earth to be forgiving, since I am compelled in this, will obey those in office. For acting uselessly has no sense. ἐγὼ μὲν οὖν αἰτοῦσα τοὺς ὑπὸ χθονὸς ξύγγνοιαν ἴσχειν, ὡς βιάζομαι τάδε, τοῖς ἐν τέλει βεβῶσι πείσομαι. τὸ γὰρ περισσὰ πράσσειν οὐκ ἔχει νοῦν οὐδένα. (65–8)

Ismene implicitly agrees with Antigone that burying Polynices would be the right thing to do, but argues that the dead will forgive her because she is under situ­ ation­al constraint. But she then moves to a new argument by designating the pro­ posed action as περισσὰ πράσσειν (acting not just uselessly, but excessively and so wrongly). This argument is subordinate to the claim of constraint. It appears to be a Belief reason based on a common Greek moral assumption. Ismene does not just refuse the action for herself, but cites a general principle. She invokes the same principle when she says at 92 ‘It is not fitting to hunt the impossible in the first place’ (ἀρχὴν δὲ θηρᾶν οὐ πρέπει τἀμήχανα). Under this principle, action that has no chance of achieving its goal is futile and senseless, as Ismene says. It is also morally suspect, because a person who refuses to acknowledge that an action is excessive or who seeks the impossible may be refusing to recognize the limits on mortals.19 This attribution can easily go farther to claim that the agent is disposi­ tionally inclined to ignore these limits (trait Causal History of Reason). Ismene does not, however, make such an attribution. While she accuses Antigone of 19  Lefèvre 2001, 100. The theme is closely related to the ‘Near and Far’; see Hubbard 1985, 11, Pfeijffer 1999, 389–91.

Attribution and Antigone  91 being ‘in love with impossibilities’ (ἀλλ’ ἀμηχάνων ἐρᾷς, 90), which implies that Antigone’s desires are abnormal, she does not attribute this to Antigone’s disposition. Antigone responds with an argument that does not deny the constraint, but treats it as insufficient as a deterrent: It is noble for me to die doing this. I will lie dear with him, with one dear … If you think it best, hold what is honoured by the gods in dishonour. Is:  I do not make these dishonoured, but I was born to be incapable of acting against the citizens. καλόν μοι τοῦτο ποιούσῃ θανεῖν. φίλη μετ’ αὐτοῦ κείσομαι, φίλου μέτα, … σὺ δ’ εἰ δοκεῖ τὰ τῶν θεῶν ἔντιμ’ ἀτιμάσασ’ ἔχε. Ισ.   ἐγὼ μὲν οὐκ ἄτιμα ποιοῦμαι, τὸ δὲ βίᾳ πολιτῶν δρᾶν ἔφυν ἀμήχανος.  (72–3, 76–9)

It may be worth thinking about this point, since the most famous claim of attribu­ tion theory has been that people tend to undervalue the situational constraints on others but not on themselves. Here the situational constraint is identical for the sisters, and they both recognize it. Antigone, however, claims, first, that such a death would be καλόν, which recalls her earlier claim about εὐγένεια and would seem to point to an argument based on heroic values. Then she expands not on the need for courage, as a spectator might expect, but on the value of inclusion in the family and Ismene’s alleged failure to honour the gods enough. The disagreement could be understood as a disagreement based on Theory of Mind. The sisters both imagine how the gods and perhaps the dead will judge them, but they evaluate differently how the gods and the dead think and so how they will attribute motive to the sisters. In Ismene’s view, the gods will recognize that she has Valuing and Desire reasons to bury Polynices and has no Desire to dishonour them, but the effort is too dangerous and impossible. At this point, it is impossible for her, because of her nature (she will claim only a few lines later, at 90, that success is impossible for anyone), which is unable to act against the citizen body. There is a lot to unpack here. It is, obviously, a trait Causal History of Reason—­it is not her nature to do this. The Causal History of Reason and the reason are very close here: Ismene’s reason for not  burying Polynices is that it is forbidden, and it is her disposition not to commit forbidden acts. Just as she earl­ier referred to nomos, here she identifies

92  Ruth Scodel Creon’s decree with the will of the citizens and so attributes general support of the edict to them. Ismene inserts a request to the dead to forgive her because of the circumstances (65–6, cited above). She thereby attributes some kind of consciousness to the dead, though it is not clear how far the implications of her request should be taken or to what extent the attribution of such a power of judgement to them makes a difference. If both characters have a fully coherent belief that the dead are conscious and aware of the actions of their living survivors, Antigone’s arguments would have considerably more power, but nobody in the play really articulates what kind of consciousness the dead have. Antigone responds to Ismene’s request to the dead to pardon her by claiming that, because she will be dead longer than she’ll be alive (as indeed all humans will be), it is more important to please the dead than the living (‘since there is more time during which I should please those below than those here’, ἐπεὶ πλείων χρόνος | ὃν δεῖ μ’ ἀρέσκειν τοῖς κάτω τῶν ἐνθάδε, 74–5). Yet, in describing her imagined future affectionate relationship with her dead brother, she uses the verb κείσομαι, ‘I shall lie’—κεῖμαι is standard in fu­ner­ eal epigrams and points rather to the immobility of a corpse than to a conscious afterlife.20 It does not deny the possibility of an afterlife, but it does not imply it, either. Although she speaks as if ‘pleasing’ were simply her own reason, she is twisting Ismene’s argument of constraint and she reiterates the theme in 89, ‘but I know I am pleasing those whom I must especially please’ (ἀλλ’ οἶδ’ ἀρέσκουσ’ οἷς μάλισθ’ ἁδεῖν με χρή), again implying that Ismene also has made a choice of whom to ‘please’. However, Ismene has not expressed any Desire to please Creon or the Thebans, and passive obedience to Creon’s decree is unlikely to provide any positive pleasure to anyone. Antigone responds to Ismene, however, by describing Ismene’s arguments—­ that is, her reasons—­as ‘excuses’, as not her real reasons but something she puts forward in order to conceal her real reasons: You could make these excuses. But I will go to pile a tomb over my dearest brother. σὺ μὲν τάδ’ ἂν προὔχοι’· ἐγὼ δὲ δὴ τάφον χώσουσ’ ἀδελφῷ φιλτάτῳ πορεύσομαι. (80–1)

By calling her explicit reasons ‘excuses’, Antigone implies that she can perceive the motives Ismene has not acknowledged, but these are likely to be Causal History of Reasons—­not the reasons themselves, but the character dispositions that underlie these reasons. Ismene, after all, has already given a general Causal History of 20  Griffith 2006, following Johnson 1997, attributes incestuous desire for Polynices on the basis especially of this line.

Attribution and Antigone  93 Reason that leads to all her reasons: she is not capable of acting against the citizen body. φιλτάτῳ, ‘dearest’, implies one such Causal History of Reason. By defining the choice as a decision about who should be ‘pleased’, Antigone has made it easy for herself to reiterate from 45 her Causal History of Reason claim that Ismene does not adequately love her dead brother. Antigone does not, however, say anything that implies cowardice in Ismene. This is remarkable. Electra, in her play, tells her sister ‘I envy you your sense, but I hate you for your cowardice’ (ζηλῶ σε τοῦ νοῦ, τῆς δε δειλίας στυγῶ, Soph. El. 1027). Ismene’s recital of the family’s ill fortune, concluding with the claim that the sisters will die ‘most miserably’ (Ant. 59), shows that she is willing to present a desire not to die as an explicit reason. When Antigone insists on going out to bury Polynices, Ismene exclaims, ‘how much I fear for you!’ (82), and Antigone tells her not to fear on Antigone’s behalf (83). Ismene then says that she will not reveal the deed to anyone (84–5). She is openly concerned for her sister’s survival. The audience can make an attribution: a desire to live is clearly a reason for refus­ ing to participate. However, the spectator should not infer that she is acting ­simply out of fear since she also offers a Belief Reason for her refusal: she believes the action is inappropriate and that success is impossible. It is impossible to know what level of risk she would find acceptable if she thought that the attempt was reasonable. Still, it would be easy for Antigone to jump from ‘you don’t want to die’ to ‘you are afraid’ to the trait Causal History of Reason ‘you are a coward’, and she does not. Both εὐγενής and καλόν point this way, but she does not expand on these implications. She does not make this move in the later scene between them, either. At 543 she says ‘I do not care for a friend who loves in words’ (λόγοις δ’ ἐγὼ φιλοῦσαν οὐ στέργω φίλην), and at 549 she accuses Ismene of caring about Creon, ‘for it’s him you care about’ (τοῦδε γὰρ σὺ κηδεμών), instead of herself or Polynices. She thereby reiterates with variation her earlier accusation that Ismene did not love her brother properly. This is an example of the phenomenon I discussed earlier, the tendency for rhetorical attributions to become extreme. No spectator is likely to imagine that Antigone believes that Ismene loves Creon more than herself or more than Polynices. She probably does believe that Ismene does not love her as much as she should, and her frustration drives her to a cruel hyperbole.21

5.5  Antigone’s valuing But although she refers explicitly to Ismene’s desire to live, when Ismene asks how she can benefit Antigone, she answers ‘Save yourself; I do not begrudge your 21  Even Müller 1967 on 549–51, for whom Antigone can do no wrong, admits that this is ‘streng genommen’ untrue.

94  Ruth Scodel escape’ (σῶσον σεαυτήν· οὐ φθονῶ δ’ ὑπεκφυγεῖν, 553). Antigone’s famous ‘For you chose to live, but I to die’ (σὺ μὲν γὰρ εἵλου ζῆν, ἐγὼ δὲ κατθανεῖν, 555) presents the choice simply as a choice, a preference that each had. Earlier, though, she pre­ sented a cost-­benefit analysis to Creon: I knew that I would die, how not?— even if you had not made your proclamation. If I will die before my time, I say that is a gain. Whoever like me lives amid many evils, how does he not win a gain by dying? So for me, to meet this fate is a grief of no account. But if I had endured that the child of my mother were unburied after his death, I would have been grieving at that. θανουμένη γὰρ ἐξῄδη, τί δ’ οὔ; κεἰ μὴ σὺ προὐκήρυξας. εἰ δὲ τοῦ χρόνου πρόσθεν θανοῦμαι, κέρδος αὔτ’ ἐγὼ λέγω. ὅστις γὰρ ἐν πολλοῖσιν ὡς ἐγὼ κακοῖς ζῇ, πῶς ὅδ’ οὐχὶ κατθανὼν κέρδος φέρει; οὕτως ἔμοιγε τοῦδε τοῦ μόρου τυχεῖν παρ’ οὐδὲν ἄλγος· ἀλλ’ ἄν, εἰ τὸν ἐξ ἐμῆς μητρὸς θανόντ’ ἄθαπτον ἠνεσχόμην, κείνοις ἂν ἤλγουν· (460–8)

This is an Enabling Factor, presented in the language of bookkeeping. Antigone does not calculate the gain of early death as so great that she would have sought it in any case. Leaving her brother unburied would cause her pain that would make her life even worse, but she does not state the argument here that way. She argues more generally that death is preferable for someone who lives amid evils, and that would be a reasonable account of her situation even before Creon’s edict. Her low estimate of the worth of her own life made it much easier for her to decide to perform the burial, but it was not her reason for doing it. She presents this, not surprisingly, as a rational evaluation that would have general application to any­ one in a similar situation. The spectator may suspect, however, that a Valuing may actually precede the calculation. Antigone does not have a strong positive affect about remaining alive. It is impossible to distinguish reasoning from rationaliza­ tion. Not having a positive affect towards life, Antigone can argue that she has no grounds to have such an affect, but we know from experience, as Sophocles’ audi­ ence knew, that many people who have many reasons to be miserable nonetheless continue to value their lives. The following interchange, though, takes a different direction. Since the choice has been presented as based on a rational consideration of advantages and

Attribution and Antigone  95 disadvantages based on valuing, it can be treated as entirely about reasons. Antigone agrees that different groups would approve the reasoning of each sister (‘You seemed to think well to some, I to others’, 557). Ismene’s response is very odd: ‘and truly, the mistake of both is equal’ (καὶ μὴν ἴση νῷν ἐστιν ἡ ’ξαμαρτία, 558). Jebb and Kamerbeek both take this as meaning that Ismene’s sympathy for Antigone/acquiescence in the deed makes her equally guilty (a position even Creon in the end does not take), while Griffith thinks that it continues the point about differing judgements—­each is wrong in the judgements of different people. This makes better sense of καὶ μήν. I would suggest yet another view, however. While Ismene’s use of ἐξαμαρτία could imply ‘in Creon’s view’, a spectator could legitimately infer that Ismene still believes that Antigone’s action was an error, but that both she and her sister have failed to value adequately different aspects of the complex situation. From Ismene’s perspective, the calculation of value depends in part on Antigone’s decision and situation. While her life was worth saving as long as Antigone was alive, she no longer thinks that her own life is worth living with­ out Antigone. So her decision not to help Antigone, once it became clear that Antigone was absolutely determined, was also a mistake. This understanding of Ismene’s point allows it to motivate Antigone’s response. Antigone rejects this argument by giving a Causal History of Reason for her own action that is not a trait, but a process: Take heart. You are alive, but my soul has been dead for a long time, so that it can benefit the dead. θάρσει. σὺ μὲν ζῇς, ἡ δ’ ἐμὴ ψυχὴ πάλαι τέθνηκεν, ὥστε τοῖς θανοῦσιν ὠφελεῖν. (559–60)

Antigone has not yet literally died, of course, so her metaphor requires in­ter­pret­ ation. Commentators have tended to assimilate this statement to a Desire self-­ attribution, so that her ‘death’ is the outcome of her decision to help the dead, not a reason for it. The Greek can be read more precisely if she is offering a Causal History or even an Enabling Factor: because her grief caused her to feel as if she were already dead, she had no impulse to self-­preservation. This death of the psy­ che consists primarily in a lack of concern for her future life, the low value she has placed on it, which made it natural that she would help the dead without concern for self-­preservation.22 She has lost what is popularly called the will to live. Her mental state does not resemble clinical depression, for Antigone is certainly not suffering from emotional numbness or lack of motivation. This happened ‘long ago’, and if she is giving a Causal History we do not need to make πάλαι refer only 22 Jebb 1900 and Kamerbeek 1978 both give the ὥστε-­clause consecutive force. Griffith 1999 glosses ‘decided to devote her life (ψυχή) to the dead’.

96  Ruth Scodel to the time since the edict.23 It is impossible to know at what point in the ca­lam­itous history of the family she feels that she ‘died’. Some unusual state of mind is ­certainly understandable in someone who has experienced as much trauma as Antigone, but she says nothing to imply that this mental death was to be expected, or was in any way better than remaining committed to life. Nothing indicates that it was intentional. It is simply something that happened to her but did not happen to Ismene. Such a view of her own mental state means that Ismene’s mental state—­being psychologically alive—­would seem to be normative. Antigone is not saying that Ismene was right, and she is certainly not saying that she was wrong, but she may be acknowledging that Ismene’s reasoning was not wrong for Ismene, because their Causal Histories were very different. If Ismene’s refusal to defy the edict was based mainly on her valuing of her life, and not on a failure to value her family—­the reason Antigone has attributed to her—­Antigone does not need to condemn her. The entire passage shows a remarkably swift development from harsh criticism of her sister to a kind of understanding. The importance of valu­ ing helps explain why courage and cowardice are not raised. If Antigone lacks will-­to-­live, it is not especially brave for her to risk death. If Antigone’s action, with its poor chance of any kind of success, would only seem wise for a person who did not value her life, Ismene is rational rather than cowardly. It is probably Antigone’s exceptionally low valuing of her own life that makes others judge her ἄνους, even though she presents this as not just rational, but the product of calculation. It is, of course, an old difficulty in understanding the play that the Antigone of the latter part of the play laments her death (e.g. 806–16). I am not going to dis­ cuss this issue here, but only to point very briefly at some ways in which the care­ ful study of character attributions may help us understand the problem. Antigone attributes to herself two different kinds of reason for her action. She believes that the gods want all corpses to be buried, a belief she expresses most vehemently in her speech to Creon about the ‘unwritten laws’ (450–7). Her Belief reason, how­ ever, might not require that she, personally, risk her life for the principle. She chooses to die for it based on two Valuings: a very high valuing of family ties, and a very low valuing of her life. She revises her valuation of her life in her lament. She does not change either her Belief or her Valuing of her brother.

5.6 Conclusion Attribution theory helps clarify why Antigone’s view of herself and her motives is so different from the way others see her. Individual Valuings are premises on the 23  Jebb 1900 and Kamerbeek 1978 take πάλαι as ‘since the edict’. Ellendt 1965 indeed gives many examples of ‘subjective’ πάλαι: Sophoclean characters often use the word of mental states that are long-­standing only in contrast to an immediate situation.

Attribution and Antigone  97 basis of which an actor makes decisions, just as in economics the same good has different utility for different people. Because Antigone places a very low value on her own life and an exceptionally high value on loyalty to her family, her reason­ ing does not make sense to others, although she regards herself as rationally assessing her available choices. Ismene does not share Antigone’s unusually low valuing of individual life. Antigone recognizes that Ismene is in this way very dif­ ferent from herself. That recognition helps explain why, despite the angry accusa­ tions she makes against her sister, she does not accuse her of cowardice and seems eventually to accept Ismene’s choice.

6

‘Remember to What Sort of Man You Give this Favour’ Looking Back on Sophocles’ Ajax Lucy Van Essen-­Fishman

6.1 Introduction Sophocles’ Ajax is a figure defined by his past. To some extent, this is true of most tragic characters, inasmuch as they are shaped by a mythological backstory, but Ajax’s past is something of a special case; not only is Telamonian Ajax a prom­in­ ent Homeric hero whose Iliadic exploits will be familiar to the audience, but the fact that those heroic exploits are in the past is itself of crucial importance to the presentation of the hero in Sophocles’ tragedy.1 When the play begins, Ajax is certainly no longer the heroic figure who is described in the Iliad as the bulwark of the Greek army (Il. 3.229, 6.5, 7.211); we first see him suffering under the mad­ ness inflicted upon him by Athena, and he comes to his senses amidst the corpses of the animals he has mistaken for Greek generals. Whether Sophocles’ Ajax ever was the hero who appears in the Iliad is less clear; over the course of the play, vari­ ous figures, including Ajax himself, look back at his past and come to very differ­ ent conclusions.2 Taking this uncertainty about Ajax’ heroic status as my starting point, I will be examining the divergent ways in which different characters remember Ajax. I will argue that the selective and subjective memories of the hero which emerge over the course of the play and the rhetorical deployment of those memories at key moments together provide an important indication of how different figures understand the world in which the tragedy takes place. Although Sophocles’ Ajax is not a ‘memory play’ in the same sense in which Tennessee Williams uses the term in the opening scene of The Glass Menagerie—­ the events which take place onstage over the course of Ajax are not remembered events narrated by the characters involved—­looking at Ajax as a type of memory 1  The importance of the Iliad and Homeric models of heroism to Sophocles’ play is frequently noted; see e.g. Lattimore 1958, 68, Knox 1961, 20–2, Brown 1965, Winnington-Ingram 1980, 15–19, Easterling 1984, Hesk 2003, 24, Nooter 2012, 30–41, Schein 2012, 429–31. 2  Murnaghan 1989, 171 and Burian 2012, 69 both comment on the insistent focus of other charac­ ters on Ajax, even after his death. Lucy Van Essen-­Fishman, ‘Remember to What Sort of Man You Give this Favour’: Looking Back on Sophocles’ Ajax In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0006

LOOKING BACK ON SOPHOCLES’ AJAX  99 play can help to shed new light on the characters of the tragedy and their relation­ ships with one another.3 Every speaking character in Ajax looks back on the past at some point in the play, and, for most characters, the past which they remember serves as justification for their present beliefs and behaviour. Perhaps more inter­ estingly, as different characters in the play reconstruct remembered people, rela­ tionships, and events, their reported memories exhibit various traits which contemporary cognitive scientists have identified as typical of the workings of memory. I would therefore like to suggest that, if we examine the memories com­ municated onstage in terms of contemporary ideas about the construction and presentation of memory, we can see more clearly both what is at stake when a character engages in an act of memory and how differences in memory can be indicative of broader differences in perspective. I am first going to discuss a few general points about the workings of memory which are particularly relevant to Sophocles’ tragedy, and I will then look in detail at the various ways in which Ajax is remembered over the course of the play and at how memories of the hero are used for rhetorical effect. Ajax is often discussed in terms of his place in the Greek force as a whole, and the memories of Ajax recounted by characters and chorus often depend on a sense of collective identity. At the same time, most of the characters in Sophocles’ play are personally invested in their assessments of the figure of Ajax, and when they look back at Ajax’ past, their accounts of the hero exhibit traits which are common to remembered accounts of emotionally charged episodes. In particular, different figures seem to remember different versions of Ajax, and the versions which they remember seem to depend not only on their personal relationship with the hero but also on the particular contexts in which they remember him.

6.2  The workings of memory Of the many aspects of memory processing which have been identified by psych­ olo­gists and cognitive scientists, a few are particularly prominent in the memories which emerge over the course of Ajax. The first aspect is the close connection between memory and identity, which can be seen in Sophocles’ play as various characters use their memories of the past to justify their sense of who they are and how they fit into the world around them. The second is the variability of memory, inasmuch as individual memories often change over time through retelling and the acquisition of new information; tightly linked to this aspect of memory is the close connection between memory and emotion. Thus over the course of the play, characters remember a past which changes to match both their

3  Williams 1945, 5.

100  Lucy van Essen-Fishman current understanding and their current emotional states. Set alongside these three interconnected aspects of memory and cognitive processing is a final point about the rhetorical status of memory: despite growing awareness of the flexibil­ ity and subjectivity of memory, clear memories are often equated with authority. Thus Agamemnon, Menelaus, and Teucer, as they present shifting and selective memories of Ajax, all use their memories to defend their own positions and undercut the authority of those who seem not to remember the past as it really was. Memory is an important aspect of both individual and group identity; as Assmann puts it, ‘Memory is knowledge with an identity-­index, it is knowledge about oneself . . . be it as an individual or as a member of a family, a generation, a nation, or a cultural and religious tradition’.4 Although individual everyday mem­ or­ies may not seem to have much to do with who we think we are, Barclay and DeCooke argue that, taken together, our memories become ‘the fabric of a self-­ knowledge system’; our sense of ourselves, that is, derives in large part from auto­ biographical memories of people, places, and events in our past.5 Memory, however, is a social phenomenon as well as an individual one; although in­di­vid­ uals form their own memories from personal experience, they also partake in the collective memory of various groups. Inasmuch as groups are often defined in large part by the sense of a shared past, membership in such a group can be an important facet of individual identity.6 Collective memory provides a framework within which individuals can process their own experiences, while personal memories can, by imposing a degree of separation between the individual and the group, allow people to define their own relationship to their social context.7 Thus while the collective memory of the attack against Troy in the previous generation may colour the later Trojan War in the eyes of most of the Greek force, it becomes apparent in Sophocles’ play that Ajax’ particular perception of the Trojan War and his place in it is coloured both by his own individual memory of particular events and by a perspective on the previous Trojan War—­as the son of one of the heroes involved—­which he may not share with his former allies.8

4  Assmann 2008, 114. 5  Barclay and DeCooke 1988, 91. 6  The importance of group membership to individual memory is laid out by Halbwachs 1992, 38–9; see Assmann 1995, 127 and also Kirk 2005, 4. Assmann draws a distinction between Halbwachs’ idea of collective memory, according to which ‘we can base ourself on the memory of others’ (Halbwachs 2011, 141), and a narrower concept of ‘cultural memory’, which is ‘exteriorized, objectified, and stored away in symbolic forms’ (Assmann 2008, 110); see also Assmann 1995, 128–9. 7  Favorini 2008, 2 notes the individual and social elements of memory; see also Ostrom 1989. Schudson 1995, 346–7 argues that memory is inherently social. 8  Halbwachs 2011, 142–3 notes that collective memory ‘retains from the past only what still lives or is capable of living in the consciousness of the groups keeping the memory alive’; at the time when Ajax takes place, the first Trojan War, as a matter of collective memory, has remained continuously accessible to the Greeks participating in the later expedition. Ajax’ personal memories of Telamon will be discussed further below; Finglass 2012, 70 discusses the shadow of Telamon’s prowess as part of a wider theme of father–­son relationships in the play.

LOOKING BACK ON SOPHOCLES’ AJAX  101 Despite their importance in the construction of identity, the memories upon which we build our sense of ourselves are not fixed.9 Instead of remaining unchanged over time, memories are constantly reconstructed through the pro­ cesses of recollection and retelling; every time we remember something, our new recollection, which can be influenced by more recently acquired knowledge, takes the place of the original memory.10 As we form and reform memories of the past, we thus often adjust our memories in order to take into account our changing understanding of ourselves and the world around us.11 This adjustment, however, is typically unconscious; so long as a memory fits with our sense of the world and what we believe is likely to have happened, we will accept that memory as ac­cur­ate.12 Thus when Agamemnon denigrates Ajax’ importance to the Greek army (Aj. 1236–8), it is possible to see his account of Ajax’ limited value as a reconstructed memory of the hero, which is sincerely believed, but based more on Ajax’ recent attempt to slaughter his former allies than on his earlier deeds at Troy. In addition to the effect of new information on the reconstruction of mem­or­ ies, memories can also shift as a result of a range of emotional factors. Although scholars agree that there is a strong connection between memory and emotion, the exact nature of that connection is still not entirely clear. One model which has gained support in recent years is the theory of mood congruence, which suggests that people are generally much more likely to recall memories that match their current emotional state; when asked to recall incidents from their past, people who are depressed are likely both to remember sad incidents and to have diffi­ culty calling to mind happy memories.13 Current emotional states, moreover, affect not only which events people remember but also how they remember cer­ tain events, and depending on the emotional state in which it is remembered, the same event can be variously perceived as happy or sad.14 When this aspect of mood congruence is combined with the fact that memories get rewritten as they are recalled and retold, it becomes apparent that the emotional charge of a given memory can evolve over time. When Ajax looks back on his father’s glorious 9  Neisser 2009, 80–1 notes that, despite the flexibility of memory, most metaphors for memory, such as the tape recorder and the photograph, tend to assume permanence and accuracy. Freeman 2010, 263 notes that increasing awareness of the changing nature of memory has expanded the rele­ vance of memory studies beyond psychology and into fields as diverse as cultural history and literary criticism. 10  Casey 2000, 271–2. Foster 2009, 14 comments on the importance of reconstruction in the pro­ cess of retrieving a memory, while McCauley 1988, 134 comments on the importance of information acquired after the fact. As Erll 2010, 7 puts it, ‘The past is not given, but must instead continually be re-­constructed and re-­presented’. See also Schacter 1995, 16–17. 11  Searleman and Herrmann 1994, 250. Schacter 1995, 17 collates a number of studies on retro­ spective bias in memory. 12  Barclay and DeCooke 1988, 92–3. 13  Baddeley 1997, 281, Hertel 2004, 193. The evidence for an alternate model, mood dependency, in which stored memories remain tied to the emotional state in which they were originally formed, is somewhat more equivocal; see Singer and Salovey 1993, 136–7 and Baddeley 1997, 281–3. 14  Singer and Salovey 1993, 132, Baddeley 1997, 281.

102  Lucy van Essen-Fishman deeds at Troy, after first being denied the arms of Achilles and then failing to avenge himself on the Greek generals, the emotional charge of his memory of Telamon’s prowess may depend to some extent on his awareness of his own cur­ rent status as a failed Greek hero.15 The flexibility of memory is widely acknowledged by cognitive scientists, but individual memories nevertheless continue to be widely accepted as authoritative sources.16 In particular, the assumed reliability of memory is often important to the legal process; eyewitness accounts of crimes are granted considerable author­ ity, and detailed accounts are often taken by juries to be particularly reliable on the grounds that elaborate memories are unlikely to be fabricated.17 When there is a question about who has the right to tell a given story, moreover, it is generally accepted that those who personally experienced the event in question have a degree of authority over those whose knowledge of the event is only second-­ hand.18 Individual memory thus has a certain amount of automatic rhetorical force; when we say that we remember something, we are making an implicit claim to be giving a reliable account. Given that others may not remember the same things, however, we may also be setting ourselves up for conflict.19 When Teucer claims that Agamemnon no longer remembers Ajax’ heroic deeds (1266–79), he suggests not only that Ajax has been dishonoured but also that Agamemnon’s authority is not what Agamemnon thinks it should be. Over the course of Ajax, every character, including Ajax himself, at some point looks back on the figure of Ajax as he was before the Judgement of the Arms. Since each character has or had a different relationship with Ajax, and since most of those relationships are highly emotionally charged, the Ajaxes remembered by individual characters are very different. All these different Ajaxes, however, are remembered with confidence; although characters on several occasions question each other’s memories, no character ever admits that his or her own memory of Ajax might not be correct. Given that the play begins in medias res, after Ajax has been snubbed by the Greek leaders and driven mad by Athena, the audience has no way to know for certain whether Sophocles’ Ajax ever really was as either his friends or his enemies remember him, and I will not be attempting to argue that 15  Schacter 1995, 16 comments on the importance of the retrieval environment in the construction of memory; Freeman 2010, 267 notes that ‘the vantage point from which one remembers . . . represents a kind of ending, which in turn serves to transform, and perhaps falsify, the meaning of the events of the past’. 16  Schacter 1995, 20–1, arguing for the ‘fragile power’ of memory, comments, ‘memories are often ephemeral and distorted, on the one hand, yet subjectively compelling and influential, on the other’. Owens 1999, 315 notes that we tend to grant authority to our memories unless we are given conflict­ ing evidence. 17  Neisser 2009, 81, Shaw and Porter 2015, 291. 18  Zandberg 2010, 5, 11. Freeman 2010, 273–4 discusses the qualitative differences between first-­ hand and ‘mediated’ memories. 19  Ostrom 1989, 215 notes that two people who were both present for an event may not both remember that event after a significant lapse of time.

LOOKING BACK ON SOPHOCLES’ AJAX  103 any individual character’s memories are more or less reliable than any other’s.20 Instead, I will be arguing that each character remembers the Ajax who fits best with his or her individual world view, and that characters regularly deploy their memories of Ajax when they feel that their world views are under threat.

6.3  Memories of the living Ajax Both the importance of determining who Ajax was and the difficulties inherent in relying on memory to develop a clear picture of Ajax become apparent as early as the prologue. When Odysseus asks Athena not to call forth the maddened Ajax, her subsequent question, although it does not refer specifically to memory, requires Odysseus to draw on his past knowledge of Ajax; she asks, ‘Was he not previously a man?’ and Odysseus replies, ‘Yes, he was an enemy to me, and he still is even now’ (77–8). Less than fifty lines later, after Ajax has gone back inside, Athena again questions Odysseus, asking whether there was ever a man more prudent than Ajax was or better at doing what was necessary (119–20); Odysseus first replies as Athena’s question prompts him, but then veers away from Athena’s suggestion and says that he pities Ajax, ‘even though he is hostile [to me]’ (καίπερ ὄντα δυσμενῆ, 121–2). Although Odysseus still acknowledges that Ajax was and is his enemy, his perception of Ajax seems to have shifted.21 Having seen Ajax’ cur­ rent misfortune, Odysseus focuses less on past enmity than he did before, and, rather than exulting in the fall of a rival, he uses his awareness of Ajax’ good qualities as the basis for a sympathetic formulation of the human condition.22 While Odysseus’ comment can be taken as a general indication of the difficulty of passing judgement on another human being, he also exhibits, for the first time in the play, the general human tendency to reconstruct memories according to the circumstances at hand.23 As the play continues, Odysseus’ particular ability to revise his own assessment of his former enemy in the light of current circum­ stances will become one of his defining characteristics.24 After the prologue establishes the importance of the character of Ajax as he used to be, much of the rest of the play is filled with recollections of different past Ajaxes. From the first episode and right through until the suicide of Ajax, the audience is confronted, on the one hand, with Ajax’ own memories of who he was before his humiliation and, on the other hand, with both Tecmessa’s differently

20  Poe 1987, 35 similarly suggests that the fact that the play begins after the quarrel makes it impos­ sible to assess the wrong which Ajax claims to have suffered at the hands of the Greek leaders. 21  Blundell 1989, 63 notes that, for Odysseus, ‘enmity is significantly, and emphatically, distin­ guished from the total indulgence of hatred’. 22  Burian 2012, 72; see also Blundell 1989, 63–4. 23  Searleman and Herrmann 1994, 250, Schacter 1995, 16. 24  Hesk 2003, 45–6 discusses Odysseus’ ability to pity Ajax as a matter of intellectual rigour.

104  Lucy van Essen-Fishman focused recollections of Ajax and her assertions about what it is that Ajax ought to remember about his own past. After Ajax’ suicide, the memory of the hero remains a contested issue; Menelaus and Agamemnon, in turn, each look back on an Ajax who was always a problematic member of the Greek force, while Teucer tries to rehabilitate the memory of his heroic but dishonoured brother. Given the emotional investment of all involved, neither the tension between Teucer’s mem­ ory of Ajax and that of the Atreidae nor the similar tension between what Ajax remembers at the end of his life and what Tecmessa remembers of him can be resolved, and, at the end of the play, it falls to Odysseus to find a way to reconcile the conflicting aspects of Ajax’ identity. When in the prologue Athena and Odysseus look back on Ajax as he used to be, they do so by way of commenting on the magnitude of his fall. As Ajax re­covers his senses, he too draws attention to what he was before his recent mis­ fortune. In considering his downfall, however, Ajax focuses on different elements of his past from those on which most other figures in the play focus. Although he does lament the events of the immediate past, culminating in his failed attempt on the lives of Odysseus and the Atreidae, he does not identify his recent delusion as an isolated turning point in his fortunes.25 While Odysseus and Athena seem to view Ajax’ downfall as something which has taken place since the slaughter of the livestock, Ajax looks back further, to the moment when he was denied the arms of Achilles, and places this loss in the context of both his own and his father Telamon’s prior gains at Troy. In his lament over his current fortunes, Ajax presents with equal conviction two distinct sides of his experience at Troy, one in which he is a great hero, and another in which none of his heroic deeds count for anything. Near the end of the kommos, Ajax, apparently fully committed to the idea of his own greatness, looks back on his own arrival at Troy and addresses the waters of Scamander, saying that never again will they behold such a man as he was, such a man, he says, as Troy never saw coming from the land of Greece (418–26). Ajax admits that this statement of his past prowess is a bold one—‘I shall utter a great boast’ (ἔπος ἐξερῶ μέγα, 422–3), he says—­and it seems also to entail a selective memory; as is often noted, Ajax is in Homer the second best of the Achaeans, after Achilles, but Sophocles’ hero for the time being either forgets or chooses to ignore the exist­ ence of the greater warrior.26 In the iambic speech which follows, Ajax reinstates Achilles in his view of the Greek army, but he does so as a way of affirming his own status. Looking back at the contest over the arms, Ajax speculates about what Achilles would have done if he could have awarded the arms himself, saying that

25  Ajax’ regret focuses not on the fact that he attempted to attack his former allies, but on the fact that the attempt failed; see Winnington-­Ingram 1980, 27, Hesk 2003, 43. 26 Winnington-­Ingram 1980, 14–15. See also Lawrence 2005, 21, who suggests that Ajax here ‘defines himself as he was before his humiliation and in a sense eternally is (in his own view at least)’.

LOOKING BACK ON SOPHOCLES’ AJAX  105 Achilles would never have awarded anyone but Ajax the first place in valour (442–4). In speaking about Achilles’ arms, Ajax of course cannot ignore Achilles, but, as he recalls the hero, he reconstructs an Achilles who agrees with his own present perception of the situation and is willing to ratify Ajax’ status as his right­ ful heroic heir.27 In addition to looking back on his past role as a great hero amongst the Greeks, Ajax remembers a very different version of his time at Troy, in which all of his past glories have been negated by the loss of the arms. This version of Ajax’ past is tied closely to his memories of his father Telamon, and Ajax follows his first state­ ment of his own misfortunes after the kommos with an account of his father’s glory: Telamon, says Ajax, went home from Troy having won ‘the first and fairest prizes of the army’ (τὰ πρῶτα καλλιστεῖα . . . στρατοῦ, 434–6). Ajax contrasts his own current dishonour with his father’s success, lamenting that although he came to Troy ‘not less in strength, and carried out deeds with [his] hands which were no lesser’ than his father’s (437–9), now he has been brought low, ‘without honour among the Argives’ (ἄτιμος Ἀργείοισιν, 440).28 Although Ajax has recently declared himself to be such a warrior as never came to Troy from Greece, all of his own past achievements seem to disappear when he looks back on the achieve­ ments of his father, which, in turn, are magnified by comparison with Ajax’ pre­ sent circumstances. In his current state of despair, Ajax’ emotions colour his memories, and he struggles to remember a past in which he was a worthy succes­ sor to his famous father.29 Ajax’ memories of his deeds at Troy continue to resurface as he moves towards his own death. In the deception speech, and again in his suicide speech, Ajax pre­ sents a view of his duel with Hector which is, like many of his memories of Troy, emotionally congruent with his present assessment of his own place in the Greek army. In both speeches, the duel with Hector becomes another turning point in Ajax’ heroic career; while the duel in Book 7 of the Iliad is one of the moments when the Homeric hero appears to his best advantage, Ajax in the deception speech claims that since he received Hector’s sword, ‘I have received no benefit from the Argives’ (οὔπω τι κεδνὸν ἔσχον Ἀργείων πάρα, 663), while Hector, whom Ajax has no particular reason to hate, except as a prominent representative of the Trojans, becomes ‘most hated Hector’ (Ἕκτορος . . . δυσμενεστάτου, 662).30 Similarly, in the suicide speech, Ajax describes the sword as the gift of Hector, the most hated and hostile of all Ajax’ guest-­friends (ξένοι, 817–18). Ajax’ hatred of

27  See Barclay and DeCooke 1988, 92–3 on the reconstruction of memory to align with a world view; Chater and Loewenstein 2016, 141–9 discuss the converse, in which people may avoid informa­ tion which threatens their understanding of the world. 28  Heath 1987, 180 suggests that the contrast between Ajax and Telamon is ‘inherently pathetic’. 29  Baddeley 1997, 281; Hertel 2004, 193–4 notes the particular link between rumination upon unhappiness and the generation of negative memories. 30  Knox 1961, 16, Winnington-­Ingram 1980, 19.

106  Lucy van Essen-Fishman Hector is connected to his perception of the Trojan land—­and all that comes out of it—­as inherently hostile; remembered places often retain a strong emotional charge, and, while Ajax’ own downfall has come from conflict with the Greeks, Troy gains a sinister significance through its connection with Ajax’ change in fortune.31 As recalled by an Ajax who has been broken by subsequent events, the  duel with Hector, which might otherwise be seen as an unequivocal ­affirmation of his heroic status, becomes the beginning of his death in a foreign and hostile land.32 Ajax’ memories are all to do with military prowess, whether his father’s or his own. His initial accounts of his past, however, are bracketed by speeches from Tecmessa, who affirms Ajax’ heroic stature while reminding him of the non-­ military obligations to which a hero ought to be subject. Tecmessa, like many fig­ ures in the play, frequently looks towards the past; over the course of the tragedy, she will become notable not just for what she remembers about Ajax, but also for her insistence that the act of remembering is itself important. As she tells the Chorus at the very beginning of the first episode about Ajax’ maddened slaughter of the livestock and his subsequent gradual recovery of his senses, Tecmessa focuses for the most part on the recent past, but she also refers briefly to how Ajax used to be before he was driven out of his mind. There is some continuity between the maddened Ajax and the Ajax who is more familiar to Tecmessa; although she describes questioning his armed nocturnal foray as ‘unbidden’ (289–90), suggest­ ing that it was a departure from Ajax’ usual practice, her description of the gno­ mic words with which he silences her as ‘brief, but often repeated’ (βαι’, ἀεὶ δ’ ὑμνούμενα, 292) suggests that such advice from Ajax is not atypical.33 It is Ajax’ violent lamentation as his senses return to him that is unusual; before his mad­ ness, she says, ‘he always taught that such grieving belonged to base and down­ hearted men’ (319–20) and therefore restrained himself.34 Tecmessa’s remembered Ajax, it seems, is not just a paradigm of strength and self-­contained masculinity, but is also concerned both with how his actions will appear to others and with encouraging others to behave in accordance with his values. Tecmessa’s references to Ajax’ recent departure from his typical character establish her credentials as someone who both cares about Ajax and knows how he should be expected to act. Her remembered experiences give force to her argu­ ments, while at the same time allowing her to present a version of herself which,

31  Neisser 1989, 74. 32  See Singer and Salovey 1993, 132 on the ability to remember a single event as either happy or sad depending on the circumstances under which the event is recalled. 33  Finglass 2011, ad 292, although Hesk 2003, 55 notes that Ajax’ injunction is typical not just of Ajax, but of contemporary male rhetoric more generally. De Jong 2006, 88–9 comments on the emphasis on Tecmessa’s surprise and confusion in her account of Ajax’ behaviour. 34  Hesk 2003, 56 comments, ‘Ajax may be returning to some form of sanity, but he is not himself ’.

LOOKING BACK ON SOPHOCLES’ AJAX  107 inasmuch as it is defined by past suffering, has much in common with Ajax.35 More importantly, though, Tecmessa’s focus on the past, which is inherently social, establishes a network of obligation by which Ajax ought to be bound.36 After her initial summary of the reversal of fortune which she has suffered and Ajax’ role in it, Tecmessa begs Ajax to spare her the further misfortune which she is likely to suffer after his death.37 Her account of the future that awaits her is peppered with references to what will soon be her past; as she speculates that one of her captors will comment, ‘Look at the bedmate of Ajax, who was the strongest in the army—­what servitude she endures who was once so envied!’ (ἴδετε τὴν ὁμευνέτιν | Αἴαντος, ὃς μέγιστον ἴσχυσε στρατοῦ, | οἵας λατρείας ἀνθ’ ὅσου ζήλου  τρέφει, 500–2), she urges Ajax to remember their shared past in the light of how it appeared at the time to Tecmessa and others in the Greek army.38 Tecmessa’s subsequent account of the central place which Ajax—­ as the destroyer of her home, her captor, and now her sole protector—­occupies in her life (514–19) concludes with the play’s first sustained appeal to memory.39 She follows her assertion, ‘I depend on you entirely’ (519), with a demand on Ajax, followed by an explanation: ‘But hold onto your memory of me; for a man should remember, if he experiences something pleasing’, (ἀλλ’ ἴσχε κἀμοῦ μνῆστιν· ἀνδρί τοι χρεὼν | μνήμην προσεῖναι, τερπνὸν εἴ τί που πάθοι, 520–1). After this first sug­ gestion that people ought to feel an obligation to remember past favours, Tecmessa makes the role of memory as a marker of noble character more explicit a few lines later, commenting, ‘when a man loses his memory of good treatment, that man can no longer be noble’ (ὅτου δ’ ἀπορρεῖ μνῆστις εὖ πεπονθότος, | οὐκ ἂν γένοιτ’ ἔθ’ οὗτος εὐγενὴς ἀνήρ, 523–4).40 For Tecmessa, memory is both a facet of identity and an important force in the maintenance of social relationships; as scholars of collective memory have argued, remembering—­and fulfilling—­obligations to friends and family is a crucial aspect

35  Barclay and DeCooke 1988, 92 discuss the ability to use ‘generalized life experiences to convey one’s sense of self to an audience’. 36  Heath 1987, 182 notes, ‘Tecmessa is made to appeal to just those points which might sway an Ajax’, arguing that, as a hero, Ajax ought to be moved by an appeal to his sense of responsibility for his dependents. 37  Foley 2001, 91 notes, ‘both [Ajax’] life and his death seem extensions of her own’. Welzer 2008, 208 suggests that ‘references to the past are in fact a constitutive part of shared experience’; by drawing on the past which she has shared with Ajax, Tecmessa suggests that Ajax’ social memory ought to prevent him from trying to separate himself from his connections. 38  Ormand 1999, 110–14 argues that the relationship between herself and Ajax which Tecmessa recalls here is a rhetorical construct which may bear little resemblance to her actual past relationship with Ajax. 39  Both Tecmessa’s hypothetical account of what will be said about her after Ajax’ death and her account of her sole reliance on him have clear parallels to Andromache’s address to Hector in Iliad 6; on the relationship between the two scenes, see e.g. Lattimore 1958, 68, Brown 1965, Winnington-­ Ingram 1980, 15–19, Easterling 1984, Burian 2012, 76. 40  Poe 1987, 49. Blundell 1989, 75 notes that Tecmessa uses the ideas of nobility which Ajax used to justify his desire for death as an argument for continuing to live.

108  Lucy van Essen-Fishman of remaining part of a community.41 Tecmessa suggests that if Ajax partakes in her memory of a shared past, he will come to understand that his community still needs him. She therefore urges Ajax to remember not so much particular events in his life, but rather the bonds which he has formed with those to whom he is—­ or ought to be—­close, chief among them Tecmessa herself, but also including their son Eurysaces and Ajax’ own parents.42 The vision of Ajax’ parents which Tecmessa conjures up for Ajax—­of his father, whom Ajax is ‘leaving behind in bitter old age’ (506–7), and his mother, ‘who prays over and over to the gods that [Ajax] will come home alive’ (508–9)—is strikingly different from the vision of Telamon which Ajax himself called up earlier in the episode; whereas, for Ajax, Telamon is simply an example to whom he will never measure up, Tecmessa sug­ gests that Ajax and his parents all have a stake in each other’s well-­being.43 Ajax draws upon his memory of Telamon’s past glory and comes to the conclusion that he can no longer occupy the place at the top of the Greek hierarchy which should be reserved for his father’s son, but Tecmessa suggests that if he remembers his family, Ajax should be able to see that he is still bound by a web of reciprocal obligation. As Tecmessa and Ajax look back on the past in the first and second episodes of the play, they not only focus on different past events and figures but also seem to process their memories differently and use them to different rhetorical effect. For both Ajax and Tecmessa, memory is a crucial factor in determining identity; Ajax, however, focuses for the most part on himself, arguing that he cannot live up to the identity which he has retrospectively constructed for his father, whereas Tecmessa’s view of identity is largely communal and based on memories of a shared past.44 Similarly, although both remember in emotional terms, their mem­ or­ies focus on different emotional states. For Ajax, even the memory of past glory grows bitter in the context of present humiliation, while Tecmessa, despite the largely sorrowful nature of her own past experience, argues that remembered pleasure should be enough to compel Ajax to carry on living and caring for his family (520–4).

41  Schudson 1995, 346–7 discusses the importance of memory to the construction of groups. See also Assmann 2008, 114. 42  Blundell 1989, 75. 43  Hesk 2003, 64 notes ‘she is picking up Ajax’s point that he is ashamed to face his father, and she is standing it on its head’. See also Holt 1981, 278–9 on Tecmessa’s appropriation of Ajax’ moral ter­min­ ology. As she reminds Ajax of how his parents must be thinking of him, Tecmessa draws on a kind of collective memory of Ajax as he used to be; Halbwachs 2011, 141 argues that ‘the events of our life most immediate to ourself are also engraved in the memory of those groups closest to us’, and Tecmessa suggests that the collective memory of Ajax’ family ought to count for as much as Ajax’ own recollections of himself. 44  Ajax and Tecmessa here approach the question of how one determines which memories are really one’s own from very different angles; on the complexities involved in that question, see Freeman 2010, 273–4 and Halbwachs 2011, 140–1.

LOOKING BACK ON SOPHOCLES’ AJAX  109

6.4  Debating the memory of Ajax Although Ajax is moved to express a degree of pity for Tecmessa (652–3), he is not sufficiently swayed by the memories by which she suggests that he can and should live, and he goes to his death still speaking about a past defined by his military achievements, in which the duel with Hector was a turning point on the road to dishonour.45 Ajax’ death, however, by no means settles the question of his memory; while Ajax’ own memories of his past are fixed as they were when he committed suicide, the dispute over whether to allow him burial is in large part a dispute over which elements of his character and history ought to be remem­ bered.46 Both Menelaus and Agamemnon, when they arrive on the scene, remem­ ber an Ajax who is defined almost exclusively by his recent actions; although they do speak about Ajax as he was before the Judgement of the Arms, the Ajax whom they remember is compatible with his more recent attacks against the other lead­ ers of the Greek army, which suggests that their memories of Ajax may have been revised in light of recent events.47 Teucer, on the other hand, in an opposite effort to maintain a consistent view of his brother, glosses over the deeds which set Ajax at odds with the rest of the Greek force and remembers an essentially Iliadic Ajax who is defined by his noble deeds at Troy.48 It is only Odysseus who can clearly remember Ajax as he was both before and after his madness, and his ability to remember two very different Ajaxes is crucial to his stance on the question of burial. The question of how Ajax will be remembered comes to the fore almost as soon as his corpse is discovered. Shortly before Teucer’s entrance, Tecmessa speculates about the likely reaction of the Atreidae; although she initially assumes that Agamemnon and Menelaus will rejoice at Ajax’ misfortune, she then considers the possibility that ‘perhaps, too, even if they did not want him while he was liv­ ing, now that he is dead they will lament for him in the necessity of battle’ (ἴσως τοι, κεἰ βλέποντα μὴ ’πόθουν, | θανόντ’ ἂν οἰμώξειαν ἐν χρείᾳ δορός, 962–3).49 This possibility, as Tecmessa expresses it, is part of a broad trend in human perception, and she explains that, in general, ‘men of bad understanding do not know the good they have in their hands before they lose it’ (οἱ γὰρ κακοὶ γνώμαισι τἀγάθ’ ἐν

45  Poe 1987, 75 argues that Ajax in his suicide speech tries to distance himself from his current circumstances. 46  Burian 2012, 81 comments, ‘After [Ajax’] death . . . those who survive can begin to reconsider the meaning of his life from their own perspectives’. 47  McCauley 1988, 134 notes changes in memory based on information acquired long after the remembered event. 48  Teucer here exhibits what Chater and Loewenstein 2016, 137–8 discuss as the ‘drive for sense-­ making’, although they are primarily concerned with the tendency to interpret new information in a manner that is consistent with prior belief. 49  Garvie 1998, ad 961–3 comments on the fact that Tecmessa’s ability to look past the likely laugh­ ter of Ajax’ enemies sets her apart from Ajax.

110  Lucy van Essen-Fishman χεροῖν | ἔχοντες οὐκ ἴσασι πρίν τις ἐκβάλῃ, 964–5).50 People’s perceptions of the past change according to the needs of the present moment, and, as Tecmessa had earlier hoped that Ajax might remember his family after losing the support of his former allies, she now hopes that the Greek leaders will miss Ajax’ prowess in battle and remember him more favourably.51 While Tecmessa’s theory that people will remember what it suits them to remember at any given moment is confirmed in the ensuing debate over Ajax’ burial, it does not play out in accordance with her hopes. The present moment, for Menelaus and Agamemnon, is coloured not by grief at the loss of a valuable fighter, but by anger at Ajax’ recent deeds, tinged with relief that Ajax is no longer in a position to trouble the commanders of the Greek army.52 When Menelaus enters, he justifies his injunction forbidding the burial of Ajax on two fronts; first he says that he issued the injunction ‘because it seemed right to me, and to him who rules over the army’ (δοκοῦντ’ ἐμοί, δοκοῦντα δ’ ὃς κραίνει στρατοῦ, 1050), and then, displaying a memory of Ajax coloured by his present anger, he goes on to explain that Ajax never really did anything to deserve honour at the hands of the Greeks.53 After admitting that he and Agamemnon had hoped to lead Ajax to Troy ‘as an ally and friend to the Achaeans’ (Ἀχαιοῖς ξύμμαχόν τε καὶ φίλον, 1053), Menelaus elides the majority of Ajax’ time at Troy in order to claim, ‘in our ex­peri­ence we found him more hateful than the Phrygians’ (ἐξηύρομεν ξυνόντες ἐχθίω Φρυγῶν, 1054). In the light of Ajax’ recent actions against his former allies, Menelaus remembers a consistent version of the fallen hero; Menelaus is deter­ mined to affirm after Ajax’ death the hierarchy that, at least according to Menelaus, Ajax was never willing to accept in life (1067–70).54 Agamemnon, when he enters, likewise denies that Ajax was ever truly valuable to the Greek force. Unlike Menelaus, who simply passes over the period between the arrival of the fleet at Troy and the Judgement of the Arms, Agamemnon con­ siders the deeds which Ajax did on behalf of the Greek force and suggests that they were not the deeds of an outstanding hero.55 After insulting Teucer’s status and casting doubt on his assertion that Ajax sailed to Troy ‘as his own ruler’ (αὐτὸς ἄρχων, 1234), Agamemnon begins to question Ajax’ heroic credentials. 50  Finglass 2011, ad 962–3 compares Achilles’ repeated assertions that he will be missed by the Achaeans (Il. 1.240–1, 11.609–10), noting that the belief that great men are only truly appreciated after they are gone is a commonplace. 51  See Schacter 1995, 17 on retrospective bias. 52  Holt 1981, 283 draws attention to ‘the highly personal character of the Atreidai’s resentment’. 53  Hesk 2003, 110 observes that, in 1050, Menelaus displays ‘the unattractive rigidity and intransi­ gence which he and his brother will maintain throughout’; in the light of his current anger, Menelaus’ hostile memory of Ajax can be seen as an example of mood congruence in memory, as discussed by Singer and Salovey 1993, 132. 54  Blundell 1989, 91–2 notes that, while Menelaus’ formal accusation deals with Ajax’ betrayal of his former allies, his main concern is with the question of Ajax’ status; on remembering to suit the needs of the present moment, see Searleman and Herrmann 1994, 250 and Schacter 1995, 17. 55  Burian 2012, 79.

LOOKING BACK ON SOPHOCLES’ AJAX  111 After demanding of Teucer, ‘What sort of man are you shouting about so arrogantly? Where did he go or stand where I did not?’ (ποίου κέκραγας ἀνδρὸς ὧδ’ ὑπέρφρονα, | ποῦ βάντος ἢ ποῦ στάντος οὗπερ οὐκ ἐγώ; 1236–7), Agamemnon reminds Teucer that there are other men among the Greeks and that Ajax was therefore not indispensable (1238).56 The most recent episode in Ajax’ life, in which he attacked his leaders ‘with a trick’ (σύν δόλῳ) and refused to accept that he was ‘one of the defeated’ (οἱ λελειμμένοι, 1245), is the only one which matters in Agamemnon’s assessment of the fallen hero. In his debates with Menelaus and Agamemnon, Teucer presents himself as cor­ recting the record in the light of the accounts which the Atreidae have used to defend their decision to forbid Ajax’ burial.57 After Menelaus’ dual assertion that he and Agamemnon brought Ajax to Troy as their ally and that Ajax was therefore at fault for not accepting a subordinate role in the Greek hierarchy, Teucer passes over the issue of Ajax’ attack on the Greek leaders in favour of questioning the assumptions that lie beneath Menelaus’ account.58 Teucer first reminds Menelaus that Ajax was not brought to Troy at all but came of his own accord, ‘as his own master’ (αὑτοῦ κρατῶν, 1099), and then calls Menelaus’ own place in the Greek army into question.59 Since, according to Teucer’s version of the Greek hierarchy, ‘You came here as the king of the Spartans, not as our ruler’ (Σπάρτης ἀνάσσων ἦλθες, οὐχ ἡμῶν κρατῶν, 1102), Menelaus never had any right to expect obedi­ ence from Ajax. Menelaus’ attack on Ajax as dangerously insubordinate stems, in Teucer’s view, from a fundamental misunderstanding both of who Ajax was and of how Ajax fitted into the Greek force at Troy. In his subsequent debate with Agamemnon, Teucer more explicitly frames his rehabilitation of his brother in terms of correcting Agamemnon’s memory; since Agamemnon’s memory is flawed, Teucer suggests, his assessment of Ajax cannot be taken seriously.60 At the beginning of his response to Agamemnon’s questions about what Ajax ever did for the Greek army, Teucer apostrophizes Ajax, lament­ ing that kharis must be fleeting, if this man here . . . no longer has any memory of you, Ajax, although on his behalf you many times, risking your life, toiled with your spear 56  Holt 1981, 285, in support of the continuity between the two debates, notes that Agamemnon starts out by rebutting a claim which Teucer made to Menelaus. 57  Burian 2012, 79 notes that Agamemnon’s denigration of Ajax’ services to the Greeks gives Teucer an opening to remind him and the audience of Ajax’ noble deeds. 58  Burian 2012, 78. 59  Blundell 1989, 92 calls Teucer’s claim ‘a sophistic quibble’ and argues that, even if Ajax was not subordinate to the Atreidae, he would still have been bound to help them by bonds of philia. 60  Owens 1999, 325 notes, ‘a faulty memory cripples reason’; Teucer’s assessment of Agamemnon seems to be aligned with this view of the importance of reliable memory.

112  Lucy van Essen-Fishman εἰ σοῦ γ’ ὅδ’ ἁνὴρ οὐδ’ . . . Αἴας, ἔτ’ ἴσχει μνῆστιν, οὗ σὺ πολλάκις τὴν σὴν προτείνων προὔκαμες ψυχὴν δορί (1268–70).

He then turns directly to Agamemnon and asks whether he remembers any of a litany of Ajax’ famous deeds against the Trojans.61 With the disbelieving ‘do you not remember anything any more?’ (οὐ μνημονεύεις οὐκέτ’ οὐδέν, 1273) with which he begins his account of Ajax’ services to the Greek army, Teucer suggests that Agamemnon’s failure of memory is also a serious failure of loyalty and leadership; in the account which follows, Ajax is of critical importance to the Greek cause, at one time single-­ handedly saving his allies from Hector’s attempt to burn the ships and at another time willingly meeting Hector in ­single combat (1273–87).62 Agamemnon, like Menelaus, has forgotten the real Ajax, without whom the other Greek leaders might never have made it to this point in the Trojan War, and, when Teucer asks, ‘You wretch, where can you look as you say these things?’ (δύστηνε, ποῖ βλέπων ποτ’ αὐτὰ καὶ θροεῖς; 1290), he suggests that Agamemnon’s ability to forget the truth disqualifies him not only from dis­para­ging the fallen Ajax but also from making any claim about his own status. Although they differ in how they remember Ajax, Teucer and the Atreidae are in some ways similar in the workings of their memory; whether they are for or against Ajax, they seem to agree that one must pick one aspect of the fallen war­ rior to remember and pass over the rest. The aspects of Ajax on which they focus, however, depend on the emotions called up by his memory; for the Atreidae, Ajax’ actions since the contest over the arms blot out any services he previously rendered for the Greek force, while for Teucer, Ajax is so purely heroic that, avoiding all mention of Ajax’ attempt against the lives of the Greek leaders, he can ask Menelaus whether Ajax was ever openly his enemy (1133).63 Both Teucer and the Atreidae have a great deal riding on the treatment of the fallen Ajax; the Atreidae suggest that not only their own honour but also the orderly functioning of the Greek force as a whole is closely linked to their ability to deny burial rites to a man who denied their authority (1067–70, 1087–8, 1246–9), while Teucer sug­ gests that his standing in both the Greek army and Telamon’s family has been

61  Hesk 2003, 123 notes that the list is made up of ‘episodes which Agamemnon has conveniently forgotten’. 62  Murnaghan 1989, 181 suggests that Teucer’s account of Ajax’ noble deeds allows him, in a way, to reverse the Judgement of the Arms. Finglass 2011, ad 1272–9 notes differences between Teucer’s account and the events of the Iliad, suggesting that Teucer is here magnifying Ajax’ role. 63  Blundell 1989, 92–3 comments on Teucer’s word choice, noting that a πολέμιος is a military opponent and that as such his question is, strictly speaking, valid. On emotional memory, see Singer and Salovey 1993, 132 and Baddeley 1997, 281.

LOOKING BACK ON SOPHOCLES’ AJAX  113 materially damaged by his brother’s ignoble fate (1006–22).64 For Teucer, as for the Atreidae, changing his understanding of Ajax would constitute a major shift in his world view, and it is in the light of this threat to his understanding of the world that Teucer doubles down on his memory of Ajax. He alone remembers who Ajax really was, and, he suggests, Agamemnon’s ability to forget Ajax’ noble past makes him unfit to exercise authority over Ajax and his kin.

6.5  Odysseus and the possibility of impartial memory It is into this highly fraught standoff—­in which the ability to remember Ajax as he really was has become symbolic of the ability to participate honourably in the army and Greek society in general—­that Odysseus enters.65 With him, Odysseus brings a way of recalling the past which is distinct from that exhibited by Ajax’ other friends and foes. Whereas both Teucer and the Atreidae speak as though their own reputations hinge on defending a single authoritative version of Ajax’ character, Odysseus seems to have little personal stake in the issue. He explains to Agamemnon that refusing Ajax’ burial would not harm Ajax, but it would be an affront to the gods (1343–4), and he suggests that it would not be worthwhile to insist on symbolically dishonouring Ajax at the risk of divine displeasure. As he was in the prologue, moreover, Odysseus is willing to accept the duality of Ajax’ character; although he freely admits to Agamemnon that Ajax was once his worst enemy in the Greek army, he acknowledges that he was also ‘the best of all the Achaeans who came to Troy, except for Achilles’ (1340–1).66 Odysseus’ ability to remember an Ajax who was hostile but nevertheless w ­ orthy of honour amongst the Greeks puts him starkly at odds with Agamemnon. In fact, Odysseus’ recommendation that Ajax should be allowed proper burial despite his past hostility towards the Atreidae is so antithetical to Agamemnon’s tendency to view the world in absolute terms that Agamemnon questions whether Odysseus is really remembering the same Ajax, and urges him, ‘remember to what sort of man you give this favour’ (μέμνησ’ ὁποίῳ φωτὶ τὴν χάριν δίδως, 1354). By challenging Odysseus’ memory, Agamemnon can suggest that his own world view, in which friends are rewarded, enemies are punished, and the ­distinction between the two categories is clear, is the only sensible one; if Odysseus should remember correctly who Ajax really was, then he would surely agree that

64  Finglass 2011, ad 974–1027 argues that Teucer’s worries about his fate after Ajax’ death ‘[mag­ nify him] by association’ with Ajax, inasmuch as Teucer will take on the same problems with Telamon and the Atreidae which troubled Ajax before his death. See also Hesk 2003, 108–9. 65  On Odysseus as a counterpoint to the other major characters in the play, Winnington-­Ingram 1980, 66 comments, ‘Odysseus enters to meet a certain situation, to achieve a certain purpose’. 66  Blundell 1989, 96 notes that Odysseus ‘distinguishes both enmity and hatred from the evaluation of worth and proper distribution of honour’.

114  Lucy van Essen-Fishman Ajax should not be entitled to proper burial.67 Memory is for Agamemnon—­as it was for Teucer less than a hundred lines previously—­a means of affirming the legitimacy of his world view in the face of conflict, and a treacherous Ajax with­ out redeeming features is an incontestable element of that world view. As Odysseus looks back on Ajax, acknowledges that he was an enemy, and admits that his hostile status does not negate his good qualities, commenting, ‘This man was my enemy, but he was noble’ (ὅδ’ ἐχθρὸς ἁνήρ, ἀλλὰ γενναῖός ποτ’ ἦν, 1355), he engages in a very different kind of memory processing from those so far exhibited by other figures in the play.68 In remembering Ajax as he used to be, Teucer, the Atreidae, and Ajax himself are all subject to the natural tendency to be swayed by their current emotions, and each consequently remembers an Ajax who is distorted to fit the needs of the present moment; even Tecmessa, as she looks back on her relationship with Ajax and urges him to remember his obliga­ tions to his kin, paints a picture of domestic happiness which is uniquely suited to her current need to impress upon Ajax the importance of his familial connec­ tions.69 All of these characters, moreover, seem to be unaware of just how well their memories fit their present needs, and they use their contradictory memories of Ajax confidently in support of contradictory opinions about how Ajax should be treated after death.70 Odysseus, however, seems to resist the general tendency to remember selectively; when asked to remember Ajax, he remembers both his own hostile personal experience with Ajax and the broader truth about Ajax’ valu­able place in the Greek force at Troy. Having acknowledged both sides of Ajax’ character (1336–41), he advises Agamemnon to allow the burial of Ajax’ corpse not on the basis of his own emotionally charged memories of Ajax—­Ajax was, after all, ‘the most hateful member of the army’ (ἔχθιστος στρατοῦ, 1336) as far as Odysseus was concerned—­but on the basis of divine law (1343–4).71 Both Odysseus’ ability to see multiple aspects of Ajax’ character and his ability to step back and examine his own mental processes make him an anomaly in the world of the play. Unlike Agamemnon and Menelaus, he can remember Ajax as an enemy without erasing the memory of his nobility; unlike Tecmessa and Teucer, he can argue that Ajax is deserving of honour without ignoring his crimes against his former allies. As he explains to Agamemnon when he urges him, 67  Finglass 2011, ad 1354 makes a comparison to Tecmessa’s earlier appeal to Ajax’ memory and notes, ‘In Agamemnon’s view, Ajax has failed to show the χάρις which would merit χάρις in return (522).’ 68  Jebb (1896), ad 1355 notes that ποτέ refers both to ἐχθρός and to γενναῖος. 69  Poe 1987, 49, Ormand 1999, 112–14. Singer and Salovey 1993, 132, Schacter 1995, 16–17, and Baddeley 1997, 281 discuss biases in memory stemming from the emotional state and other circum­ stances under which an event is recalled. 70 Barclay and DeCooke 1988, 92–3 note both that memories tend to be congruent with the rememberer’s world view and that memories which are congruent with the rememberer’s world view are typically accepted as accurate; see also McCauley 1988, 134 on the influence of new information on memory recall. 71  Hesk 2003, 126.

LOOKING BACK ON SOPHOCLES’ AJAX  115 ‘Do  not let violence overcome you so that your excessive hatred leads you to trample on justice’, (μηδ’ ἡ βία σε μηδαμῶς νικησάτω | τοσόνδε μισεῖν ὥστε τὴν δίκην πατεῖν, 1334–5), Odysseus can see the risk of letting memory become clouded by personal grievance, and he is therefore uniquely situated to under­ stand those of Ajax’ friends and foes who can now see Ajax only through the lens of their own emotional responses. Unlike those who, depending on their own experiences, remember either a purely heroic Ajax or a purely hostile one, however, Odysseus has since the prologue been able to put the fallen hero’s story in a broader human context. While Agamemnon and Teucer continue until the very end of the play to operate on the assumption that alternative views of Ajax are inherently threatening, Odysseus’ more nuanced approach to the memory of Ajax allows him to emerge from the play with his world view more or less intact. Moreover, by facing the subjective nature of human memory head-­on, Odysseus can provide both his fellow characters and the audience of the tragedy with a model for engaging constructively with the ambiguity of the past.

6.6  Conclusion: looking at memory through Ajax As a kind of memory play, Ajax has much to say both about the particular ways in which characters depend on their memories as they navigate the world of the play and about the more general workings of memory itself. Memory processing, in the play as in real life, is highly individual; not only does each character have his or her own memories and draw upon those memories in the construction of his or her own identity, but also, and perhaps more interestingly, each character uses his or her individual memories in highly individual ways. Thus Ajax takes his memories of Telamon’s glory and his own prior achievements as proof of his cur­ rent isolation and hopelessness, while Tecmessa tries to argue that Ajax’ mem­or­ies, which are—­or should be—­in some ways parallel to her own, ought in fact to cement his connections to his family. After Ajax’ death, Teucer, Menelaus, and Agamemnon all weaponize their memories of the fallen hero; their argument over how Ajax ought to be remembered becomes a proxy battle in a broader argument about the leadership of the Greek force at Troy. Throughout the play, moreover, Tecmessa, Teucer, and Agamemnon argue that their own ways of remembering are the only correct ways, and it is only Odysseus who ac­know­ ledges multiple ways of remembering the same past. For the most part, memory as we see it in Ajax is so subjective as to be useless for reconstructing past events with any degree of reliability. On the basis of the conflicting stories told by most characters about Ajax’ past contributions—­or lack thereof—­to the Greek effort against Troy, we cannot determine whether he ever was, as Teucer and Ajax himself suggest, the bulwark of the Greek army familiar from the Iliad, or whether he was, as the Atreidae argue, merely one more soldier

116  Lucy van Essen-Fishman with an overblown idea of his importance to the war. What each character remembers is the version of Ajax which fits best with his or her world view; Tecmessa remembers an Ajax who was concerned with his nobility and would therefore want to honour his familial obligations, Ajax remembers a version of himself who surpassed all other Greeks at Troy, the Atreidae both remember an Ajax who was merely filling a subordinate role—­and not filling it very well—­and we are given no way to decide which version is most accurate. Memory, in the play, is not a route to objective truth about the past, but it is instead a window into how characters understand themselves and their world.72 If Odysseus’ memory is more reliable than that of other figures in the play, it is insofar as his memory of the antipathy between himself and Ajax does not prevent him from remembering that Ajax once seemed—­to Odysseus as well as the rest of the Greek force—­to be the best of the Achaeans after Achilles (1336–42). The conflicts of memory—­in which one character suggests that another has forgotten something crucial or is otherwise relying on a false version of the past—­ which occur throughout the play serve to highlight and intensify more obvious conflicts between characters. Divergent memories can drive a wedge between people; implicit in the connection between memory and personal identity is the idea that people whose memories differ are themselves different.73 In this way, both Tecmessa’s claim that Ajax, if he is truly noble, ought to remember past kindnesses (520–4) and Teucer’s disbelieving question about whether Agamemnon can really have forgotten all that Ajax did for the army (1273) highlight the gulf in understanding which has developed between figures who, for one reason or another, ought to be on the same side. Assmann argues that ‘Remembering is a realization of belonging, even a social obligation’; seen in this light, when Ajax no longer remembers the favours he received from Tecmessa or Agamemnon forgets his obligation to Ajax, their failures of memory represent a wider disintegration of social bonds.74 When he remembers both sides of Ajax’ character and encourages Agamemnon to do the same, Odysseus shows the Greeks a way to recover from the damage done by their divergent memories and to move forward again as allies.

72  As Barclay and DeCooke note, ‘Autobiographical recollections are not necessarily accurate, nor should they be; they are, however, mostly congruent with one’s self-­knowledge, life themes, or sense of self ’ (1988, 92). 73  On the link between memory and personal identity, see Barclay and DeCooke 1988, 91–2 and also Assmann 2008, 113–14. 74  Assmann 2008, 114.

7

Thinking Through Things Extended Cognition as a Consolatory Fiction in Greek Tragedy Anne-­Sophie Noel

7.1 Introduction Writing about the loss of his father, Paul Auster acutely captured the affective and cognitive impact of objects on bodies and minds. There must have been more than a hundred ties, and many of them I remem­ bered from my childhood: the patterns, the colors, the shapes that had been embedded in my earliest consciousness, as clearly as my father’s face had been. To see myself throwing them away like so much junk was intolerable to me, and it was then, at the precise instant I tossed them into the truck, that I came closest to tears. More than seeing the coffin itself being lowered into the ground, the act of throwing away these ties seemed to embody for me the idea of burial. I finally understood that my father was dead.1

Mundane objects can play a crucial mediating role in an affective and intellectual process. Imbued with childhood memories, the ties act as surrogates for Auster’s father, strangely surviving the deceased and sharply challenging the living.2 Physical interactions with them are more effective in making the death ac­know­ ledged and understood than the funeral itself. Auster thus clearly articulates the capacity of objects to materialize (‘embody’) ideas, so that those ideas become real, tangible things, whose physical presence cannot be evaded. This literary intuition predates recent scientific and philosophical approaches to the roles of materiality in thinking. The idea of smart or cognitive artefacts, which contribute to our intellectual functions, emerged when scientists started to explore interactions between humans and computers.3 Man-­ made, physical 1  Auster 1988, 11. 2  See Gibson 2008 for a sociological enquiry on how ‘death reconstructs our experience of personal and household objects in particular ways’ (1). 3  Norman 1991. Anne-­Sophie Noel, Thinking Through Things: Extended Cognition as a Consolatory Fiction in Greek Tragedy In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0007

118  Anne-Sophie Noel objects like the aptly named smart phones, but also more basic office supplies such as notebooks, calendars, or to-­do lists, are said to enhance our cognitive abilities by, for instance, stabilizing and focusing attention, stimulating memory, helping information processing or problem-­solving.4 The theory of extended cognition, proposed by Andy Clark and David Chalmers in the late 1990s, argues for the extension of cognition across brain, body, external objects, and environment.5 Applying it to the field of cognitive archaeology, Lambros Malafouris and Colin Renfrew have forged the concept of the ‘cognitive life of things’, which questions the prevailing assumption that only humans can engage in cognitive processes.6 Drawing on prehistorical artefacts such as the Stone Age Acheulean ‘handaxe’, Malafouris describes the act of knap­ ping a stone as a cognitive one, where the labour is shared between human and thing. Far from being a passive agent or a blank surface, the stone is ‘a tightly coupled, transparent and intrinsic part of the knapper’s cognition’.7 Objects in ancient Greek tragedy have recently been studied through the vari­ ous lenses of performance studies, affect studies, or cognitive psychology.8 New Materialist and posthumanist ideas, which strive to go beyond the dichotomy of subject and object in favour of a more balanced collaboration between humans and nonhumans, have also been applied to this specific corpus of objects.9 This proves well enough the depths of meaning involved in the handling of objects onstage. In this chapter, I nevertheless aim to add a further dimension, by dem­ onstrating how close readings of some tragic scenes can be deepened by the the­ or­et­ic­al framework of extended cognition. I develop a series of case studies drawn from Aeschylus’ Oresteia and Euripides’ Heracles, in which characters seek to include objects in their cognitive processes (related to memory, reflection, and problem-­solving).10 In the texts under discussion, however, I will show that objects resist as much as they ‘collaborate’. Tragic characters’ interactions with objects challenge some aspects of the ‘cognitive life of things’. This no doubt results partly from the diversity of objects considered (a house, a lock of hair, a piece of clothing, a bow and its arrows), and partly from the fact

4  Norman 1991, 17; Clark 2010, 23. 5  Clark and Chalmers 1998. 6  Malafouris and Renfrew 2010. For an account of ‘4E cognition’ (embodied, embedded, enacted and extended mind) applied to a wide range of ancient phenomena, see Anderson et al. 2019; applied to theatre, see Meineck 2018b, 6. In this monograph, Meineck explicitly states that the topic of ‘props from cognitive perspectives’ falls outside the scope of his book (Meineck 2018b, 2). See also the Introduction to this volume, p. 1–2, 5. 7  Malafouris and Renfrew 2010, 18. 8  Chaston 2010, Fletcher 2013, Noel 2013, Mueller 2016, Noel 2016, Noel 2018, Telò and Mueller 2018. Chaston focuses on a different cognitive aspect, i.e. the function of imagery in reasoning (2010, 4–5). 9  Telò and Mueller 2018, Canevaro 2019, Noel 2019a, Chesi and Spiegel 2019, Bianchi, Brill, and Holmes 2019. 10 On modern ‘cognitive artefacts’ and memory, reflection, and problem-­solving, see Norman 1991, Beach 1993, Clark 2010, 27, Heersmink 2013, Vaccari 2016, 887.

Thinking Through Things  119 that these objects are not properly tools, as are the stone tools or the potter’s wheel in the case studies of Malafouris. This diversity can nevertheless prove to be a good test for the validity of his general theory and an occasion for discussing and refining some aspects of it. In particular, even if objects, such as Auster’s ties, can evidently be external catalysts for our mental functions, the idea that objects become ‘internal to the cognitive process’,11 or even, parts of an ‘extended mind’,12 involves a further leap, and one that one may feel reluctant to make. As a matter of fact, this proposition is still debated among philosophers. For instance, Michael Wheeler argued that objects (like the flight strips used by air traffic controllers) remain distinct from cognition even though their materiality can implement and enhance performance in the storage and manipulation of information.13 As part of this debate, I aim to show that Malafouris’ seductive metaphors, expressing a ‘synergistic’ continuity between mind and objects (for instance, call­ ing the latter a ‘continuous part of the machinery’),14 are rather contradicted by tragic case studies. A distance remains between humans and objects in the plays under discussion, even when the latter mediate the characters’ cognitive pro­ cesses. The way thinking with things is represented in Greek tragedy, I argue, emphasizes the fact that it is human beings who choose to engage in a form of make-­believe, which involves the anthropomorphizing of the environment. By situating these observations within recent psychological accounts of anthropo­ morphism as a speculative fiction that aims to reduce the otherness of the exter­ nal environment, I seek here to refine Malafouris’ conception of the ‘cognitive life of things’ by placing it within the framework of make-­believe and the consolatory mechanisms of the human mind.

7.2  Partnering with knowledgeable objects in the Oresteia? The Oresteia trilogy offers opportunities to think about the way fictional charac­ ters integrate material objects into their cognitive endeavours. Its representation of humans and objects complicates any strict ontological differentiation. When Heraclitus criticizes those who pray to statues as if they were gods, he claims that this practice makes no more sense than having a conversation with a house.15 Like other inanimate objects, houses are excluded from the domain of knowledge and cognition. It is certainly possible that Heraclitus’ rejection of statues as a medium for communication with divinity was idiosyncratic, but it is equally pos­ sible that his dictum reflects a more widespread attitude towards material

11  Malafouris 2013, 85. 12  Clark and Chalmers 1998. 13  Wheeler 2010, 32. 14  Malafouris 2004, 54, 58. 15 Heraclitus fr. B 5.22 DK: καὶ τοῖς ἀγάλμασι δὲ τουτέοισιν εὔχονται, ὁκοῖον εἴ τις δόμοισι λεσχηνεύοιτο, οὔ τι γινώσκων θεοὺς οὐδ’ἥρωας οἵτινές εἰσι.

120  Anne-Sophie Noel objects.16 When Demosthenes in his speech Against Aristocrates speaks of the practice of trying implements such as ‘a stone, a piece of wood, or iron’ for ­homicide at the Prytaneion he emphasizes that these are ‘inanimate beings that do not take part in reasoning’ (τῶν ἀψύχων καὶ μὴ μετεχόντων τοῦ φρονεῖν, Dem. 23.76).17

7.2.1  The house of Agamemnon In a different way, the Oresteia confronts us with the representation of a house that is invested with cognitive abilities. A ‘house is an intelligent object, . . . having an undoubted effect on human bodies and brains. Houses are not mental exten­ sions, but containers, shapers and complex products of human action all at once.’18 In this formulation by the cognitive archeologist Chris Gosden, ‘object’ is taken in the broad sense of a material, bounded, fabricated thing (a car, a plane or a skyscraper are objects, by this definition). His description of the contribution a house makes to cognition provides an interesting frame for an interpretation of the house of Atreus, which was given material representation by the skênê at the back of the orchêstra. Far from being meaningless decor, the skênê offered new dramaturgic possibilities (playing on the rooftop, screaming from inside, rolling out the ekkuklêma to reveal the result of an offstage action, etc.) and could become the locus of striking symbolism (the doors of the house, for instance, being merged with the doors of Hades).19 A material building constantly visible onstage, it is also a ‘recurrent element in the discourse’, as noted by Kitto, so that one might almost think of it as one of the most important actors of the play.20 The house of Atreus is a complex extension, not of a single character, but of the several gen­er­ ations that have lived under its roof: in containing the ‘old insufferable pains’ (παλαιὰ ἄλγη δύσοιστα, Cho. 744–5), the house becomes inextricably linked with them, as indeed it gets connected to its own crimes (παλαιὰς τώνδ’ ἁμαρτίας δόμων, Ag. 1197). The house comes to be described both as a personified victim and perpetrator of these past crimes.21 In this elaborate metonymic relationship, the house is qualified as a cognitive entity with which human beings may interact. From the prologue of the Agamemnon onwards, it is addressed as a potential interlocutor that would have plenty to divulge. ‘Were it to find a voice’ (εἰ φθογγὴν λάβοι, Ag. 37), the Watchman 16  On attitudes to statues in the Greek world, see Vernant 1965, Donohue 1988, Steiner 2001, Bouffartigue 2007, Johnston 2008, Kindt 2010, Gaifman 2012, Bremmer 2013. 17  Unless otherwise indicated, all translations are my own. 18  Gosden 2010, 41. 19  Ag. 1291. Cf. Taplin 1977, 276, 452–9; Vasseur-Legangneux 2004, 61, Sommerstein 2010, 154–7, Bakola 2017. 20  Kitto 1956, 31–2, Sommerstein 2010, 154–7. 21  Ag. 1309; ‘the house breathes murder’ (φόνον δόμοι πνέουσιν), but is also wrecked by the crimes of the Atreidae (Ag. 1532–3, Cho. 50, 262–3, 843).

Thinking Through Things  121 hopes, the house could very clearly tell what it knows (σαφέστατ᾽ ἂν λέξειεν, 38) and disclose the highly dysfunctional family relationships inside it, generation after generation. The voice of the Watchman sleeping on the roof (στέγαις Ἀτρειδῶν, Ag. 3) for months may even be interpreted as the voice of the house itself.22 Yet his voice is stifled and cannot unveil the secrets of the house, while the  house itself lacks the physical speech organs to tell what it knows: this is a thwarted partnership since the surrogate voice of the Watchman remains in­hibit­ed and silenced.23 The ‘cognitive life’ of the palace is built step by step as we watch various other characters engage with it. When Cassandra enters, she describes the house as ‘sharing the knowledge’ (συνίστορα, Ag. 1090) of the many crimes that it con­ tained.24 This is the only use of the compound συνίστωρ (‘who knows along with another’) in Aeschylus’ extant plays; elsewhere in tragedy, the word occurs a handful of times to refer to the superior knowledge of the gods from whom noth­ ing can be hidden.25 This confirms the palace’s representation as a supernatural cognitive, memory-­enabling entity. A storehouse was a common metaphor for memory in antiquity:26 it seems to be taken in a literal sense when Cassandra discovers the dark memories hidden in the attic—­the souls of the dead children of Thyestes that have stayed attached to this house, as well as the Erinyes, waiting for revenge on the roof.27 At the conclusion of his ground-­breaking study of art and material agency, Alfred Gell explains, apropos of New Zealand Maori meeting houses, that these houses are like ‘containers’ and even ‘bodies’: ‘Houses are bodies . . . because they have organs of sense and expression—­eyes which peer out through windows and spyholes, voices which reverberate through the night. To enter a house is to enter a mind, a sensibility’.28 Aeschylus’ staging of the house is in line with this in­ter­ pret­ation. Throughout the trilogy, little by little the palace gains anthropomorphic ‘organs of sense and expression’: the Chorus exhort it to raise its ‘head’ (ἀνιδεῖν, Cho. 808) in order to see ‘from its friendly eyes’ (φιλίοις ὄμμασιν, 810–11) the light of the coming freedom. At the end of Choephori, they hope for the ‘eye of the house’ (ὀφθαλμὸν οἴκων) to be saved by the horrendous, yet necessary, murder of Clytaemnestra:29 ὀφθαλμός designates both Orestes, who comes to retrieve his

22  Ag. 18; Sommerstein 2010, 155. 23  Ag. 26–7. 24  In Sophocles’ Electra, the ‘old’ (παλαιά, 483) double axe used by Clytaemnestra and Aegisthus to kill Agamemnon comparably ‘remembers’ the murder it has enacted (οὐ γάρ ποτ᾽ ἀμναστεῖ, 481), as if the object were a container for the revengeful, ‘brazen-­clawed’ Erinys (491). 25 Soph. Ant. 542, Phil. 1293; Eur. Supp. 1174. 26 Cf. Rhet. Her. 3.16–22; August. Conf. 10.8.12; Yates 1966, 2–26; on the persistence of this image in neurosciences, see Markowitsch 1994 and Dressler et al., 2017. 27  Ag. 1091–7, 1192, 1217–22, 1242–4 (Thyestes’ children); 1117–18, 1186–90 and Cho. 971 (the Erinyes are described as μέτοικοι, ‘resident aliens of the house’). 28  Gell 1998, 252–3. 29  Cho. 934.

122  Anne-Sophie Noel lineage and patrimony,30 and the house, itself called upon to rise up from the ground where it has fallen (ἄναγε μὰν δόμοι, Cho. 963).31 Between Heraclitus’ rejection of houses as unsuitable interlocutors and Aeschylus’ verbal and visual staging of a living house, we may detect some traces of a dialogue about the ontology of material things. The interactions with the pal­ ace challenge a restrictive conception of a house as an ἄψυχον, excluded from cognition, notably by the anthropomorphic projection involved, while suggest­ ing, at the same time, that a ‘house’s mind’ cannot be accessed by anyone.

7.2.2  Electra thinking through things This same dialectical tension between the possibility and impossibility of access to objects, affording (or not) a cognitive collaboration, structures Electra’s in­ter­ pret­ation of tokens in the famous recognition scene of Choephori. It is a well-­ known fact that Electra and Orestes’ anagnôrisis, achieved through the discovery and exposition of a series of tokens, involves an elaborate cognitive activity dependent on material objects.32 Goldhill’s now classic reading has demonstrated Aeschylus’ acute awareness of the ambivalence of signs and the challenges of communication through language.33 As Electra acts as a decipherer of signs, struggling to link a ‘signifier’ and a ‘signified’, her cognitive investment is at its highest level. However, the cognitive support offered by the objects involved is more elusive than critics have noted. For Electra, thinking about the lock of hair, and recognizing the cloth that she had woven with her own hands, are distinct cognitive operations that are not equally successful: this will lead me to suggest that anthropomorphic projection onto objects happens when characters actually fail to decipher them. When Electra finds a lock of hair on the grave of Agamemnon, it seems to present itself as a direct response to her recent prayer to her father and the nether gods. Its meaning nevertheless remains mysterious. Aristotle’s categorization of this recognition as one ‘which results from inference’ (ἐκ συλλογισμοῦ) does just­ ice to the complex cognitive processes involved.34 Electra first uses deductive reasoning—­the process of knowledge acquisition is emphasized by the double use of the verb μαθεῖν in the choral responses to her investigation.35 Electra’s guesses 30  Orestes is called ‘eye of my joy’ (τερπνὸν ὄμμα) by Electra, Cho. 238. 31  In a fragment of Aeschylus’ Edonians, the house of Lycurgus is described as participating in the Bacchic frenzy (TrGF fr. 58). 32  Goldhill 1984, 120 reads in it a ‘complex process of recognition’, contra Fraenkel 1950, 815–16, Appendix D; Garvie 1986, 86f. For other attempts to make sense of these tokens by putting them back in their cultural context or showing the complex semiotic operations involved in their interpretation, see Bond 1974, Roux 1974, Ronnet 1975, Goldhill 1984, 120–37, Jouanna 1997. 33  Goldhill 1984, 126. 34  Poet. 1455a4–6. 35  Cho. 171, 175.

Thinking Through Things  123 are at first based on the norms of hair offerings in the context of funeral rituals: only a close parent may honour the grave of a dead kin with this specific type of offering.36 Her second series of hunches is based on her visual perception:37 Electra perceives a visual resemblance that is striking enough to make this guess ‘easy for anyone to make’ (εὐξύμβολον τόδ᾽ ἐστὶ παντὶ δοξάσαι, Cho. 170).38 Yet, with the verb δοξάσαι, Electra remains in the realm of supposition and opinion rather than certainty: the adjective εὐξύμβολον also contains the idea of an ac­cess­ible omen that nevertheless needs to be interpreted.39 At this stage, her deductions lead her to an aporia, not to the acquisition of positive knowledge (Cho. 187–94). The object resists her attempts to engage with it intellectually. This failure of deductive reasoning effects a cognitive shift in Electra. She dis­ tances herself from the rational mindset she adopted initially in her hermeneutic attempts and projects cognitive functions onto the object itself: Ah! If only it had a mind and a voice like a messenger, so that I wouldn’t be tossed about in two minds, but would know for sure that I should reject this lock, if it really was cut from the head of an enemy, or else, if it was my kin, it would be able to join in my mourning, giving glory to this tomb and honour to my father! We appeal to the gods who know in which kind of storms, like ships, we are whirled around—­though if we are destined to find safety, from a tiny seed a great tree-­trunk can spring.40 φεῦ· εἴθ᾽ εἶχε φωνὴν ἔμφρον᾽ ἀγγέλου δίκην, ὅπως δίφροντις οὖσα μὴ ᾽κινυσσόμην, ἀλλ᾽ εὖ σαφ᾽ ἤινει ἢ τόνδ᾽ ἀποπτύσαι πλόκον, εἴπερ γ᾽ ἀπ᾽ ἐχθροῦ κρατὸς ἦν τετμημένος, ἢ ξυγγενὴς ὢν εἶχε συμπενθεῖν ἐμοὶ ἄγαλμα τύμβου τοῦδε καὶ τιμὴν πατρός. ἀλλ᾽ εἰδότας μὲν τοὺς θεοὺς καλούμεθα οἵοισιν ἐν χειμῶσι ναυτίλων δίκην στροβούμεθ᾽. εἰ δὲ χρὴ τυχεῖν σωτηρίας, σμικροῦ γένοιτ’ ἂν σπέρματος μέγας πυθμήν. (Cho. 195–204)

36  Cho. 172. 37  Cho. 168 (ὁρῶ), 176 (ἰδεῖν), 178 (προσείδεται), 187 (ἰδούσῃ). 38  Cho. 174, 176. 39  Ag. 8, the beacon of light is a σύμβολον (cf. Goldhill 1984, 120). 40  Trans. Sommerstein, Loeb Classical Library, modified.

124  Anne-Sophie Noel The ontological status of the lock of hair undergoes several transformations. Electra envisages it as having a voice (φωνήν) and a mind (ἔμφρον᾽), freeing itself from the condition of a mute and soulless thing. The comparison to the messen­ ger confirms that the object is the bearer of some knowledge and lacks only a voice to reveal it clearly.41 Like the house of Atreus, the lock of hair is imagined as having the potential to become a speaking object,42 an ‘intelligent’ object (ἔμφρον᾽ ) that may solve Electra’s dilemma, expressed here with a telling hapax: δίφροντις suggests a divided mind, balancing (᾽κινυσσόμην) two opposite attitudes. The object is therefore imagined as a potential interlocutor that would solve Electra’s dilemma and re-­ establish unity and stability in her mind.43 This ongoing an­thropo­morph­ic transformation eventually results in an affectionate compan­ ionship: the lock of hair itself becomes a relative (ξυγγενής); no longer is it a mere material token thought to have been left by her brother.44 The repetition of the prefix συν- (συμπενθεῖν; ξυγγενής) hints that Electra and the object that is her companion have the same status. As fellow mourners and kin, they experience similar emotions. As further material evidence of the kinship between brother and sister, the woven textile that Orestes shows to his sister triggers another response. In the whole trilogy, textiles are multi-­layered and versatile objects, endlessly prone to transformation.45 Scholars have focused on the symbolic interpretation of the fabric in relation to Orestes himself: his ὕφασμα connects him with a network of other ambivalent pieces of clothing. The woven patterns of wild beasts foreshadow cunning action and hunting, a metaphoric leitmotiv repeatedly associated with intra-­familial murders.46 But a close relationship is also suggested between this artefact and Electra herself: And look at this piece of weaving, the work of your hand, the strokes of the weaving sword and the picture of a beast.47 ἰδοῦ δ᾽ ὕφασμα τοῦτο, σῆς ἔργον χερός, σπάθης τε πληγὰς ἠδὲ θήρειον γραφήν. (Cho. 231–2)

41 In Ag., the Chorus have similarly contrasted the ‘voiceless’ smoke (ἄναυδος) of the beacons with the voice of a messenger that can speak clearly to them (496–8). Theognis also called a beacon an ἄγγελος ἄφθογγος (549). 42  And not necessarily as a ‘speaking subject’ as suggested by Goldhill 1984, 124. 43  Cho. 197 (εὖ σαφ᾽ ἤινει); 201 (εἰδότας μὲν). 44  This seems to be confirmed by line 226, a subsequent comment made by Orestes on the scene: Electra ‘seemed to see [him]’ when she was ‘looking at the lock’ (κουρὰν δ’ ἰδοῦσα τήνδε, 226). 45  Noel 2013, with a list of references; Mueller 2016, 42–69. 46  Lebeck 1971, Vidal-Naquet 1972, Goldhill 1984, Noel 2013, 168–9. 47  Trans. Sommerstein, Loeb Classical Library, modified.

Thinking Through Things  125 In these lines uttered by Orestes, with the emphatic possessive adjective (σῆς) and the mention of Electra’s creative hand (χερός) that is responsible for this work, the textile is immediately connected with Electra herself. With the children’s clothing that Orestes places before her eyes, he appeals to her exclusively and this is what makes this proof irrefutable: only she can recognize the piece of clothing as the one she wove for her brother. This memory-­enabling artefact does not function as passive external storage, but it is through a dynamic, physical interaction with its creator that it suddenly provides access to a common past.48 Orestes’ verbal evo­ cation of the σπάθη may indeed activate an embodied and enactive memory of her past textile work in Electra’s mind. Relying on experimental research to inves­ tigate ancient Greek warp-­weighted weaving looms, Susan Edmunds comments on the technical similarities between the σπάθη and the κερκίς (the ‘pin beater’).49 The σπάθη was nothing more than a flat stick inserted between the warp threads to separate them into upper and lower. But according to Edmunds’ research, each woman used her weaving sword or her pin beater in a unique manner: ‘the beat­ ing in of the weft is a mark of the weaver’s hand, a sign of her skill or perhaps even a signature to the knowing eye.’50 The recognition takes place, therefore, primarily through self-­recognition: when Electra sees the cloth, she first recognizes her own hand, work, and self in the patterns of the fabric, and hence infers from it that the person before her is her brother.51 Her cognitive process undeniably encompasses a material compo­ nent: this is one case of illustrating how ‘brain, body and thing conflate, mutually catalyzing and constituting each other’.52 In acting as the decisive proof that puts an end to Electra’s uncertainty, the woven textile resolves her experience of a divided mind (δίφροντις), bringing her back to the sense of self-­unity she was calling for.53 In this successful cognitive collaboration, the absence of anthropomorphic projection is noticeable. One may hypothesize that an anthropomorphic attitude to objects, which may as here involve direct-­speech address, is prompted when a character experiences resistance in the object’s mute presence and struggles to enter into communication with it. Rather than being a sign of successful com­ munication, anthropomorphic imagination would then be a strategy that charac­ ters adopt when confronted with the distance and otherness of objects.

48  Malafouris 2004, 57. 49  Edmunds 2012, § 47 (relying on Arist. Ph. 243b11). 50  Edmunds 2012, § 46–7. 51 Aristotle saw in this scene a recognition διὰ μνήμης: ἰδὼν γὰρ τὴν γραφὴν ἔκλαυσεν, Poet. 1454b37–55a4. 52  Malafouris and Renfrew 2010, 15. 53  McLure 2015, 221 associates woven tokens of recognition in general with ‘female memory and the construction of social identity within the domestic sphere’.

126  Anne-Sophie Noel

7.3  Anthropomorphic thinking and embodied memory in Euripides’ Heracles At the end of Euripides’ Heracles, Heracles is in a climactic situation of crisis and isolation after the killing of his own wife and children. Collapsed on the ekkuklêma among scattered debris and corpses of his relatives, he has his whole body and head draped in a cloak, in a posture of grief loaded with visual resonances.54 In view of our previous observations, one may ask whether his choice to engage in a dialogue with his bow and arrows is rather an embodied cognitive strategy for dealing with an intense identity crisis or whether it reveals an incapacity to integrate objects into cognition without anthropomorphizing them.55 How painful the pleasure of kissing you, how painful the embrace with these weapons! I do not know whether I should keep them or let them go, since they will hang at my flanks and say to me, ‘By means of us you killed your children and your wife: you hold us, the slayers of your children.’ Shall I then carry these in the hollow of my arms? What will I say? Or shall I strip myself of the weapons with whose aid I performed my glorious exploits in Greece, put myself at the mercy of my enemies, and thus meet a disgraceful death? I must not let them go but must in misery keep them.56 ὦ λυγραὶ φιλημάτων τέρψεις, λυγραί τε τῶνδ᾽ ὅπλων κοινωνίαι. ἀμηχανῶ γὰρ πότερ᾽ ἔχω τάδ᾽ ἢ μεθῶ, ἃ πλευρὰ τἀμὰ προσπίτνοντ᾽ ἐρεῖ τάδε· ἡμῖν τέκν᾽ εἷλες καὶ δάμαρθ᾽· ἡμᾶς ἔχεις παιδοκτόνους σούς. εἶτ᾽ ἐγὼ τάδ᾽ ὠλέναις οἴσω; τί φάσκων; ἀλλὰ γυμνωθεὶς ὅπλων, ξὺν οἷς τὰ κάλλιστ᾽ ἐξέπραξ᾽ ἐν Ἑλλάδι, ἐχθροῖς ἐμαυτὸν ὑποβαλὼν αἰσχρῶς θάνω; οὐ λειπτέον τάδ᾽, ἀθλίως δὲ σῳστέον. (HF 1376–85)

54  See Taplin 1972 and 2016; for similar scenes elsewhere in Euripides’ plays, see Cairns 2011a. 55  On this identity crisis, see Padilla 1992, 11, Dunn 1996, 122–6, Papadimitropoulos 2008, 138; Mueller rightly speaks of a ‘cognitive impasse’, 2016, 37. 56  Trans. Kovacs, Loeb Classical Library, modified.

Thinking Through Things  127 Just as Megara and her children were united in death in a ‘dire embrace’ (κοινωνίαν δύστηνον, 1363), Heracles and his weapons are joined in a ‘painful embrace’ (λυγραί τε τῶνδ’  ὅπλων κοινωνίαι, 1377): the choice of word suggests an unex­ pected and unnatural communion between a man and his weapons, the very instruments which have killed the family he should be uniting with instead. The tragic irony is breathtaking, but the noun κοινωνία, which covers various kinds of ‘communion’ or ‘partnership’, often with a physical aspect (‘embrace’ or even in some cases ‘sexual intercourse’) invites us to analyse the nature of Heracles’ collaboration with his weapons.57 Configured by Heracles’ verbal address to these objects, this ‘partnership’ is characterized by a fluid ontology: the objects are not only an addressed ‘you’, they become an active, collective, interlocutor speaking in the first person (ἡμῖν, ἡμᾶς)—with the limitation that it is always the human voice that grants them this status of active subject. In this ‘role-­play’, Heracles is the second-­person addressee, but he remains the subject of the two verbs (εἷλες and ἔχεις). This complex distribution of speech stirs a ‘cognitive life’ for the objects. The language of deliberative discourse is also informed by the sensual presence of objects. The arrows are felt dangling against Heracles’ sides (πλευρὰ τἀμὰ προσπίτνοντ᾽, 1379), a contact that triggers memory: it ironically duplicates the games of his children clinging to his clothes and his knees,58 inducing a painful reminiscence of past sensory interactions between father and sons. This memory seems to inform the very words he uses: when he envisages holding his weapons ‘in the hollow of his arms’ (ὠλέναις οἴσω, 1381–2), the phrasing also recalls his former playful interactions with his children.59 Memory is therefore described as an embodied experience that tricks Heracles, through the heart-­breaking false belief (and feeling) that he is holding his children. In helping his decision-­making, the integration of the objects into Heracles’ thinking has an ambiguous outcome. Endowing the weapons with speech is a way of discouraging himself from keeping them: otherwise, for the rest of his life, they will remain the gruesome objects embodying the killing of his kin. Yet, at the same time, the fact of attributing to them voice and thought, i.e. an­thropo­morph­ic traits, helps to shape a relationship of surrogacy between weapons and children—­ both scandalous and somehow comforting. Heracles’ effort to engage cognitively with his weapons therefore rather leads to an aporia, which is finally solved by the consideration of another, practical argument (weapons will preserve his life against his enemies). 57  Aristotle acknowledges the affective scope of κοινωνία, when he claims that, as an ‘association’, a ‘partnership’, κοινωνία is a precondition for friendship (ἐν κοινωνίᾳ μὲν οὖν πᾶσα φιλία ἐστιν, Eth. Nic. 1161b11). 58  HF 986 προσπεσών; see also for the same action, HF 79; Alc. 948; Soph. Trach. 904. 59  ὠλένη has clear affective overtones in Eur. Tro. 762, 1142; Phoen. 300, 306.

128  Anne-Sophie Noel

7.4  Make-­believe in drama: extended cognition as an ‘as-­if ’ fiction The framework of the ‘cognitive life of things’ envisions a perfectly smooth rela­ tionship between human and nonhuman agents. In the example of the Acheleuan axe, as Malafouris puts it, ‘sometimes it is the stone that becomes the “extension” of the knapper. At other times, however, it is the knapper that becomes the “exten­ sion” of the stone’.60 With the same idealizing rhetoric of reciprocity, he argues for a ‘continuum of potter’s brain-­body-­clay-­wheel’ in the chain of cognitive pro­ cesses that leads to the production of a clay pot.61 Taking his propositions as heuristic premises for analysing objects in Greek tragedy, I expanded his concept by applying it to objects of fiction, whereas the archaeological investigation into prehistorical cognitive artefacts excludes de facto literary evidence. In Greek tragedy, moreover, characters are not shown making things or tools, but interacting with them physically and verbally (to this extent, Electra’s engagement with fabric made by her own hands is an exception). Exploring the ‘cognitive life of things’ in ancient Greek tragedy therefore in­ev­it­ ably leads one to highlight points of contact with and divergence from Malafouris’ concept. In the Oresteia, I hypothesized that anthropomorphism and the vocal address­ ing of objects were signs of a challenging, rather than successful, affective and cognitive partnering with objects. In Euripides’ Heracles, the hero makes use of the same rhetorical devices in a deliberative discourse in the attempt to solve a dilemma. His decision-­making is arguably impacted by the embodied memory enclosed within the arrows and, simultaneously, by the rhetorical elaboration of these objects as surrogates for his lost sons. But does that mean that we can go as far as to say that the limits between the human being (and her/his cognition) and the external environment are removed entirely?62 An objection to such a conclusion becomes obvious, I believe, when one con­ siders the specific discursive and emotional framework in which this cognitive partnership takes place. With his favourite examples of the Acheulean handaxe or the potter’s wheel, Malafouris argues for an embodied and extended thinking ‘through things, in action, without the need of mental representation’.63 This imme­ diacy is absent from our textual evidence. When tragic characters are represented in the act of thinking with objects, mental representations occur in the form of anthropomorphic imagining—­the self-­consciously make-­believe projection of 60  Malafouris and Renfrew 2010, 19. 61  Malafouris and Renfrew 2010, 20, 22. 62  This question echoes a larger debate on whether it is possible ‘to make the transition from embodied-­embedded cognition to extended cognition’ (Anderson et al. 2019, 5, with bibliographical references). 63 Malafouris 2004, 58. The idea of ‘thinking outside the brain’ is equally radical (Malafouris 2013, 57).

Thinking Through Things  129 human traits onto objects. A self-­aware as-­if is a recurrent stylistic feature in the examples analysed: ‘were it to have a voice’ (εἰ φθογγὴν λάβοι, Ag. 37), the house could reveal the secrets of the Atreidae; ‘if it had a mind and a voice like that of a messenger’ (εἴθ᾽ εἶχε φωνὴν ἔμφρον᾽ ἀγγέλου δίκην, Cho. 195), the lock of hair would tell Electra the secret of its origin. In these two passages, the potential and counterfactual modes inscribe the dialogue with objects within the realm of the unfulfilled. The use of the future in Heracles’ speech may hint at a stronger cer­ tainty: dangling at his sides, the arrows ‘will tell him these words’ (ἐρεῖ τάδε, HF 1379), but the dialogue he sets up with the bow and arrows remains his very own. It is like a miniature theatrical play that he creates and directs himself as a response to the dilemma he encounters. Another example of this would be Philoctetes’ wistful address to his bow, the very bow he had received from Heracles and which is sought by the Greeks as the weapon destined to defeat the Trojans, in Sophocles’ tragedy. Deprived of his companion bow by Neoptolemus, Philoctetes ascribes to it mental faculties, but only in a hypothetical mood (‘if you have any mind’, φρένας εἴ τινας ἔχεις, Phil. 1130–1); imagining the bow in the devilish hands of Odysseus (an event that will in fact never come to pass outside his fantasy), he projects empathy onto his like-­minded weapon (‘you look with pity’, ἦ που ἐλεινὸν ὁρᾷς, 1130),64 consoling himself, like Electra, with this make-­believe integration of the weapon into an emotional community.65 If these material objects possess a ‘cognitive life’, therefore, this life remains on the plane of the imagination. Onstage, the cognitive labour is distributed between human agents and material ones only through the medium of a consoling fiction invented by the characters. I therefore find it productive to combine Malafouris’ concept with recent social-­ psychological research on anthropomorphism—­ a recurrent attitude towards objects in the passages analysed.

7.4.1  Anthropomorphism as a cognitive reflex It was odd, she thought, how if one was alone, one leant to things, inanimate things; trees, streams, flowers; felt they expressed one; felt they became one; felt they knew one, in a sense were one; felt an irrational tenderness thus . . . as for oneself.66

64  In Sophocles’ Ajax, when Ajax buries his sword so that ‘it very mindfully helps [him] to a speedy death’, (εὐνούστατον τῷδ’ ἀνδρὶ διὰ τάχους θανεῖν, 822) it is also a make-­believe fiction on his part that pursues a specific strategy (e.g. staging his suicide as a substitute for heroic fight). Contra Mueller 2016, 15–41, who emphasizes the weapon’s autonomous animacy and agency. 65  Contra Aristotle, who excludes objects from relations of φιλία (Eth. Nic. 1161a34–62b9). 66  Virginia Woolf, To the Lighthouse (1968, 74).

130  Anne-Sophie Noel In recent years, anthropomorphism has received most scholarly attention in the fields of anthropology, religion, and social psychology. In the Piagetian theory of child development, anthropomorphism was believed to be natural and intuitive, anchored in an animistic tendency perceptible in children’s thought processes.67 However, this way of thinking has been debunked by more recent experimental research in cognitive development. In his cognitive and evolutionary approach to religion, Pascal Boyer states that even when they attribute humanlike traits to their toys and engage in a dialogue with them, children do not expect these objects actually to respond to them and would probably react with horror if they did.68 He suggests that anthropomorphism is a universal tendency (albeit with cultural variants), which enhances the human cognitive response to an insecure or uncanny external environment.69 In the anthropomorphic projection, he argues, children and adults alike entertain fictions, ‘salient counterfactual specu­ lations’, which retain attention and are transmitted culturally from generation to generation, in spite of their fictitious, counterfactual nature.70 Other experimental research in social psychology supports this conclusion, by shedding some light on the social circumstances and motivational states that favour this anthropomorphic speculation. As Bartz and his team put it, social disconnection prompts efforts to forge new bonds (…); what is more remarkable, however, is the flexibility and ingenuity of some of these attempts. . . . Some people anthropomorphize inanimate objects to serve as sources of social connection—. . . lonely people (compared with nonlonely ­people) were more likely to ascribe humanlike traits (e.g. free will) to an alarm clock, battery charger, air purifier, and pillow.71

On the contrary, when the participants in their experience were reminded of a close, familiar social connection, it attenuated their propensity to an­thropo­ morph­ize objects.72 These consistent findings from social psychology resonate with the situation of Greek tragic characters, in plays that draw their dramatic intensity from the ex­acer­ba­tion of misfortune and pathos. When one is alone, ‘one lean[s] to things, inanimate things’: the intuition of Woolf ’s Mrs Ramsay is in line with that of 67  Boyer 1996, 83–4; for Lakoff and Johnson 1980, 25–35, the ontological metaphor of personifica­ tion is also a basic cognitive mechanism, rooted in the hyperactivity of human beings’ agent-­detection capacities. 68 Boyer 1996, 86. See also Boyer 2001, Boyer 2018 (in particular 93–124), Boyer and Baumard 2016. 69  Guthrie 1993, Gell 1998, 121–2. 70  Boyer 1996, 89. 71  Bartz et al. 2016, 1644. Their experience was based mainly on questionnaires and fMRI neuro­ imaging, see also Maner et al. 2007, Epley et al. 2008a and 2008b, Waytz, Epley et al. 2010, Waytz, Morewedge et al. 2010; Waytz, Heafner, and Epley 2014. 72  Bartz et al. 2016.

Thinking Through Things  131 Greek tragedians, who worked with hyperbolic solitude as an almost inevitable ingredient of their plots. In the Oresteia, Electra is not alone in a literal sense, since she is accompanied by a chorus of female libation-­bearers. Still, she is utterly marginalized in her opposition to Clytaemnestra and Aegisthus, and being re­united with her brother is the only prospect that may put an end to her isolation. In Euripides’ play, Heracles’ solitude is brought to light by his staging as physically separated from his father and friend (Amphitryon and Theseus), collapsed on the ekkuklêma, behind a barrier of debris and corpses. His veiling is another significant action that emphasizes his seclusion.73 In Sophocles’ Philoctetes, the solitude of the hero abandoned on a desert island is a leitmotiv, constantly brought to mind by the repetition of the adjective ἔρημος (‘isolated’).74 When Philoctetes, the one ‘who loves his possession’, is deprived of his weapon, he experiences a climactic moment of despair: a remarkable juxtaposition of epithets conveys his exclusion from friendship, civilized society, and finally, the world of the living (‘deprived of friends, isolated, deprived of a city, a dead man among the living’, ἄφιλον, ἔρημον, ἄπολιν, ἐν ζῶσιν νεκρόν, 1018).75 These dramatic circumstances are a stimulus for a heightened level of emotions and the creation of counterfactual anthropomorphic fictions. In the absence of friends, objects become deeply longed-­for allies. Electra’s emotional projection onto the lock of hair, envisaged as a kin and a partner in mourning, eloquently illustrates the ‘flexibility and ingenuity’ of forging new bonds. The evidence from social psychology further suggests that even as a speculative fiction, an­thropo­ morph­ic thinking proves to be a good strategy to reduce the otherness of the external environment and enhance the efficacy of our dealings with it. Anthropomorphism would then be part of our cognition-­enabling system of adaptation. We know that the objects around us do not have a mind; yet treating them as if they possessed feelings or speech fosters more appropriate cognitive responses and emotional adjustment than treating them as external materialities without anything in common with us.76 This cognitive gain is perhaps more difficult to detect in the scenes we have discussed. Still, it seems that Electra’s affective and cognitive engagement with the lock of hair precedes (and in some ways, foreshadows) her later effusive reunion with her brother. The same could be said of Philoctetes and his bow, later returned by Neoptolemus, and of Heracles, who eventually accepts Theseus’ offer of friend­ ship, support, and asylum. In these cases, one may say that this make-­believe

73  Cairns 2011a, 16–19. 74  Phil. 268–9, 280–4, 470–1, 486–7, etc. 75  According to the testimony of Dio Chrysostom (52.8), who, in his time, was still able to compare the Philoctetes plays of the three canonical tragedians, this solitude might even have been a dramatic feature enhanced by Sophocles. While he represented Lemnos as a wild and hostile place devoid of human life, in the lost play by Euripides, the island was inhabited by Actor, a Lemnian who pitied Philoctetes and brought him assistance. 76  Guthrie 1993, Gell 1998, 121–2.

132  Anne-Sophie Noel cognitive and affective partnering with objects comes with a social benefit: through its anthropomorphic features, it is capable of fostering favourable conditions for a later renewal of human bonds.

7.5 Conclusion For many, the familiar presence of things is a comfort . . . . Their long association with us seems to make them custodians of our memories; so that sometimes, as in Proust, things reveal us to ourselves in pro­ found and unexpected ways. Yet all this does not mean that things reveal themselves, only our investments in them.77 In spite of the limitations that I have outlined over the course of this chapter, the heuristic scope of my investigation has been twofold: to refine our understanding of the cognitive contribution of some famous objects of Greek tragedy, and to revise Malafouris’ concept by expanding it to this specific corpus of case studies. Greek tragedy does indeed allow in-­depth analysis of the rhetorical and fictional dimension of the ‘cognitive life of things’. Tragic characters express an awareness of forging a fiction when they imagine that objects can be their thinking partners. In the plays that I explored, the ‘cognitive life of things’ may be better called a make-­believe fiction in which things are momentarily envisaged as human ‘minds on stage’. This shift through fiction and dramatization is in line with current accounts of anthropomorphism as one of the consoling and adaptive mechanisms of the human mind. Ancient Greek tragedies give us insights into our tendency to ‘lean to’ objects, and their puzzling revealing power. But it is the characters’ ‘investments in them’ (as Peter Schwenger puts it), both emotional and cognitive, that turn them into emotive and intelligent objects.

Acknowledgements I address my warmest thanks to Ineke Sluiter and Felix Budelmann for having invited me to make a late contribution to this volume. I also greatly benefited from their judicious critique and incisive comments. I am indebted as well to Al Duncan, to esteemed colleagues gathered for a workshop on Greek tragedy at Northwestern University (Marianne Govers Hopman, Sarah Nooter, Jonah Radding, Angeliki Tzanetou) and anonymous readers, for their very valuable feedback.

77  Schwenger 2006, 3.

PART III

PER F OR MA NC E , SPE CTAT I NG , A N D C O GNIT ION

8

Spectating Ancient Dramas The Athenian Audience and its Emotional Response Hanna Gołąb

8.1 Introduction In this chapter I aim at assessing and adapting one of the most successful cognitive theories to the analytical framework of spectating in ancient theatre. I argue that the theory of conceptual blending can serve as a useful study tool of the cognitive processes of ancient audiences, and that building on this theory we can better account for the affective power of ancient dramas.1 This problem is known more broadly as the paradox of fiction, that is, the question why people can ex­peri­ence strong emotions that are caused by an openly fictional story, and a satisfactory answer is yet to be provided. While I do not claim to have a universal answer to that problem, I show that the cognitive approach can complement and refine our understanding of the phenomenon that is customarily called in classical studies ‘generic expectations’2 and the role of emotions in those mental processes. Building upon previous scholarship on metatheatricality in ancient theatre, I argue that in the case of theatre in classical Athens the engagement of audiences in dramatic narratives was an active process to a degree that we have not yet suspected. My dynamic model of spectating stands in opposition to the widely used paradigm of suspension of disbelief, and demonstrates that there were at least two basic modes of spectating contingent on the theatrical genre being presented. The tragic gaze generally favoured immersion in the events (the degrees of which varied from tragedy to tragedy and staging to staging) and, subsequently, the immersion encouraged certain affective responses which we know well from ancient literary theory. The comic gaze was more interactive, contextualizing, self-­reflecting, and amusement-­oriented.3 Those features had to compensate for an increased distance to the comic narrative. 1  This is not the first time that conceptual blending is adapted to the field of classics, see e.g. Pagán Cánovas 2011, Budelmann and LeVen 2014, Bonifazi 2018 and 2019. 2  On generic expectations, see also the chapter by Seth Schein in this volume. 3  Our sources on satyr drama are so exiguous, that I am not able to make any conjectures on the audience’s attitude to this genre, but for the possible relationships between satyr plays and their

Hanna Goła ̨b, Spectating Ancient Dramas: The Athenian Audience and its Emotional Response In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0008

136  Hanna Gołąb

8.2  Theory of conceptual blending The conceptual blending theory was first formulated by Gilles Fauconnier and Mark Turner in a series of publications, but most notably in their 2002 monograph titled The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities.4 In it, they built an interpretative framework of how humans create new conceptual constructs which contain information from two or more different sources. In their model Fauconnier and Turner assume the existence of mental spaces, which are used during the cognitive process of blending. By ‘mental spaces’ they mean small packages of concepts that consist of elements activated simultaneously as single integrated units (for example the mental space of brushing your teeth can consist of the physical space of the bathroom, the repetitiveness of the activity, and the motion itself). The process of blending of those mental spaces into a new construct is not a unidirectional mapping of a source space onto a target space, to which George Lakoff and his followers subscribed.5 Instead, Fauconnier and Turner want to see it as a multi-­directional exchange of information from multiple spaces or sources, which converges in a resulting blended space, creating in this way a new cognitive construct. This process of blending should be thought of as a web of elements, contexts, assumptions, and interferences (called by them integration networks) rather than a neat geo­met­ ric­al model.6 The authors often emphasize that in their opinion this process stands behind any creative and artistic activity of humans.7 Fauconnier and Turner assert that humans create and use such conceptual blends in language and creative thinking from early childhood. For instance, talking animals in children’s stories are a blend of a zoomorphic body and both

spectators see Griffith 2002. The subject of metatheatricality generated excellent contributions to the study of ancient dramas, although from a customary point of view of literary texts, not that of the audience, see e.g. Taplin 1986, Bain 1987, Gutzwiller 2000, Slater 2002, among many others. Taplin’s description of comedy that invades the world of the audience and tragedy that binds a spell (Taplin 1986, 164) is a characteristic elision of spectators’ agency in the accounts of metatheatre. 4 Turner and Fauconnier 1995, Turner 1996, Fauconnier and Turner 1996, Fauconnier 1997, Fauconnier and Turner 1998, 2002, 2008. 5  In the classical formulation of Lakoff and Johnson 1980, conceptual metaphors such as ‘life is a journey’ are a mapping of the concept of journey (source domain) onto the concept of life (target domain), which results in the concept of life organized accordingly to our concept of journey. This approach is still a useful tool in close readings of metaphors in classical literature, see the chapter by Michael Carroll in this volume. 6  Integration networks are succinctly described in Fauconnier and Turner 2008, 53–4. 7  Their hypothesis was received quite enthusiastically, but it has to be noted that there are some issues that cognitive theories at large have to face: Glebkin 2015 points out that even though conceptual blending assumes airs of a scientific theory, it is hardly falsifiable and as such remains just an interpretation of cognitive processes. Furthermore, Glebkin claims that mental spaces have an unclear ontological status—­are they theoretical abstracts or actual cognitive processes? Perhaps Glebkin’s strongest critical remarks are that the theory cannot provide for emergence of a completely new knowledge that is not a result of blending and that the process mostly describes how a problem might be represented (a visual simplification).

Spectating Ancient Dramas  137 animal and human behaviour. Aesop’s fable of the ant and the grasshopper describes a hard-­working ant and a careless grasshopper—­one collects food for winter, while the other idly hops around. Our mind takes from two separate concepts of the animal and the human any information it needs to complete the blend and to create a new image, in this case one that is impossible in reality. The result, however, is not a purely visual representation—­the conceptual blend is more complicated than that. It allows us to ethically judge the grasshopper for being too reckless, something that itself is possible only within the logic of the peculiar animal-­human blend.8 Even though Fauconnier and Turner do not focus on theatrical experiences, it is clear that they think that similar processes are in play in the realm of dramatic performances just as in any other human creative activity.9 Following them, scholars working in the field of performance studies tried to apply conceptual blending in their theories of dramatic performance.10 In simple terms, the theory deepens our understanding of how performers retain their defined selves as e.g. Athenian citizens and, at the same time, enter a performed role with its characteristic peculiarities. On stage we see a blend of a performing body ruled both by the individual’s will and the requirements posed by the fictional character he is playing.11 This process constitutes the ‘doubleness’ of the theatre—­when looking at an actor we see both the actor and the tragic hero.12 Within the same interpretative framework we can also easily grasp how the audience recognizes the change from an Athenian citizen to the role he is playing, and of telling apart different personae during the performance, even if they are played by the same actor.13 Bruce McConachie, a forerunner in using cognitive approaches in per­form­ ance studies, emphasizes that in the case of theatrical spectators the conceptual blending framework allows for different perceptions of the same blends. For example, depending on the degree of familiarity with the performer, an observer might be able to discern gestures or tone of voice that are characteristic of this individual, while for another spectator this unknown actor is only his enacted persona.14 For this very reason the emperor Nero could never be treated as a regu­lar actor. Aside from the fact that his talents were apparently rather meagre, his real-­life character was too strongly rooted in the audience’s minds. This is why, during the staging of the Hercules Furens, when the emperor entered the stage bound in fetters, a soldier ran to his assistance. It does not make any difference if 8  Turner 1996, esp. 11 and 57–61, provides an interesting insight into such fables. 9  Fauconnier and Turner 2002, 266–7. Turner 2016 pays more attention to performance and the­ atre, but in rather general terms. 10  See e.g. Blair 2008, McConachie 2008, Cook 2010. 11  Blair 2009, 94. Dancygier 2016, 24 adds that the acting body and theatrical costume is a material blend that belongs both to the story world and the reality. 12  This phenomenon is similar to what art historians call ‘seeing double’: when looking at a painting an observer sees both the canvas and the imagery, see Maynard 1994. 13  Turner 2016, 69. 14  McConachie 2008, 40–7.

138  Hanna Gołąb the soldier’s reaction was a true naiveté or a calculated demonstration of faux-­ innocent loyalty—­both interpretations show the difficulty of a perfectly balanced blend in the case of a well-­known powerful individual. The mental space of real-­ life Nero must have provided much more input to the blended space of Nero-­ Hercules on stage than the mental space of the mythological hero. This imbalance must have been additionally amplified by the fact that the tragic mask of Hercules most likely imitated Nero’s own facial features.15 From our ancient sources we can get a confirmation that the ancient Greeks were aware of the doubleness of the theatre. Dicaeopolis from Aristophanes’ Acharnians reflects upon this mental process in fitting words: ‘Today I have to look like a beggar. I must be who I am, but not appear so’ (δεῖ γάρ με δόξαι πτωχὸν εἶναι τήμερον, | εἶναι μὲν ὅσπερ εἰμί, φαίνεσθαι δὲ μή, 440–1).16 In Aristophanic comedy paradoxical blends of mutually exclusive identities are often a source of entertainment and laughter. In the opening scenes of Aristophanes’ Ecclesiazusae, Praxagora dresses and trains her companions to behave like men and sneak into the Assembly. The staged cross-­dressing is so entertaining, because in reality they are men playing women playing men. Their fake beards and hair had to be attached to female masks hiding their real beards and hair, thus creating a kind of a ‘matryoshka doll’ holding different personae inside itself. The incongruities between the various layers, or in other words a failure of perfect blending, evoke laughter when female gender-­determined behaviour (such as using women’s speech) contaminates the seemingly male surface.17 15 Suet. Ner. 21.3: ‘he also sang tragedies, wearing masks of heroes and gods, and also of heroines and goddesses. Those masks were shaped in resemblance to his own features and features of a woman, whom he favored at the moment. Among others he sang Canace in Labour, Orestes the Matricide, Oedipus Blinded, and Hercules Mad. The rumour says that during the last performance a young soldier put on guard of the entrance saw Nero in a costume and bound in fetters, as the plot required, and ran to his assistance’ (tragoedias quoque cantauit personatus heroum deorumque, item heroidum ac dearum, personis effectis ad similitudinem oris sui et feminae, prout quamque diligeret. Inter cetera cantauit Canachen parturientem, Oresten matricidam, Oedipodem excaecatum, Herculem insanum. In qua fabula fama est tirunculum militem positum ad custodiam aditus, cum eum ornari ac uinciri catenis, sicut argumentum postulabat, uideret, accurrisse ferendae opis gratia). The mask resembling the face of the starring actor is a perplexing experiment and poses the question of whether Nero actually wanted to create a perfect blend of Hercules on stage; or maybe he did not wish to fully suppress his identity to begin with. An anonymous reviewer of this chapter kindly suggested a comparandum recounted by Stendhal (1928, 17): during a staging of Othello in Baltimore, a soldier shot the leading actor who was about to kill Desdemona. 16  Lada-Richards 2002, 396 writes about this passage that ‘[l]ike a seasoned actor, the Aristophanic Dicaeopolis reflects self-­consciously upon this twofold way in which the elements of “actor” and of “character” can co-­exist in a performer’s stage presence’. It is only a small, conceptual as it were, step between her analysis and the theory of blending. 17  For example in lines 296–9c (ed. Henderson): ‘after we get the token, let us men sit close to each other in order to vote in favour of all the things necessary for our dear ladies – but what am I saying? I should have said our dear lads!’ (ὅπως δὲ τὸ σύμβολον | λαβόντες ἔπειτα πλησίοι | καθεδούμεθα ὡς | ἂν χειροτονῶμεν | ἅπανθ᾽ ὅποσ᾽ ἂν δέῃ | τὰς ἡμετέρας φίλας | – καίτοι τί λέγω; φίλους | γὰρ χρῆν μ᾽ ὀνομάζειν). Vase iconography shows that there was a sense of a real identity hidden behind the comic masks and costumes. Perhaps the most interesting scene comes from an Attic bell krater dated to 370/360 bce (Heidelberg  B.134, Trendall 1959, no. 7), which shows two comic actors in female garments and

Spectating Ancient Dramas  139 Of course, blended identities of the performers are not the only dramatic elem­ent that the cognitive tool of conceptual blending helps us interpret. The theatrical space is similarly transformed into whatever a given drama requires, be it a Trojan camp in Pseudo-­Euripides’ Rhesus or an Argive palace in Aeschylus’ Agamemnon. The place of action is often mentioned in the prologues, together with significant landmarks or buildings, in order to fill in the gaps in the audience’s knowledge, and is often referred to in the course of performance to re­inforce the theatrical illusion. This practice prompts the audience to mentally construct the blend of the physical space of theatre and the performed space of drama. It is worth adding that not only the visible space was blended with the imagined one. The performed space extended to the imaginary interiors of palaces and faraway cities or oracles outside the field of vision of the spectators. Later, in the post-­classical theatre, this conceptual blend of physical and mental space eventually became so conventional that it governed the stage movement of the actors and consequently its decoding by the audience—­the left parodos led to the city of the dramatic narrative, while the other one to the harbour and countryside.18 This, naturally, is just a simplified description in broad strokes of the cognitive processes that govern our perception of the theatre. There are different kinds of conceptual blends and also different aspects of blending (e.g. composition, completion, elaboration, etc.).19 All of them create unique ways of spectating, but, as Amy Cook eloquently puts it, the power of conceptual blending as applied to per­ form­ance arts lies not in the taxonomy, but rather in how it uncovers the ceaseless flux of possible mental spaces and their connections.20 The framework of blending, moreover, allows us to gather together elements of spectating that until now were considered separate phenomena. Both blended identities and blended space when embedded in a dramatic narrative combine to produce an experience of performance that McConachie called a subjunctive reality.21 It is a temporary space, which adheres to its own rules and logic. This phenomenon is not the simple suspension of disbelief to which literary criticism is accustomed. It requires the audience’s assistance in creating the performative space. The onlookers do not willingly suppress any of their conscious or subconscious knowledge about external reality. On the contrary—­many elem­ ents of external life are crucial in providing the structure for the new mental space. The audience does not need to be explicitly told every time that in a fictional dramatic narrative severely wounded people tend to die. They masks, but one of them in a dancing pose lifts his mask up and shows his real face. Young 1991, 300 also reports a vase that shows a comic scene, in which a captured Antigone removes her wig and reveals the bald head of the male actor. 18  Poll. 4.126–7; Vitr. 5.6.8; Kocur 2001, 194–5. 19  For the full list of elements of blending see Fauconnier and Turner 2002, 44–50. 20  Cook 2007, 584. 21  McConachie 2011, esp. 38–43.

140  Hanna Gołąb subconsciously assume that fact, confirmed by their daily experience, until they are told that a wounded hero will not die, unless he is shot in his heel. This way arises the so-­called emergent structure of the fictional narrative—­an internal framework that not only keeps being updated with more information from new/other mental spaces and blends but also develops its own logic which was not an original element of its input spaces.22 The information from the blended space can then be transferred back to the input spaces, so next time a person hears ‘Achilles’, the fatal heel can become an inherent part of this mental image. Coleridge’s suspension of disbelief assumed that a reader or a spectator would have to repress their cognitive faculties in order to enjoy the fictional narrative. It seems, however, that the actual psychological phenomenon is far more complex: one’s cognitive faculties are still employed to the same degree, if not more, when spectating a dramatic performance. Entertainment and enjoyment do not depend on repressing of one’s (dis)beliefs.23 At least in the case of theatrical performances, moreover, the subjunctive reality can extend its reach over the spectators, who modify their behaviour in order not to interfere with the performed narrative or to participate in it according to the performance’s rules. Thus, bodily movements of both actors and onlookers, speech, music, space, and the response of the audience are regulated by the subjunctive reality, which exists in the participants’ minds. Fauconnier and Turner do not discuss theatrical audiences except to briefly mention that spectators ‘live in the blend’ of a performance by selective projection (which is, in fact, a principal characteristic of conceptual blending in general). This selectivity means that in the case of the onlookers many aspects of their own existence remain outside of the performed blend. According to Fauconnier and Turner, the audience is conceptually separated from the drama’s narrative, but only to a certain extent: their identity remains intact and the power of reacting to the performed narrative is limited.24 In this model, spectators should retain the capability of distinguishing reality from performed narrative. Fauconnier and Turner admit that sometimes audiences can lose the framing of themselves as spectators and, for example, rush on stage to save a hero from being murdered, but they do not try to explain this phenomenon. They also leave out the detailed questions of why and how that would happen. I want to argue that using such non-­standard emotional responses as a starting point we can build on and eventually refine their theoretical approach, at least as it applies to the theatrical audiences in ancient Greece.

22  For the emergent structure see Fauconnier and Turner 2002, 42–4, 48–9. 23  McConachie 2011, 43–4. 24  Fauconnier and Turner 2002, 267.

Spectating Ancient Dramas  141

8.3  Emotions of ancient audiences A few ancient anecdotes report a thoroughly unreserved emotional involvement of the Athenian audiences in the dramatic narratives enacted onstage. Perhaps the most notorious one comes from Vita Aeschyli 9, a short passage, which tells us of the audience’s response to Aeschylus’ Eumenides: ‘some say that during the staging of the Eumenides he brought the chorus on stage in a scattered manner, scaring people to such an extent that infants fainted and fetuses were miscarried’ (τινὲς δέ φασιν ἐν τῆι ἐπιδείξει τῶν Εὐμενίδων σποράδην εἰσαγαγόντα τὸν χορὸν τοσοῦτον ἐκπλῆξαι τὸν δῆμον ὡς τὰ μὲν νήπια ἐκψῦξαι, τὰ δὲ ἔμβρυα ἐξαμβλωθῆναι). Even though we have early modern parallels to such a strong effect of an unknown medium or a terrifying plot, the anecdote was easily dismissed by most classics scholars.25 It is not my purpose here to prove its veracity, but I want to point out that for the anonymous author of Vita Aeschyli this reaction was certainly possible, if not definitely confirmed (he distances himself from the account saying τινὲς δέ φασιν). He does not dwell on the description of the chorus members, giving us  only one detail—­they apparently entered the stage in a scattered manner (σποράδην). It is an unlikely cause of the strong emotional reaction, unless the author awkwardly meant something like an entrance not through a parodos, but from scattered spots in the audience. In any case, it would seem that this oversensitive response was not something that would require a long explanation, i.e. it was not unheard-­of. Other sources attest to similarly strong emotions. For example, Herodotus in chapter 6.21 of his Histories describes an intense feeling of sadness and mourning evoked by Phrynichus’ Sack of Miletus: The Athenians made clear their deep grief for the taking of Miletus in many ways, but especially in this: when Phrynichus wrote a play entitled ‘The Fall of Miletus’ and produced it, the whole theater fell to weeping; they fined Phrynichus a thousand drachmas for bringing to mind a calamity that affected them so personally, and forbade the performance of that play forever. Ἀθηναῖοι μὲν γὰρ δῆλον ἐποίησαν ὑπεραχθεσθέντες τῇ Μιλήτου ἁλώσι τῇ τε ἄλλῃ πολλαχῇ καὶ δὴ καὶ ποιήσαντι Φρυνίχῳ δρᾶμα Μιλήτου ἅλωσιν καὶ διδάξαντι ἐς δάκρυά τε ἔπεσε τὸ θέητρον καὶ ἐζημίωσάν μιν ὡς ἀναμνήσαντα οἰκήια κακὰ χιλίῃσι δραχμῇσι, καὶ ἐπέταξαν μηκέτι μηδένα χρᾶσθαι τούτῳ τῷ δράματι. (6.21.2, transl. A. D. Godley).

Shortly after Persians razed Miletus to the ground in 494 bce, Phrynichus ­produced his drama about the unfortunate fate of the city. The tragedy was 25  For the general dismissal of the anecdote see Calder 1988.

142  Hanna Gołąb apparently gripping to such an extent that the whole audience burst into tears (ἐς δάκρυα τε ἔπεσε τὸ θέητρον). For this blunder not only was the dramatist fined, but also the subject itself was banned from ever being staged by anyone else. For the same reason playwrights would be held accountable for the ill effects of their plays on the audience. If we believe Plutarch’s account, Solon was outraged at the ‘lies’ staged by Thespis, fearing that they would corrode the morality of Athenian citizens.26 A charge of immorality could have also been behind the failure of Euripides’ first Hippolytus, which met with a public damnatio, perhaps because of Phaedra’s explicit sexual advances towards her step-­son. The inappropriate conduct of Euripides’ heroes had to be corrected in what was a kind of dramatic palinode in his Hippolytus Garlanded.27 Even spectators who were famous for their cruelty and cold-­heartedness could break into tears, when watching a well-­enacted tragedy. Alexander, the tyrant of Pherai, was allegedly moved by the acting skills of the tragic actor Theodorus to such a degree that he had to leave the theatre, lest his tears be noticed by the rest of the audience. Plutarch’s description of this event contains particularly emotive language. According to him Alexander ‘watched the tragic actor with more empathy and because of this pleasure he was moved to pity’ (θεώμενος τραγῳδὸν ἐμπαθέστερον ὑφ’ ἡδονῆς διετέθη πρὸς τὸν οἶκτον).28 Each of those passages on its own could be dismissed as a fabricated anecdote, and each of them could be rebutted because of their particularities. In the first example, Aeschylus affected only those who in ancient opinion were feeble of mind, that is, children and women, Phrynichus reminded the audience about events that actually happened, and Alexander as an individual with a political agenda could not be representative of all ancient spectators. Nevertheless, I do believe that they reflect a historical reality of the dynamics between the staged narratives and their spectators. Both Plato in his Republic and Aristotle in Poetics and Rhetoric attest to the importance of emotions in theatrical spectating.29 Some of the aforementioned reactions (the tears, the fainting) are well-­situated within Aristotle’s classical formula of fear and pity as the two main emotional responses to the tragic genre. Some of them, however, go well beyond that to the registers of disgust and outrage, which in the Aristotelian formulation should not have taken

26 Plut. Vit. Sol. 29.4–5. 27  Ar. Byz., hypothesis to the Hippolytus Garlanded: ‘this is the second Hippolytus, which is called Garlanded. It appears that it was written later, because what used to be indecent and worthy of reproach in this tragedy stands corrected’ (ἔστι δὲ οὗτος Ἱππόλυτος δεύτερος, καὶ Στεφανίας προσαγορευόμενος. ἐμφαίνεται δὲ ὕστερος γεγραμμένος· τὸ γὰρ ἀπρεπὲς καὶ κατηγορίας ἄξιον ἐν τούτῳ διώρθωται τῷ δράματι). Cf. Ar. Ran. 1043–4. 28 Plut. De Alex. fort. 334a; a similar account is in Ael. VH 14.40. See also Duncan 2005, 61–3. 29  The limitations of this chapter do not allow to fully engage with the philosophers’ stance on emotions. A thorough treatment of emotions in Plato and Aristotle, and their relation to ancient spectating can be found e.g. in Heath 1987, 11–17, Belfiore 1992, Lada 1994, Konstan 1999, Halliwell 2011, 208–66.

Spectating Ancient Dramas  143 place in theatre. As W. B. Stanford points out, emotions in general were central to the experience of ancient spectating, thus singling out only fear and pity can obscure the view rather than enlighten it.30 To further complicate matters, Plutarch’s passage adds yet another layer to the ancient emotionalism: pleasure (ἡδονή). In his description, it is ultimately empathic pleasure, not fear or pity, which brings Alexander to tears and, in consequence, secondary embarrassment and a hasty exit from the theatre. This brings us to the vexed and much-­debated ancient concept of katharsis, interpreted in many different ways—­a purgation in a medical sense, a release from emotions (but not an emotion per se), an intellectual purification, but also in some in­ter­ pret­ations a sort of affective power as well.31 Whatever katharsis might have been exactly, it indicates strong reactions to fictional narratives either because it is a strong emotion or because it purges the audiences of such affect.32 It is clear, then, that the emotional response in ancient theatre must have been an incredibly complex process, which cannot be reduced solely to Aristotelian terms. While em­pathy is a well-­recognized emotional reaction,33 there is a plethora of other possible reactions ranging from fear and pity, through sadness and anxiety, to anger, admiration, pleasure, shame, delight, and laughter, all of which could manifest themselves in an unwelcome excess. It should not come as a surprise that emotions can significantly influence our mental processes. Recent cognitive research prefers to talk of an ‘embodied mind’ or ‘embodied cognition’ rather than of a purely computational model, so popular in the early years of the discipline and still strongly influential in Fauconnier and Turner’s publications.34 Despite the fact that the two authors mention that there is a possible transfer of emotions from the input space to the blended space,35 they do not elaborate on that process. The closest they get to the paradox of fiction is their fascinating description of ‘lottery depression’—an affliction in which people who lost in a lottery suffered from loss of their imaginary life, even though they had no hope of winning in the first place.36 Even though Fauconnier and Turner 30  Stanford 2014. 31  The interpretation of katharsis as an aesthetic pleasure can be found in e.g. Schaper 1968, Ferrari 2019; and katharsis as an ecstatic kind of power in Ford 2016. 32  The intensity of those emotional responses and, on the part of writers, the need to correct what has been deemed unpleasant show that we cannot subscribe to Walton’s description of affective response as quasi-­emotions, which are only a wilful make-­believe on the part of the audience (Walton 1978 and 1990, 240–67). The main point of Walton’s solution to the paradox of fiction is that the emotions that we feel when confronted with fictional narratives, are not a ‘real’ affect, but only a pretended one in a game of make-­believe, where an object of art is a prop. This hypothesis does not explain an unwanted or unexpected emotional response, which cannot be a wilful make-­believe game of the audience. Cf. Campeggiani 2020. 33  See e.g. Lada 1994. 34  Turner 2016, 68–9 criticizes the early stages of cognitive studies and admits that he himself had been too engrossed in the computational approaches to the human mind. 35  Fauconnier and Turner 2002, 49; see also brief remarks in Fauconnier and Turner 1998, 135, 155. 36  Fauconnier and Turner 2002, 231–2.

144  Hanna Gołąb clearly allow for emotions in their model, many questions remain unanswered: how does the transfer change emotions? How does the affect change the perception of the blended space? Can emotions originate from the blended space itself? Since affect was not within the primary scope of the cognitive blending theory, those concerns have not yet been addressed in a satisfactory manner.37 Meanwhile, as we have seen from the anecdotal evidence, those factors are especially important for different modes of spectating in the ancient world. What was the reason for those conventional emotions to turn into some kind of a disturbing affect? It seems that in the case of ancient audiences the answer lies in the degree of conceptual distance of the spectators from the enacted narrative. To speak in Fauconnier and Turner’s terms, the theatre-­goers ‘lived in the blend’ of the theatrical spectacle, but to what extent exactly did their consciousness live in the full blend of the performance? Perhaps the ritual origins of Athenian drama resulted in a lack of conceptual distance between theatrical fiction and the audience from the very beginning, i.e. a staged myth was a reality for the ancient Greeks. It would mean that from the very outset the whole theatrical space was supposed to be transformed into the performed space of the myth. The followers of Mircea Eliade’s theory of sacred time would maybe agree with this statement. According to Eliade, the myth is not simply performed or re-­enacted. It is brought back as a sacred time, during which the believers are able to return to the origins of their beliefs. Thus, scholars who represent a ritualist approach to ancient tragedy would like to see ancient Greek tragedy as a ritual embodiment of myth onstage. In that process the whole space around the ritual’s participants, possibly including even the theatrical audience, would be transformed into a sacred space, and the narrative would be a myth come true, to which spectators are the witnesses. At the same time the identities of the participants could undergo a certain metamorphosis—­be it a temporary or a permanent one, as in the case of rites of passage. Both the sacred time and the sacred space would have to be a particular type of mental state and mental space in the minds of the participants. We can glimpse this kind of cognitive process in an ancient Greek mystic ritual in honour of Demeter Kidaria in Arcadian Pheneos, which very clearly displays several important features of the theatre (Pausanias 8.15.1–3): Next to the temple of Eleusinian Demeter there is a sacred place called Petroma—­two big stones well fitted together. Every other year they celebrate some kind of festival that they call Bigger Mysteries, when they open those stones. From them they take out writings that pertain to the celebration and after reading them to the initiated, put them back the same night. . . . And on top 37  Pagán Cánovas 2010 points out that even though conceptual blending theory has a potential to become a tool in the research on emotions, so far it has been underexplored.

Spectating Ancient Dramas  145 of it there is a spherical container holding inside a mask of Demeter Kidaria. During the Bigger Mysteries a priest puts it on and according to some myth beats the underworld people with staves. παρὰ δὲ τῆς Ἐλευσινίας τὸ ἱερὸν πεποίηται Πέτρωμα καλούμενον, λίθοι δύο ἡρμοσμένοι πρὸς ἀλλήλους μεγάλοι. ἄγοντες δὲ παρὰ ἔτος ἥντινα τελετὴν μείζονα ὀνομάζουσι, τοὺς λίθους τούτους τηνικαῦτα ἀνοίγουσι. λαβόντες δὲ γράμματα ἐξ αὐτῶν ἔχοντα ἐς τὴν τελετὴν καὶ ἀναγνόντες ἐς ἐπήκοον τῶν μυστῶν, κατέθεντο ἐν νυκτὶ αὖθις τῇ αὐτῇ. [. . .] καὶ ἐπίθημα ἐπ’ αὐτῷ περιφερές ἐστιν, ἔχον ἐντὸς Δήμητρος πρόσωπον Κιδαρίας· τοῦτο ὁ ἱερεὺς περιθέμενος τὸ πρόσωπον ἐν τῇ μείζονι καλουμένῃ τελετῇ ῥάβδοις κατὰ λόγον δή τινα τοὺς ὑποχθονίους παίει.

During the Phenean Great Mysteries a priest would wear a mask representing Demeter Kidaria and re-­enact some kind of a mythological narrative connected to or located in the underworld. The mask of the goddess was powerful enough to transform both the identity of the priest and the space around him. He and his audience were transferred together to the underworld, where the priest/Demeter smote with a staff either imagined inhabitants of the underworld or the audience playing their role (!).38 Thus, even though the audience seems to have had only a semi-­passive role in this cultic performance, conceptually and emotionally they were inside the subjunctive reality of the ritual, inside the blend of the underworld and visible landscape. What is even more interesting, their identities could have been to a certain extent transformed during the performance into those of the underworld dwellers. Even though it would be tempting to extrapolate from this mystic performance to the classical theatre of Athens, we see no such phenomenon among the Athenian spectators. As far as we can tell, they remain themselves throughout the staged performance, but what is even more important, they are also able to judge the technical and artistic aspects of a given drama. Were the audience completely immersed in the blended space of the performance without referring to their knowledge of the exterior reality, they would have no means of appreciating the art of actors and writers. It seems, then, that the theatrical spectatorship differed from the mystic gaze of the initiated. To reconcile the two seemingly opposite mindsets (a strong emotional involvement in the enacted narrative and the ability to aesthetically judge its execution) contemporary scholars of performance and theatre studies have attempted to situ­ate the spectator’s mind in between the two perspectives of inside and outside the performed narrative.39 Ismene Lada-­Richards in her discussion of emotions 38  In Athenaeus 14.30 (Kaibel) we have a short mention of an Arcadian dance called κίδαρις, which might have been somehow connected to this festival, see Dietrich 1962, 139. 39  The dichotomic model of inside-­outside perspective has a long history in theatre studies, see Saltz 2006.

146  Hanna Gołąb and cognition in classical audience notes that ‘Greek theatre appears to be ­constructing for itself an “implied” spectator who is both “engaged” in the fiction and capable of penetrating it, both bewitched and ready to understand the subtle interplays of representation-­levels’.40 Similarly, Julia Walker in an essay on the text/performance split in the theoretical works of theatre studies writes that ‘theater has the unique ability to shift us between . . . two perspectives by situating us both inside an imaginative fiction and outside the proscenium frame’.41 Walker analyses how Anna Quincy, a nineteenth-­century theatre-­goer, writes about a spectacle she had witnessed. Her short note displays a certain fluctuation between describing an emotional state of the protagonist and the actor’s technique used to achieve such an effect. This ‘oscillating dynamic’ was picked up by McConachie, who claims that the audience can consciously shift between inside and outside perspectives.42

8.4  Dynamic model of spectating Those scholars set up a binary scheme of fiction and reality. According to them, the audience willingly goes back and forth between the two. On the ­cognitive account presented here, however, the spectators’ minds can blend many elements of fictional and external reality at the same time. The fictional narrative already has many elements of real life, and through the dramatic experience it actually becomes a part of the audience’s reality. In other words, the theatre is con­ tam­ in­ ated with reality, and real life contaminated with ­fictions.43 The degree of ‘contamination’ will vary for each individual, but all will be creating a unique version of this blended experience. Moreover, as we have already seen, the mental spaces of the enacted narrative and the reality were not a stable construct—­they were constantly modified and updated. Sometimes, the two could even be merged on purpose (more on which below). The model itself, then, should be thought of as more flexible, mutable, and not so much dichotomous, as the audience’s perspective and involvement can change easily depending on the mental states of the spectators and techniques used by the playwrights.

40  Lada 1994, 122. 41  Walker 2006, 36. 42  McConachie 2008, 40–1. 43  I owe this formulation to my colleague Richard Hutchins. Cf. experiments reported in Gerrig and Prentice 1991 and Prentice and Gerrig 1999; they strongly suggest that people incorporate information from fictional narratives into their real-­world knowledge (e.g. after screenings of the movie Jaws many vacationers would avoid water, even though sharks very rarely attack humans or, on a more nefarious note, after the racist portrayal of Blacks in The Birth of a Nation movie, attitudes toward this minority significantly worsened).

Spectating Ancient Dramas  147

8.4.1  Interaction, contextualization, and separation In this dynamic model of theatrical spectating there is a plethora of factors that can influence the perception of a dramatic play, but I focus here on interactivity, contextualization, affect, and the degree of immersion or separation. Those elem­ ents created two different modes of spectating, which depended on the theatrical genre. For example, in comedy the media—­unnatural phalloi, grotesque masks, fantastic costumes, etc.—set the dramatic narrative apart from the audience. Comic entertainment is possible only thanks to this conceptual distance; otherwise amusement could easily turn into disgust, and what is grotesque and comically transgressive would seem monstrous and impious in ancient eyes when truly embodied.44 To keep the audience engaged and to compensate for the conceptual distance from the narrative, comedy often takes on a more interactive stance and directly addresses either current events or the spectators themselves in a gesture that is called a ‘positioning of the spectator’.45 We can see a striking example of the latter in the parabasis of Aristophanes’ Ecclesiazusae, when the chorus say: I have but a few words to say: let the wise judge me because of whatever is wise in this piece, and those who like a laugh by whatever has made them laugh. In this way I address pretty well everyone. If the lot has assigned my comedy to be played first of all, don't let that be a disadvantage to me; engrave in your memory all that shall have pleased you in it and judge the competitors equitably as you have bound yourselves by oath to do. Don't act like vile courtesans, who never remember any but their last lover. σμικρὸν δ’ ὑποθέσθαι τοῖς κριταῖσι βούλομαι· τοῖς σοφοῖς μὲν τῶν σοφῶν μεμνημένοις κρίνειν ἐμέ, τοῖς γελῶσι δ’ ἡδέως διὰ τὸ γελᾶν κρίνειν ἐμέ· σχεδὸν ἅπαντας οὖν κελεύω δηλαδὴ κρίνειν ἐμέ, μηδὲ τὸν κλῆρον γενέσθαι μηδὲν ἡμῖν αἴτιον, ὅτι προείληχ’· ἀλλ’ ἅπαντας ταῦτα χρὴ μεμνημένους μὴ ’πιορκεῖν, ἀλλὰ κρίνειν τοὺς χοροὺς ὀρθῶς ἀεί, μηδὲ ταῖς κακαῖς ἑταίραις τὸν τρόπον προσεικέναι, αἳ μόνον μνήμην ἔχουσι τῶν τελευταίων ἀεί.  (1154–62, transl. E. O’Neill 1938)

The chorus notice different experiences among the spectators and try to convince them that the author took care to accommodate them all. In an attempt to correct 44  Hemenover and Schimmack (2007) in a brief report on their psychological experiment on disgust and amusement in watching a comic clip say that the participants felt a mixed feeling of disgust and amusement when observing from an outside perspective, and a feeling of disgust, with amusement significantly decreased, when asked to take the protagonist’s viewpoint. 45  Grodal 1999, 157–8.

148  Hanna Gołąb the cognitive bias of the audience, they draw attention to their tendency to forget the plays that were shown first and give the prize to the ones played last. Such self-­conscious and metatheatrical remarks force the spectators’ minds outside the fictional narrative and focus their attention on the craftsmanship of the dramatist and his actors. In this case, moreover, the experience of a viewer is contextualized not only in the moment of seeing the play (those who are wise should praise the play’s wisdom and those who want pure entertainment should praise its jokes), but also within the whole comic agôn.The act of spectating is influenced and enriched by this contextualization, which requires the audience to extend their experience beyond the narrative frame of the play.46 Some ancient comedies encourage the audience to recall their different mental processes when watching tragedies.47 Timocles in his Women Celebrating the Dionysia (Dionysiazusae) makes several remarks on the cognition and emotion of tragic audiences, saying that ‘the mind forgets about its own problems when entranced by someone else’s sufferings; it leaves with a sense of pleasure, but at the same time it is educated’ (ὁ γὰρ νοῦς τῶν ἰδίων λήθην λαβὼν | πρὸς ἀλλοτρίῳ τε ψυχαγωγηθεὶς πάθει | μεθ’ ἡδονῆς ἀπῆλθε παιδευθεὶς ἅμα).48 In the following lines he jokingly enumerates the diverse forms of hardship afflicting one spectator or another that can be assuaged by the tragic narrative: a pauper will know that Telephus was poorer than he, a person who lost his child will be cheered up by Niobe, someone with an infected eye will think of the blind sons of Phineus, and so on. This take on ancient popular literary theory is of course parodic, but nevertheless functions as a productive contextualization, since it makes the audience think on their own about how tragedy works on their minds and, at the same time, implicitly brings up a question of the pleasures of the comedy.

8.4.2  Affect and immersion In the dynamic model of spectating that I am proposing, the emotions conventionally expected in a given genre pull the audience further into the realm of the play, that is, decrease the distance between audience and fiction.49 In tragic per­form­ances that of course means pity, fear, and various shades of similar reactions. Several psychological experiments confirm that the strength of the affective response is intimately intertwined with immersion and psychological 46  On the contextualization in cinema, see Grodal 1999, 157–8. 47  Cf. Revermann 2006, 101–3. 48  Timocles fr. 6 K–­A. 49  I specify ‘emotions as expected by the genre’s conventions’, because unexpected emotions such as, for example, amusement in tragedy, certainly would not help in engaging the audience in the narrative. An example of that happening is Eur. Or., in which the actor Hegelochus mispronounced his part ‘from the waves, I can again see a calm sea (γαλήν’ ὁρῶ)’ as ‘from the waves, I can again see a weasel (γαλῆν ὁρῶ)’. For his blunder he was mercilessly mocked in Sannyrion’s Danae (fr. 8) and Ar. Ran. (l. 303).

Spectating Ancient Dramas  149 transportation into the narrative,50 both of which are responsible for reduced ­dissociation from the narrative, increased perception of realism, and overall enjoyment of the plot.51 A similar phenomenon was recently reported by ­psychologists who conducted preliminary experiments on cinema audiences. The more that spectators were perceptually immersed in a movie, the stronger were the emotions evoked by the screening. Although the conceptual distance from the film’s narrative was decreased, the cognitive capabilities of the audience were not inhibited. The spectators were able to discern genres and generic expectations without fail, i.e. they could still read the stylistic and narrative cues, the degree of deviation from reality, and the different modes of acting. The experiment confirmed that fictional genres are inextricably connected to the basic emotions they were supposed to evoke. In other words, appropriate emotional reaction of the audience was one of the main elements of generic expectations.52 Ancient authors of tragedy were aware that this immersive effect had a strong emotive force (and vice versa) in viewing a tragic performance.53 Tragedies rarely, if ever, use asides directed to the audience or metatheatrical comments, and they try to induce the audience to immerse themselves in the performance. Actors, however, sometimes had a different idea. At the risk of bringing elements of the outside world into the realm of performance they at times made use of powerful real-­life emotions. For example, Polus, playing the title role in Sophocles’ Electra, delivered lamentations over the alleged ashes of Orestes while holding not a mere prop, but an urn with remains of his own son.54 Anne Duncan points out that since Polus was a celebrity, it is likely that the audience knew that the urn and the ashes were in fact real.55 In that particular staging, then, the two mental spaces of fictional narrative and reality overlapped to a great degree, and the reality of the actor and his audience merged with that of Electra. At the same time, the audience must have been also aware that it actually was not the ‘reality’ of the dramatic narrative, since Orestes was alive standing in front of his sister! The performed 50  The terminology of immersion, transportation, and identification is somewhat murky and the distinctions are often unclear, see Tal-­Or and Cohen 2010. Studies on psychological transportation in literature were initiated by Gerrig 1993; for immersion (both perceptual and psychological) see most notably Ryan 1999, esp. 114–20 on the role of affect, and Ryan 2015, esp. 85–116. 51  Green and Brock 2000, Tal-­Or and Cohen 2010, 405. 52  Visch, Tan and Molenaar 2010; see also Visch and Tan 2008. It should be noted that the experiment used a very basic computer-­generated action, which cannot equal a complex theatrical or cinematographic narrative. It is, however, a good first step towards our understanding of emotional response to fictional plots. 53  However, Aristotle in Rh. 1385b13–15 claims that some degree of distance is needed to ex­peri­ ence pity. Once there is no distance whatsoever, pity turns into fear, which still does fit into the general affective scheme proposed here. It seems that Aristotle constructed pity in a simple, self-­referential way: one can feel pity when one sees something that may happen to one, but does not. Heath 1987, 13–15 mentions that we should not put too much weight on such a restrictive understanding of pity, since not that many Athenians would consider themselves at risk of having sex with their own aged mothers, and yet they could feel pity when watching the fall of Oedipus. 54  Aul. Gell. NA 6.5. 55  Duncan 2005, 63–5.

150  Hanna Gołąb grief of Electra was a real grief of the star actor, and in a perverse way it was a truthful emotional reaction to a fiction within fiction. By using this un­usual prop as a material anchor for blending56 and his own intense affect, Polus managed to blend several layers of truth and fiction into an especially memorable and moving performance. His example also shows that the degree of affective response and involvement of the audience not only varied from play to play but also depended on the particular stagings of the same drama (we must hope that providing theatrical props did not necessitate a steady supply of actors’ dead sons).57 Thanks to the example of Polus, we can also recognize the dangers of such an affective tragic performance; an audience drawn in too close to the enacted narrative could overreact, just as in the case of the Sack of Miletus and other examples discussed earlier, especially if we believe Philochorus’ account that the Athenian spectators not only were drunk from the meal before the festival but also kept drinking wine during the performances.58 Drunkenness and physical exhaustion could greatly intensify the audience’s emotional reaction to the dramatic events. Additionally, in large gatherings like theatrical audiences of classical Athens crowd dynamics and emotional contagion could further intensify emotions.59

8.5  Emotions and conceptual blending theory The structural differences of comic and tragic plays that put emphasis on inter­ action and contextualization on the one hand, and immersion and affect on the other, are matched by two different approaches on the part of the audience: the comic gaze and the tragic gaze. Several scholars and psychologists working on various types of spectatorship emphasize that the emotional state of an onlooker might be a result of ‘playing along with the fiction’.60 Torben Grodal in his monographs devoted to the cognitive and affective aspects of cinema audiences underlines the importance of recognition of prototypical narrative genres in the 56  For material anchors in conceptual blending and theatre, see Dancygier 2016, 30–2. 57  We can also think in those terms about the three performances of Ter. Hec., two of which failed to keep their audience engaged. With each consecutive staging Terence had to ‘re-­position’ his spectators in the prologue and fight for their attention. 58  Athen. 11.13 (Kaibel) = FGrH 328 F 171: ‘about that Philochorus says thus: Athenians during the Dionysia first would eat and drink, and then they would go to the spectacles which they watched wearing wreaths. During the whole contest wine would be poured for them and dried fruits served, and even for the choruses that were coming in they poured to drink, and for the ones that just took part in the contest they poured to drink again when they were going out’ (λέγει δὲ περὶ τούτων ὁ Φιλόχορος οὑτωσί· ‘Ἀθηναῖοι τοῖς Διονυσιακοῖς ἀγῶσι τὸ μὲν πρῶτον ἠριστηκότες καὶ πεπωκότες ἐβάδιζον ἐπὶ τὴν θέαν καὶ ἐστεφανωμένοι ἐθεώρουν, παρὰ δὲ τὸν ἀγῶνα πάντα οἶνος αὐτοῖς ᾠνοχοεῖτο καὶ τραγήματα παρεφέρετο, καὶ τοῖς χοροῖς εἰσιοῦσιν ἐνέχεον πίνειν καὶ διηγωνισμένοις ὅτ’ ἐξεπορεύοντο ἐνέχεον πάλιν). 59  Stanford 2014, 15. 60  Walton 1978, 19; see however my caveats to Walton’s theory above. Cf. the chapter by Jonas Grethlein in this volume.

Spectating Ancient Dramas  151 emotional response of the spectators. He writes that ‘the main genre-­formulas and moods of fictive entertainment are often constructed to produce certain emotions, by allowing the viewer to simulate one from a set of fundamental emotions linked to basic human situations’ [my emphasis].61 Grodal’s statement is in line with psychological experiments which confirm that movie genres can be used as a way of mood management by the audience, and that recognition of a literary genre modifies the readers’ perception of the text and their reaction to it.62 Thus, the identification of a theatrical genre as a tragic one can actually enhance the desired affective result. In other words, the ancient tragic gaze favoured certain aspects of affect such as the well-­known sadness, pity, and fear of ancient Greek literary criticism, while the comic gaze preferred distance, con­text­ ual­iza­tion, and amusement.63 How can we situate the desired affect and emotional response in the cognitive theory of conceptual blending? In terms of subjunctive reality, we could say that the role the spectators give themselves in the blend of the performance and within the subjunctive reality is that of real-­life spectators, who are there in a particular capacity—­spectator, judge, critic, etc. Consequently, they construct themselves (or blend themselves in) in conformity with the emotional reactions expected in the particular theatrical genre.64 They do retain a sense (mental space) of reality, but the degree to which they let themselves be affected depends on the entire context and on the individual’s state and disposition. Thus, the active participation of ancient audiences in the construction of dramatic subjunctive reality was significantly higher than what scholars have anticipated relying upon the classical formulation of generic expectations. While the mental processes that are behind perception, cognition, and emotional response to a fictive narration are a human universal, the particular genres and the specifics of the expected affective reaction depend on the cultural, historical, and social factors.

8.6 Conclusion To conclude, conceptual blending theory can very well expand our understanding of how the spectating mind creates dramatic narrative, personae, and space in different ways for each dramatic form, all the while acknowledging different ways of blending by different spectators. In the form presented by Fauconnier and

61  Grodal 1999, 161. 62  Visch and Tan 2008, esp. 302, Zwaan 1993. 63  The question of why the audiences of tragedies, melodramas, and other types of ‘sad’ genres expect to feel unhappy and in a way revel in this feeling is beyond the scope of this chapter but considering the universal nature of this phenomenon, it is likely that it has an evolutionary explanation, see Grodal 2009, 122–42. 64  This formulation stands in contrast to those of psychologists who attempt to describe cognitive phenomena without discerning the genres of the perceived narrative, see e.g. Green and Brock 2000.

152  Hanna Gołąb Turner, however, it does not embrace all the degrees of an audience’s emotional involvement in the performance. Emotions and affective response should become a significant part of the theory, if it aims at describing the cognitive processes behind the perception of theatrical dramas. I argued that the theatrical spectators are placed in a precarious balance between the two perspectives of inside the narrative and outside it, but it is not a simple binary—­a theatrical performance was not only a constant negotiation between the two viewpoints, but at any given moment it also included elements of both mental spaces. Thus the model should be thought of as a fluid and dynamic process (as Fauconnier and Turner say themselves) with emotions steering the rest of cognition.65 Furthermore, I refined the customary approach of generic expectations, proposing that the involvement of ancient audiences in the spectacle in front of them was a more active and dynamic process than we had suspected. Ancient spectators modelled their own emotional response that was contingent on the genre of the dramatic per­form­ ance. They consciously accommodated the conventions of dramatic genres, placing themselves within the subjunctive reality of the performance, which moderated their response, but outside its fictional narrative, which allowed them to retain critical distance and a critic’s eye. Since the tragic gaze encouraged immersion and an emotional response such as pity and fear, and, as it seems, ancient techniques of acting emphasized the emotions and the audience was often drunk, we hear of excessive reactions of this type in the ancient theatre. The comic gaze, meanwhile, was more contextualized and self-­reflective. It provided a conceptual distance from the enacted fiction, which was necessary for the enjoyment of the play.

Acknowledgements This chapter underwent many changes and revisions, at each stage benefitting from the generous feedback and comments of its readers. My heartfelt thanks go to Ineke Sluiter and Felix Budelmann—­the masterminds behind the ‘Minds on Stage’ conference and the present volume—­anonymous reviewers, my mentor Andrew Ford, and my colleagues from Princeton classics department. Needless to say, all shortcomings that remain are my own.

65  For this fruitful formulation I would like to thank the chapter’s anonymous reviewer.

9 Gorgias’ ἀπάτη, Sophocles’ Electra, and Cognitive Criticism Jonas Grethlein

9.1 Introduction One of the main themes of Sophocles’ Electra is deceit. In the prologue, Orestes instructs his old teacher to go to the palace of Mycenae and to announce ‘that Orestes is dead from a fatal accident, tossed from his racing chariot at the Pythian Games’ (ὁθούνεκα | τέθνηκ᾽ Ὀρέστης ἐξ ἀναγκαίας τύχης, | ἄθλοισι Πυθικοῖσιν ἐκ τροχηλάτων | δίφρων κυλισθείς, 48–50). The long messenger speech, in which the Paedagogus gives a detailed account of Orestes’ death to Clytaemnestra and Electra, stands at the centre of the play and forms its ‘hinge’.1 In the second half of the play, Orestes himself enters the palace of Mycenae with an urn that—­so he pretends to Electra—­contains the ashes of Orestes. It is only then, after more than 1,200 verses, that Orestes discloses his identity to Electra and kills his mother. The postponement of his anagnôrisis and the late enactment of the matricide create ample space for the deception and make it the central aspect of Electra’s plot. While some scholars see Orestes’ reliance on a ruse as justified by an oracle of Apollo (36–7),2 others argue that, despite the oracle, his strategy is problematic and aligns him with sophistic figures such as Odysseus in Philoctetes3 and even creates a parallel to Clytaemnestra, who relied on deception in her murder of Agamemnon (Soph. El. 124–6).4 In this chapter, I will leave this and other much-­ debated questions aside and approach the deception in Electra from a new angle. For this, I will combine an ancient piece of reflection on tragedy, Gorgias fr. B 23 DK, with modern cognitive models of aesthetic experience in general and of reader response in particular. My chapter thus triangulates ancient literature and ancient criticism with cognitive studies. I do not claim that these three fields fully

1  Segal 1966, 479. 2  E.g. Parlavantza-­Friedrich 1969, 39, MacLeod 2001, 30–1, March 2001, ad 59–61. 3  E.g. Kells 1973, 4–12, Winnington-­Ingram 1980, 236, Segal 1981, 261. Hesk 2000, 199 contrasts Orestes’ lie with Odysseus’ deceit in Philoctetes, arguing that the former is ‘unequivocally dark and callous’. Against such interpretations, see Finglass 2007, ad 59–66 and 61. 4  See also 197, 279. Cf. Segal 1966, 510–11, Kells 1973, 6, Segal 1981, 261, Schein 1982, 72–3, Seale 1982, 57. Jonas Grethlein, Gorgias’ ἀπάτη, Sophocles’ Electra, and Cognitive Criticism In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0009

154  Jonas Grethlein map onto each other, and yet I believe that the results of cognitive research can help to flesh out Gorgias’ understanding of tragedy as ἀπάτη, and that the two taken together are able to yield a fresh perspective on Electra, notably the false messenger scene. I do not presuppose that Sophocles is in any way responding to Gorgias: the insecure dating of both texts makes this hard to argue.5 Rather, Gorgias’ reflection on tragedy will serve as a heuristic framework for sharpening our understanding of Sophocles’ play. In section 9.2, I will introduce Gorgias fr. B 23 DK, rarely taken seriously, if taken notice of at all, in modern aesthetics, and show that its premises correspond with insights of cognitive accounts of aesthetic experience. Then, in the third and fourth sections, I will reconsider Electra in the light of Gorgias fr. B 23 DK, suggesting that the false messenger speech stages the entwinement of deception and aesthetic illusion with which Gorgias plays. In section 9.3, I will first show that the messenger speech is geared towards generating a strong aesthetic illusion; an enactive model of reading will help me analyse the experiential quality of the speech. As I will then go on to argue, the very devices that serve to engross the theatre audience are simultaneously a means of deception within the play. As pointed out in the fourth and final section, this association of aesthetics with ­ethics reveals a specifically ancient approach and sheds light on the limits as well as the benefits of employing cognitive studies in classics.

9.2  Gorgias fr. B 23 DK and cognitive approaches to aesthetic experience Plutarch twice quotes Gorgias’ dictum that ‘tragedy induces a deception, in which the one who deceives is more just than the one who does not deceive, and the one who is deceived is more intelligent than the one who is not deceived’ (ἡ τραγῳδία  . . . παρασχοῦσα . . . ἀπάτην, ἣν ὃ τ᾽ ἀπατήσας δικαιότερος τοῦ μὴ ἀπατήσαντος, καὶ ὁ ἀπατηθεὶς σοφώτερος τοῦ μὴ ἀπατηθέντος).6 We do not know from which work the quotation is taken, nor even whether it was an isolated quip or formed part of a poetological treatise, as older scholarship was keen to believe.7 If one assumes that it was an acknowledged task of tragedy to improve the audience morally, then the comment is based on a chiastic inversion: on the one hand, σοφία, which is not exclusively, but especially, associated with poets,8 is assigned to the

5  Cf. the sober assessment of the evidence for the dating of Electra by Finglass 2007, 1–4. 6  De glor. Ath. 5.348c; Quomodo adul. 2.15d. Some scholars argue that the passage in De glor. Ath. cites verbatim further words of Gorgias (e.g. Dosi 1968, 37–8), but Diels-­Kranz, Laks and Most, and the majority of scholars limit the citation to the one given above. 7  See especially the influential Pohlenz 1965, 436–72 (originally published in 1920). 8  On the poet as σοφός, see e.g. Maehler 1963, 94–6, Gianotti 1975, 85–107, Stenger 2004, 326–34. For a healthy note of caution against identifying all σοφοί in Pindar with poets, see Most 1985, 144–5.

Gorgias ’ ἀπάτη, Sophocles ’ Electra, and Cognitive Criticism  155 spectators; on the other hand, the justice that the audience is supposed to learn becomes an attribute of the poet. The result is a double paradox—­how can deceit be just, and how can being its victim a sign of wisdom? The fragment plays with several traditions and topoi.9 The most obvious and prominent of these is the claim that poets lie, a notion which can be traced back to the sneering declaration of Hesiod’s Muses that they ‘know how to speak many falsehoods that are like true things’ (ἴδμεν ψεύδεα πολλὰ λέγειν ἐτύμοισιν ὁμοῖα, Theog. 27). Gorgias evokes this topos and gives it a positive twist: yes, poets lie, but their deception is just. This twist is made possible through a new meaning of ἀπάτη, namely aesthetic illusion.10 In generating an aesthetic illusion, the poet does what a poet is meant to do and is thus ‘just’. This, however, does not mean that δίκαιος in fr. B 23 DK lacks an ethical connotation and, as some scholars have claimed, merely signifies what is appropriate, befitting, or normal.11 On the contrary, Gorgias uses the word ‘just’ consciously and thereby reinforces his play with the basic meaning of ἀπάτη; couched in ethical terms, the comment on aesthetics becomes paradoxical. The ethical semantics of ἀπάτη and δίκαιος also render the reflection on the recipient puzzling—­σοφία is not only an antidote to deceit but, what is more, is characteristic of the deceiver rather than the victim.12 Plutarch is on the right track when he explains the meaning beneath this paradoxical ring: ‘. . . the one who is deceived is more intelligent, for whoever is not insensible, is captured by the pleasure of words’ (. . . ὁ δʼ ἀπατηθεὶς σοφώτερος· εὐάλωτον γὰρ ὑφʼ ἡδονῆς λόγων τὸ μὴ ἀναίσθητον, De glor. Ath. 5.348c). Σοφία is not the result of aesthetic illusion, but the requirement for its emergence.13 In reflecting on the recipient and his role in the success of ἀπάτη, Gorgias considers a key aspect of aesthetic illusion. As cognitive as well as phenomenological studies argue, aesthetic illusion is predicated on a special mode of response: the recipient needs to engage in an ‘as-­ if ’, which combines immersion with reflexive distance.14 The recipient immerses herself into the represented world, but at the same time remains aware of its representational status. Both aspects need to come together for aesthetic illusion to emerge: if a recipient concentrates only on the representation, say, the brushstrokes of a painting or the sound pattern of a poem, she does not experience aesthetic illusion. At the same time, full immersion into the represented 9  For a full interpretation of Gorgias fr. B 23 DK in the light of the topoi it evokes, see Grethlein (2021, 1–32). 10  The majority of scholars understands ἀπάτη in Gorgias fr. B 23 DK as aesthetic illusion. Against this, see, however, Rosenmeyer 1955, 234, Dosi 1968, 40–50, Garzya 1987, 253–6, Sier 2000, 602–10. 11  This is argued by e.g. Nestle 1940, 320, Rosenmeyer 1955, 227, Rösler 1980, 311n80. 12  See, for example, Pind. Nem. 7.20–3. 13  Pace Taplin 1978, 167. 14  From a cognitive perspective, see Schaeffer 1999, 133–230; from a phenomenological perspective, see Walton 1990; in literary criticism, see esp. Wolf 1993 and Wolf 2013. For the highly nuanced reflections on the balance between immersion and reflection in ancient art and literature, see Grethlein 2017.

156  Jonas Grethlein world, as when Zeuxis asks Parrhasius to draw the curtain on his painting (see Plin. HN 35.65–6), is no longer a case of aesthetic illusion, but of delusion. While the degree of immersion may increase and decrease, there is constantly a residual degree of distance, which is inherent in the attention given to the representation qua representation. A reader may concentrate on the story and have the feeling of witnessing the action narrated, but the act of reading, the book she holds in her hands, and the letters she has to process, make it impossible for her to enter fully into the represented world. In the same vein, the stage on which actors operate, like the presence of others in the theatre, sustains the spectator’s awareness that she is watching a play, no matter how strongly she is taken in by the action. A comment on the ἀπάτη of drama, which seems to be contemporary and was perhaps influenced by Gorgias fr. B 23 DK, makes it explicit that spectators remain aware of the mediating instance. When the Hippocratic treatise De victu discusses deception as an art and as an essential part of human nature, it also adduces the art of acting, ‘which deceives those who know; they [i.e. the actors] say one thing and think another thing, they enter as the same and are not the same when they leave’ (ὑποκριτικὴ ἐξαπατᾷ εἰδότας· ἄλλα λέγουσι καὶ ἄλλα φρονέουσιν, οἱ αὐτοὶ ἐσέρπουσι καὶ ἐξέρπουσιν οὐχ οἱ αὐτοί, De victu 24).15 Εἰδότας fully captures the audience’s irrepressible knowledge of attending to a representation. This awareness is also gestured to in the σοφία which Gorgias ascribes to the person who undergoes an aesthetic illusion. While referring to the reflexive distance less directly than εἰδότας in De victu, σοφία chimes well with the notion of a special and culturally refined mode of attention that is essential to the idea of aesthetic illusion. Representations such as theatre, pictures, and novels are cultural products, and, while the capacity for make-­believe seems to form in early childhood,16 the habits and understanding required for their proper reception is something that we acquire through education. Scholars tend to complement Gorgias fr. B 23 DK with Helen, the epideictic defence of Helen, who, Gorgias argues, cannot be blamed for the Trojan War. This is not unproblematic. Helen has an obviously different scope from fr. B 23 DK; it discusses speech in general and focuses on its persuasive force, whereas fr. B 23 DK addresses specifically tragedy and the aesthetic illusion it is capable of generating. Moreover, Helen’s thesis that the force of speech is irresistible clashes with the notion in fr. B 23 DK that not everybody is deceived and that those who are deceived are wiser than those that are not.17 If we keep these differences in mind,

15  De victu is generally dated at the end of the fifth and the beginning of the fourth centuries; cf. Joly and Byl 1984, 44–9. A further striking parallel to Gorgias fr. B 23 DK is found in the anonymous Dissoi Logoi 3.10. 16  Walton 1990, 11–12. 17  It also ought to be noted that ἀπάτη in Helen has only ethical significance and straightforwardly means ‘deception’.

Gorgias ’ ἀπάτη, Sophocles ’ Electra, and Cognitive Criticism  157 however, it is possible to flesh out fr. B 23 DK with the help of Helen. Pertinent is the reference to poetry, which Gorgias adduces to illustrate the power of speech: Those who hear it are penetrated by a terribly fearful shuddering, a much-­ weeping pity, and a yearning that desires mourning, and on the basis of the fortunes and misfortunes of other people’s affairs and bodies, their soul is affected, by an affection of its own, by the medium of words. 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.  (fr. B 11.9 DK)18

It is hard to avoid seeing the effects of tragedy in the response that Gorgias describes.19 Scholars have duly noted that φρίκη περίφοβος and ἔλεος πολύδακρυς prefigure the emotions that both Aristotle and Plato, albeit with opposite verdicts, assign to spectators in the theatre.20 I propose to read the discussion of the effects of poetry in Helen as an elaboration on the immersive aspect of ἀπάτη mentioned in fr. B 23 DK. While most modern commentators home in on the emotions that poetry elicits, the key aspect for my interpretation is the claim that words about the fate of others trigger in the soul ‘an affection of its own’. Gorgias ascribes a very strong effect to what, after all, is only a verbal representation. The intensity of the audience response is highlighted by the figura etymologica of πάθημα . . . ἔπαθεν, which underlines the point that narration is capable of producing a πάθημα just as actual events do. This force of the representation is tangible in the physiological dimension of its resonance in the recipient, who is said to shudder and to weep.21 Perhaps most importantly, ἴδιον is markedly juxtaposed with ἀλλοτρίων: the fortunes and misfortunes concern other people, and yet, they elicit an ‘affection of its own’ in the soul of the spectator. Here, Gorgias reconfigures a view that can be traced back to the Odyssey. When Demodocus sings about Odysseus’ experiences, only the Phaeacians enjoy the recital whereas Odysseus breaks into tears. Terpsis is generated only by the account of the experiences of others, notably when they include suffering.22 As reported by Herodotus, Phrynichus was fined because his Sack of Miletus confronted the Athenian audience with οἰκήια κακά (‘their own evils’, 6.21.2). In a fragment from Timocles’ Dionysiazusae, a speaker 18  On the role of poetry in Helen, see e.g. Halliwell 2011, 275–7, which also brings in fr. B 23 DK, but ignores the differences between the two texts. 19  E.g. Dalfen 1974, 273, Taplin 1978, 168. Halliwell 2002, 212–14 and 218 notes that Aristotle’s psychology of tragedy, notably the prominence of pity, relied on firmly established notions. 20  E.g. Pl. Ion 535c5–8; Arist. Poet. 1449b27. See, for example, Segal 1962, 131–5, Munteanu 2012, 41–2. On Plato and Gorgias, see e.g. Flashar 1958, 68–72. On Aristotle and Gorgias, see e.g. Schadewaldt 1955, 144–5, 164–5, Bona 1974, 26n6. On φρίκη, see Cairns 2017b. 21  Cf. Segal 1962, 106. 22  On the question of why Odysseus nonetheless asks Demodocus specifically to sing about the Wooden Horse, a topic that will drive him to tears again, see Halliwell 2011, 77–92.

158  Jonas Grethlein praises the effect of tragedy: ‘for forgetting his own and being transported to alien suffering, the mind returns with pleasure and simultaneously educated.’ (ὁ γὰρ νοῦς τῶν ἰδίων λήθην λαβὼν | πρὸς ἀλλοτρίῳ τε ψυχαγωγηθεὶς πάθει, | μεθ’ ἡδονῆς ἀπῆλθε παιδευθεὶς ἅμα, fr. 6.5–7 Kock). Where most authors oppose the experiences of others that are narrated or staged to the experiences that the audience have had themselves, Gorgias claims that the experiences of others induce an ‘affection of [their] own’ in the souls of the audience. In shifting the focus to the audience’s experience of the representation, which he credits with the same force as actual experiences, Gorgias dissolves the opposition between the audience’s ἴδιον and the ἀλλότρια of the characters and instead envisages the former as an effect of the representation of the latter.23 While Gorgias goes out of his way to emphasize the intensity of the response to a mere representation, it is difficult to say whether the πάθημα of the spectator is the same as the πάθημα of the characters.24 Tragedy is populated with characters and choruses who shudder out of fear, shed tears in pity for others and fully devote themselves to grieving. The response described by Gorgias could thus be a re-­enactment of what the characters on stage and the chorus in the orchestra do and feel. At the same time, pity may also describe an emotion that does not echo a feeling exhibited by characters or the chorus but reacts immediately to the suffering of characters. At least for Aristotle and later theoreticians, pity presupposes some distance: if the suffering party is too close, then the audience experiences only fear, unmixed with pity.25 But even in this case, the audience response, while not identical with the experiences of the characters, would map closely onto them. Pity requires some kind of involvement, and the fear mentioned by Gorgias indicates that the audience identifies with the characters to a certain extent. The assertion about the spell of poetry in Helen ignores the reflexive pole of aesthetic illusion, to which fr. B 23 DK at least nods, and is easy to dismiss as hyperbolic. At the same time, it resonates with a set of much-­discussed findings of cognitive research. In the early 1990s, neuroscientists identified in the brains of monkeys neurons that fire not only when an action is executed but also when the action is observed in another. The discovery of these so-­called mirror neurons gave rise to a whole programme of further research and also, it has to be said, to 23  Gorgias’ emphasis on the strong effects of representations is later adopted by Plato (Resp. 606b). For Plato, however, the effects that Gorgias adduces as evidence for the power of λόγος threaten to corrupt the spectator’s soul as they will affect his ability to control his emotions in real-­life inter­ actions. See below, p. 171. 24  I am grateful to one of the anonymous readers for alerting me to this question. S/he suggests that the juxtaposition of ἴδιον and ἀλλοτρίων indicates that the response of the audience is distinct from the emotion of the characters. However, both the thrust of Gorgias’ argument and the tradition of contrasting a person’s own experience with an experience represented by a text or performance lead me to understand ἴδιον as qualifying the audience’s response as a full experience, whether or not identical with the represented experience. See now also Schollmeyer 2021, 241–3. On the relation of the πάθημα to the emotions mentioned, see Munteanu 2012, chapter 2.2 with further literature. 25  E.g. Arist. Rh. 1386a17–24. Cf. Konstan 2001, 12–13, Halliwell 2002, 216.

Gorgias ’ ἀπάτη, Sophocles ’ Electra, and Cognitive Criticism  159 wild speculation.26 The interpretation of the experimental data continues to be controversial, and it has proved difficult to establish the salience of mirror neuro­ns for feelings and more complex processes in the human brain. That said, there are strong indications that observing an action engages our brain in similar ways to performing the same action. fMRI scans show that some of the same brain areas are involved in executing and perceiving certain activities.27 Experiments reveal that these brain areas are also put to work when we imagine these activities. While not providing any direct evidence, these empirical investigations of the human imagination are nevertheless suggestive for our understanding of responses to literature. Not only when we watch an actor carry out an action, but also when we hear him speak about an action or when we read about an action, brain areas seem to become active that are also involved when we execute this action. While not attempting to erase the differences between doing and imagining, cognitive studies emphasize the experiential character of the imagination. Needless to say, when Gorgias speaks of ‘an affection of its own on the basis of the fortunes and misfortunes of other people’s affairs and bodies’ (ἐπ’ ἀλλοτρίων τε πραγμάτων καὶ σωμάτων εὐτυχίαις καὶ δυσπραγίαις ἴδιόν τι πάθημα), he is operating in a very different epistemic field, and yet, as outlandish as his claim may sound to ears trained in traditional philosophy of the subject, cognitive research supports the idea that representations can trigger experiences that in some regards are similar to those of engaging in the represented activity itself. Just as Gorgias’ focus on the emotions in the audience response chimes with the emphasis on emotional effects of literature in cognitive narratology,28 his reference to its somatic dimension corresponds to insights into the embodied nature of reader response. Gorgias, just like Aristotle, seems to be indebted to a specific physiological model, the humour theory, which envisages the body as balancing such basic qualities as warm versus cold and dry versus humid.29 While shuddering seems to be a reaction to an increase of the cold element, tears result from too much humidity. This physiological model is obviously remote from the premises of today’s medical understanding, but this difference should not detract from the fact that Gorgias concurs with cognitive theories in stressing the emotional appeal of literature and the physical nature of our responses to it.

26  The classic statement of the thesis about mirror neurons can be found in Gallese et al. 1996. For a critique, see e.g. Hickok 2009. For a more recent survey, see Ferrari and Rizzolatti 2014. See also Budelmann in this volume, p. 8. 27  E.g. Iacoboni et al. 1999, Gazzola and Keysers 2009. 28  E.g. Hogan 2011 and 2018. 29  Cf. Flashar 1956, who argues that katharsis in Aristotle’s Poetics is not only a metaphor but also a medical term. See also Segal 1962, 104–6 on the comparison between soul and body in Helen 14 and the ‘almost physical impact’ of words on the soul, and, more recently, Holmes 2010, 212–15. The somatic dimension of ἔλεος and φόβος was also stressed by Schadewaldt 1955, who called them ‘seelisch-­leibliche Elementaraffekte’ (on Gorgias, see 144–5).

160  Jonas Grethlein A further point that is surprising from the perspective of established critical accounts of reader response, but which chimes well with cognitive approaches, is the ‘affairs and bodies’ (πράγματ[α] καὶ σώματ[α]) which Gorgias singles out as responsible for inducing in the spectator ‘an affection of its own’ (ἴδιόν τι πάθημα). Wolfgang Iser and other reader-­response theorists view the reception of literature as an intellectual engagement in which the reader fills in blanks left in the narrative and makes conjectures about the further development of the plot.30 This, however, is only part of the story. Cognitive accounts of the enactive nature of reading suggest that the representation of physical action is a similarly important means of enticing readers.31 Accounts of motion echo in the reader’s own sensorimotor system; both descriptions and the execution of simple bodily movements resonate with particular strength and therefore have a strong immersive cap­acity.32 Gorgias’ reference to ‘bodies’ besides ‘affairs’ is not as archaic as it may seem when compared to the highly sophisticated models of reader-­response theory. Not only Gorgias’ emphasis on the strong and embodied echo of dramatic performances but also the specific elements he identifies as the trigger of this response are in line with the results of cognitive research. It has proven fruitful, in this section, to read Gorgias’ reflection on tragedy and its effects on audiences in the light of cognitive theory. Cognitive models of aesthetic experience reveal that Gorgias’ claims are more than glib quips and archaic exaggerations. The reference to the recipient and his contribution to the workings of ἀπάτη in fr. B 23 DK gestures to the insight that immersion is constantly poised against a residual awareness of the representation. His elaboration on the immersive effects of poetry in Helen shows that, despite the obvious differences, Gorgias’ argument for the strong resonances of actions imagined as well as perceived, particularly bodily movements, anticipates embodied and enactive models of responses to literature. With this, we are now well-­equipped to reconsider the false messenger speech in Electra.

9.3  An enactive reading of the false messenger speech in Electra It is my thesis that the report of Orestes’ old teacher in Electra engages with the very ambiguity of deception that is showcased in Gorgias fr. B 23 DK.33 As we 30  E.g. Iser 1972 and 1976. 31 E.g. Troscianko 2014, Caracciolo 2014, Kukkonen and Caracciolo 2014b. For a thought-­ provoking comparison of reader-­ response theory with an embodied model of response, see Kukkonen 2014a. 32  For the kinesis or motor-­resonance approach, see especially Bolens 2012 and, very instructively, Cave 2016, 32–45. On the strong effect of descriptions of simple bodily movement, see Grünbaum 2007, 309–10, Kuzmičová 2012, 29, 2013, 114–16. 33 While δόλος and its cognates occur more frequently (37, 124, 197, 279, 649), ἀπάτη and its cognates are also twice used in Electra, once in referring to Clytaemnestra’s murder of Agamemnon (125), and once

Gorgias ’ ἀπάτη, Sophocles ’ Electra, and Cognitive Criticism  161 have seen, Gorgias’ comment on tragedy derives part of its paradoxical appeal from the oscillation of ἀπάτη between ethical deception and aesthetic illusion: the characterization of the poet as just clashes with the ethical meaning of ἀπάτη, but is justified by its aesthetic meaning. I now wish to show that the messenger speech is designed to create an aesthetic illusion that is tied to deception. However, this association of aesthetic illusion with deception comes with a special twist. The theatre audience is not deceived—­they are fully aware of the Paedagogus’ lie. However, the features that help to transport them also, at the level of the action, trick Clytaemnestra into feeling safe and precipitate Electra into despair. The means of transporting the theatre audience is simultaneously a means of deception within the play. The aesthetic illusion of tragedy is thus entangled with deception across the boundary between the internal and external communication systems of drama. Before discussing this entanglement of ethics and aesthetics, we will have to explore the vividness of the messenger speech, which has been duly noticed, but has neither been fully analysed nor interpretatively exploited by scholars. I will therefore first dissect the immersive appeal of the account of Orestes’ death with the help of cognitive approaches, notably an enactive model of narrative. Some scholars claim that the spectators were or could be enticed by the false messenger speech to such a degree that, at least for the moment, they were made to believe in its truth.34 This thesis, however, is flawed and confounds the distinct issues of aesthetic illusion and fictionality.35 The former describes a specific mode of attention paid to the represented scene, while the latter is the assessment of its truth status. The experiential quality of narratives is, by and large, independent of whether they are fictional or factual. We can be immersed in story worlds no matter whether or not their claim to referentiality is suspended. While immersion may deepen the appeal of an account that we consider factual, it does not make us believe the truth of an account that we know to be false. The pairing of immersion with reflexive distance discussed above reveals why spectators are in fact very unlikely to become oblivious to the deception of the Paedagogus. We follow a play and may be drawn into its action, but nonetheless remain aware of attending to a play. The case of messenger speeches is

to describe Orestes’ unfulfilled announcements of his imminent return (170, ἀπατώμενον), and thus pertinent to the central deceit of the plot. Sophocles’ oeuvre features further deception speeches, not­ ably in the false-­merchant scene of Philoctetes (542–627) and also Lichas’ lie about the true identity of Iole in Trach. 248–90, that could also be interpreted along similar lines. This chapter, however, concentrates on the false messenger speech in Electra, which is arguably the most vivid specimen of a false speech in Sophocles and best illustrates the entwinement of deceit with aesthetic illusion. For a comprehensive account of Sophoclean ‘Täuschungsszenen’, see Parlavantza-­Friedrich 1969. 34  See Kaibel 1896, ad 660, Wilamowitz-­Moellendorff 1917, 188–93, Whitman 1951, 158, Ringer 1998, 171n59 and, more cautiously, Winnington-­Ingram 1980, 236n66, Seale 1982, 65, Lloyd 2005, 69. Against the idea of a complete absorption, see Reinhardt 1947, 162n1, Parlavantza-­Friedrich 1969, 36–7, Waldock 1951, 183–4, Finglass 2007, 304n43. 35  Wolf 2008 elaborates on the differences.

162  Jonas Grethlein particularly complex but no exception, as Budelmann and Van Emde Boas compellingly argue.36 While in dramatic dialogue the attention given to the conversation of the characters is poised against the attention given to the theatrical performance, there are three different levels in the reception of a messenger speech: the spectator can concentrate on the actor and his performance, on the character and his speech act, or on the events narrated in this speech. Even if the messenger’s narrative is highly immersive, the sense of witnessing the narrated events will be modulated by a concomitant awareness of the fact that, first, this narrative is mediated by a character and, second, that behind the character there is an actor. The audience’s residual awareness of the narration of the Paedagogus, who is pretending to be a messenger from Phocis, makes it hard for them to become oblivious to the truth status of his narrative. The claim that the Paedagogus’ account induces in the audience a full illusion is thus untenable, but it nevertheless helpfully puts the spotlight on its experiential quality. Scholars have often stressed the detail-­rich character of the speech to explain its powerful effect,37 but this is insufficient, if not outright false. As cognitive studies show, it is not the amount of detail in itself that renders narrative immersive.38 On the contrary, stories highly invested in descriptive details are not, in Emily Troscianko’s terms, cognitively realistic, for they do not conform to how we seem to perceive the world around us. The mental processes involved in perception continue to be a hotly debated topic, but there is evidence that they do not involve photographic images of our environment. Enactive and embodied approaches suggest that perception is highly selective and concentrates on aspects of things that are relevant to how we actually or possibly interact with them. If we assume that imagination parallels perception, then narratives become cognitively realistic and feel experiential to the extent that they do not try to deliver picture-­ like descriptions and instead focus on things in terms of actual and possible interaction. The exploration of the style that renders narrative enactive is still in its infancy, but some important features have already been identified:39 1. A prerequisite for the recipient’s immersion in the narrated world is transparency. An overtly present narrator and strong self-­referential elements highlight the act of mediation. Only an unobtrusive narrator and the absence of self-­reference make it possible for the recipient to focus on and, potentially, to immerse herself in the represented scene.

36  Budelmann and Van Emde Boas 2020, 75–6. 37  E.g. Kells 1973, ad 680–763: ‘detailed persuasive account’; Seale 1982, 65: ‘implicit and highly seductive detail’; Ringer 1998, 168: ‘wealth of specific detail’; March 2001, ad 680–763: ‘abounding in realistic detail’. 38  See Grethlein and Huitink 2017 for a fuller explanation with references to scholarship. 39  The list given above develops further the list presented in Grethlein and Huitink 2017.

Gorgias ’ ἀπάτη, Sophocles ’ Electra, and Cognitive Criticism  163 2. Since our perception is closely tied to action, the account of action plays an important role in immersing recipients. The narration of simple, intentional bodily movements in particular seems to have a strong echo in our sensorimotor system. The narration of actions should be dynamically veracious. This means that the time that a passage takes to read should be com­men­ sur­ate with the duration of the described action in the real world. While it is hard to pinpoint when exactly narrative time and narrated time fully coincide—­except in direct speech—­the notions of pause and summary illustrate ex negativo the salience of dynamic veracity. 3. Descriptions ought to be ‘just in time’, focus on affordances, and appeal to the senses. Just as our perception attends to things that are relevant to possible interactions, a narrative should introduce objects as and when they become relevant and focus on affordances, that is, on those aspects of an object that pertain to potential and actual ways of interacting with them. Likewise, references to features that appeal to our senses make the description resonate with our sensorimotor system. 4. A shift of the deictic centre to the narrated scene reinforces the recipient’s immersion. Supported by adverbs, e.g. ‘now’ and ‘here’, tense and prefixes and prepositions create temporally and spatially an internal viewpoint that helps to transport the reader or listener. In focalization, this viewpoint is even tied to a specific character. This enactive approach to narrative can help us analyse the spell that the messenger speech in Electra casts on the theatre audience as well as the internal listeners. When we consider the Paedagogus’ report in this light, it is immediately striking that he refrains from providing detailed descriptions for their own sake. There are no descriptions of the setting, of Orestes or the other protagonists, or of objects. We learn about these things only insofar as they matter to the action. The old teacher mentions the turning post when Orestes passes it; he does not describe where it stands or what it looks like, as this plays no role in the action, but only that Orestes first grazes it with the nave of the chariot wheel and then crashes into it (720–1; 744–5). There is no depiction of Orestes’ chariot, but its rail is mentioned when Orestes, after crashing into the turning post, veers over it (746–7).40 Here as elsewhere, Sophocles has the Paedagogus confine himself to things as and when they pertain to the action in a cognitively realistic manner. Vital to the experiential quality of the messenger speech in Electra is its pacing. As cognitive scholars note, ‘for immersion to retain its intensity, it needs a contrast of narrative modes, a constantly renegotiated distance from the narrative

40  See Finglass 2007, ad 746 for the specific meaning of ἄντυξ in this sentence.

164  Jonas Grethlein scene, a profile made of peaks and valleys’.41 The Paedagogus carefully modulates his narrative between summaries, iterative passages, and highly enactive accounts of individual activities. The change of register is reflected in the choice of tense. After mentioning Orestes’ arrival (681–4) and his victory in the foot race (685–7), and after summarizing his successes in other competitions (688–95), the Paedagogus zooms in on the chariot race. First comes the catalogue of the competitors (701–8). The start of the race is narrated in the aorist (709–11), a summarizing description of the racing is cast in the imperfect (714–22). In presenting an action as unfinished and ongoing, the imperfect can be an important means of immersing the reader in the action,42 but here it rather paints the background against which the experiential account of the collision of the Aenian with the Barcaean charioteer stands out. This accident is narrated in the historical present, which, in this messenger speech as elsewhere, is used particularly for turning points in the action (723–7).43 While the Paedagogus conveys the crashing of the other chariots iteratively in the imperfect, he returns mostly to the historical present when he singles out the reactions of the Athenian and Orestes to the mass accident (731–8).44 The imperfect is used for the summarizing representation of the ensuing duel between Orestes and the Athenian (739–42). The lethal crash of Orestes is highlighted through the present tense again (743–6), which then gives way to the aorist that is used to report the dispersal of the horses and the reaction of the spectators (747–56). As this brief run-­through of the speech shows, Sophocles has the old teacher shift between a background of summarized action and foregrounded events that are presented with particular vividness. The most experiential part is Orestes’ crash, which echoes the collision of the Aenian and Barcaean chariots (745 ἔθραυσε ~ 729 ἔθραυε; 745 παίσας ~ 727 συμπαίουσι), but is narrated at greater length and split up into individual movements: while the horse was turning, Orestes ‘slackened his left rein’ (λύων ἡνίαν ἀριστεράν, 743) and ‘accidentally struck the edge of the pillar’ (λανθάνει στήλην ἄκραν | παίσας, 744–5). Then, ‘he broke his axle-­box in the centre and was thrown from the rim of his chariot’ (ἔθραυσε δ᾽ ἄξονος μέσας χνόας | κἀξ ἀντύγων ὤλισθεν, 745–6). The fall is further dissected: ‘he gets entangled in the sharp-­cut reins, and as he fell to the ground his horses bolted wildly into the middle of the course’ (ἐν δ᾽ ἑλίσσεται | τμητοῖς ἱμᾶσι· τοῦ δὲ πίπτοντος πέδῳ | πῶλοι διεσπάρησαν ἐς μέσον δρόμον, 747–8). The narrative elaborates on bodily movements which are spatially situated and therefore likely to trigger the reader’s imagination through the resonances in her sensorimotor system, an insight of cognitive studies anticipated in Gorgias’ claim 41  Ryan 2001, 137. See also Kuzmičová 2012, 43: ‘Presence cues become effective only if moderately dosed. Not only should they appear periodically, once in a while, for a continuous sense of presence to arise. They should also appear just once in a while, if presence is to be instantaneously elicited at all.’ 42  Cf. Bakker 1997, illustrating this thesis through Thucydides. 43  Cf. Allan 2011. 44  On the Athenian as a means of engaging the Athenian audience, see Ringer 1998, 169.

Gorgias ’ ἀπάτη, Sophocles ’ Electra, and Cognitive Criticism  165 that mere words can instil in the recipient’s soul ‘an affection of its own’ through the account of ‘affairs and bodies’. The narration of Orestes’ crash also coincides with a noteworthy shift of pace. The Paedagogus limits himself to six lines for the race between Orestes and the Athenian that extends over arguably six laps; he then slows down for the crash and describes it in the distinct motions that I have just listed. It is notoriously difficult to measure narrative time and narrated time,45 but the comparison with the preceding summary suggests that the description of the fall effects dynamic veracity. The assimilation of narrative time to narrated time helps to transport the reader to the plain of the race. The immersive appeal of the account of Orestes’ crash is further heightened by the presence of an internal audience through whose lens the narrative continues: But when the crowd sees him fallen from his chariot, they bewailed loudly the young man, who had achieved so much and then met with such disaster, at one moment dashed to the ground, at another with his legs tossed to the sky, until the charioteers with difficulty brought his galloping horses under control and set him free, so covered with blood that not one of his friends who saw him could have recognized his poor body. στρατὸς δ᾽ ὅπως ὁρᾷ νιν ἐκπεπτωκότα δίφρων, ἀνωλόλυξε τὸν νεανίαν, οἷ᾽ ἔργα δράσας οἷα λαγχάνει κακά, φορούμενος πρὸς οὖδας, ἄλλοτ᾽ οὐρανῷ σκέλη προφαίνων, ἔς τέ νιν διφρηλάται, μόλις κατασχεθόντες ἱππικὸν δρόμον, ἔλυσαν αἱματηρόν, ὥστε μηδένα γνῶναι φίλων ἰδόντ᾽ ἂν ἄθλιον δέμας. (749–56)

The internal spectators, a case of ‘fictional minds’,46 give the reader an anthropocentric point of reference vis-­à-­vis the objects located in space and thereby help to draw her into the narrated action, so her soul, in the words of Gorgias, can feel ‘an affection of its own’. Moreover, the response of the spectators, who contrast Orestes’ gruesome death with his extraordinary successes, raises the pathos of the scene. So far, we have seen that the immersive appeal of the messenger speech in Electra is generated through a carefully paced enactive narration that conforms with Gorgias’ emphasis on ‘affairs and bodies’ and artfully deploys fictional minds to generate in the audience ‘an affection of its own’. There is, however, a further 45  Cf. Grethlein and Huitink 2017, 12 for the suggestion that dynamic veracity is a natural ex­pect­ ation which lasts until it is broken by summary or freeze. 46  On fictional minds, see esp. Palmer 2004.

166  Jonas Grethlein feature of the Paedagogus’ report that also has a considerable impact on the response of the audience and later readers. The report is couched in language that evokes epinician and epic in general, as well as at least one particular intertext from the Iliad.47 As the commentators document in detail, Sophocles takes up and reworks elements of the chariot race in the Iliad’s funeral games. How do the epic and epinician flavour and the pointed references to the Iliad affect the response to the old teacher’s narration in Electra? The evocation of two prominent generic frameworks certainly endows the narration with plausibility. The death in a chariot race not only makes for a compelling story because it harks back to the chariot race between Pelops and Oenomaus, which, as the Chorus state, formed the beginning of the family’s suffering,48 but the young successful athlete, the idea of a chariot race, and the danger of the turning post, as well as the threat of a sudden downfall, are schemata with which most audience members will have been familiar.49 Their deployment makes the report of the Paedagogus easy to process and gives it an internal plausibility—­the narrative follows firmly established patterns and fulfils the expectations of the recipients. The foil of epic and epinician is also vital to the grandeur of the report; it generates an atmosphere that is elevated and heroic.50 The august tone inspires awe and may thereby help to captivate the audience. At the same time, the epic and epinician elements weigh against the immersive pull and strengthen the reflexive pole of the reception. While enactive features invite the audience and readers to delve into the narrated action, intertextuality alerts them to the me­di­ ation. In drawing the attention to other texts and revealing the textuality of the narration, allusions and generic colouring go against the immersive pull that experiential narratives exert on recipients. There are further features that, while contributing to the overall effect of the report, undercut its immersive appeal. The messenger ignites a rhetorical firework; some of his rhetorical devices are so arresting as to threaten the transparency at which experiential narratives aim.51 Consider for instance this sentence: ‘All in a mass together, they did not spare their goads, so that each would overtake the wheels of the others and the snorting of the horses.’ (ὁμοῦ δὲ πάντες ἀναμεμιγμένοι | φείδοντο κέντρων οὐδέν, ὡς ὑπερβάλοι | χνόας τις αὐτῶν καὶ φρυάγμαθ᾽

47  Besides the commentaries, see Davidson 1988 and Barrett 2002, 137–67 on the Homeric background of Electra. Barrett 2002 demonstrates the general affinity of messenger speeches to epic poetry. On epinician elements in the Electra’s messenger speech, see, for example, Finglass 2007, ad 686; 687 and 689. 48  Cf. 504–15. On the pointed similarities between the races, see e.g. Segal 1981, 267–8, Schein 1982, 76. 49  Scodel 1984, 82 notes that the messenger speech is ‘a fable of common Greek morality’. 50  Segal 1981, 282 argues that ‘the high registers of epic and epinician poetry play against the low registers of lie and deception’. 51  On the importance of transparency for experiential narrative, see Allan 2020, 19 with further references.

Gorgias ’ ἀπάτη, Sophocles ’ Electra, and Cognitive Criticism  167 ἱππικά, 715–17). The phrase ‘did not spare their goads’ is elegant but less enactive than a direct and unnegated expression, for instance ‘goaded the horses’, would be. The metonymy of wheels for chariots may not be a significant impediment to the imagination, as the wheels would in fact be overtaken, but the periphrasis ‘snorting of the horses’, literally ‘horsey snorting’, for ‘snorting horses’ is so startling that it commands the reader’s attention in itself, instead of straightforwardly directing it to the action.52 Despite these elements that establish a reflexive distance from the narrated action, the experiential appeal of the messenger’s report is nevertheless striking. It is in fact a speech geared towards inducing in its recipients ‘a terribly fearful shuddering, a much-­weeping pity, and a yearning that desires mourning’. As pointed out above, the theatre audience, while being invited to immerse themselves in the race, are very unlikely to succumb to the Paedagogus’ lie. That said, the very elements that serve to transport the theatre audience are crucial to the Paedagogus’ attempt to deceive the internal audience. The aesthetic illusion of drama is entangled with the deception at the level of the action. Within the play, aesthetic illusion fully coincides with deception, as the internal recipients succumb to the deception when they immerse themselves in the narrated action. At first sight, this final point may seem to be at loggerheads with the argument put forward above (p. 161) that the issue of referentiality ought to be distinguished from the issue of attention. And indeed, as I hope has become clear, the audience’s or reader’s immersion will not make them oblivious to the deceptive character of the speech. For the audience on stage, however, the Paedagogus’ report has a different status. While we know that it is a lie, the account is presented as truthful to Clytaemnestra and Electra. In this referential framework, the immersive quality of the narration feeds into its credibility. An experiential presentation reinforces the plausibility of a report that brings with it a claim about referentiality. If the narrator manages to put his audience into the shoes of the characters, they are less likely to question the truth of his account. The effect of the messenger speech is impressively attested by the response it elicits on the stage. Electra is unable to utter a word for some twenty verses and then starts to mourn, alone and together with the Chorus, with an intensity that evokes Gorgias’ ‘yearning that desires mourning’. Clytaemnestra, who had just prayed for the death of Orestes, is so affected by the messenger’s narrative that, other than relief, she initially also feels horror and sadness.53 The vividness of the narration seamlessly translates into credibility. Commentators have noticed that 52  The effect of τροχῶν βάσεις in 718 is similar. 53  Note, however, Electra’s assertion that Clytaemnestra ‘went away gloating’ (807) with Goldhill’s discussion of the various audiences on stage and their interaction with each other (2002, 49). See also Kells 1973, 7, who gives much weight to the maternal feelings that Clytaemnestra seems to have here as well as the objections to this interpretation in Finglass 2007, ad 766–8. For further literature, see MacLeod 2001, 119n33, Chaston 2010, 156n108.

168  Jonas Grethlein the Paedagogus never swears the oath that Orestes requested as a confirmation of the truth of his story (47). This is as revealing as it is striking: an oath is not necessary; the experiential feel of the narrative leaves no space for doubts about its veracity.54 The spell of the messenger speech is so powerful that it trumps the evidence for Orestes’ return that Chrysothemis had encountered.55 She enters in the subsequent episode and reveals to Electra what she deems to be ‘clear signs’ (σαφῆ σημεῖα, 885–6) that their brother has come back: at the grave of Agamemnon, she has found not only traces of an offering, namely milk and flowers, but also a freshly cut lock, which, she concludes, can only be Orestes’. Chrysothemis’ rhesis is lucid and her interpretation of the evidence plausible:56 she first recounts her visit to the grave and then considers who may have left the offerings and the lock. Carefully eliminating all other possibilities, Chrysothemis infers that it must have been Orestes. As cogent as this conclusion is, it is rejected by Electra, who even manages to convince her sister of the death of Orestes. The fact that the messenger’s report, mere words, overrides the evidence of material signs underscores the force of experiential narrative.57 When Chrysothemis tells Electra that Orestes has returned, she stresses that her belief is rooted in what she has seen.58 There are ten occurrences of words signifying vision and the eyes in the first fifteen verses of her rhêsis (892–906). The pervasiveness of sight is tangible when she recounts her discovery of the lock: The moment, poor me, I saw it, a familiar image bursts in upon my soul, telling me that I am seeing a sure sign of Orestes, dearest of all men. Taking it in my hands, I do not speak an ill-­omened word, but at once my eye fills with tears of joy. κεὐθὺς τάλαιν᾽ ὡς εἶδον, ἐμπαίει τί μοι ψυχῇ σύνηθες ὄμμα, φιλτάτου βροτῶν πάντων Ὀρέστου τοῦθ᾽ ὁρᾶν τεκμήριον καὶ χερσὶ βαστάσασα δυσφημῶ μὲν οὔ, χαρᾷ δὲ πίμπλημ᾽ εὐθὺς ὄμμα δακρύων. (902–6)59

54  Cf. March 2001, ad 47. 55  On the juxtaposition of the two scenes, see e.g. Schadewaldt 1926, 59n1, Reinhardt 1947, 164–5. There are some verbal echoes that may be more than incidental and reinforce the juxtaposition: 720 ἐσχάτην στήλην ~ 900–1 ἐσχάτης . . . | πυρᾶς; 721 ἔχριμπτ’ ~ 898 ἐγχρίμπτει; 745 παίσας ~ 902 ἐμπαίει. 56  Cf. Finglass 2007, ad 892–919. 57  For a different interpretation, see Chaston 2010, 159: ‘Elektra trusts the words of her mother’s friend over what her sister has seen. Her despair deludes her into seeing what is not there and into not seeing what is there.’ 58  Cf. Seale 1982, 67–8, Chaston 2010, 159. On vision and its ambiguity in Electra, see besides Seale 1982, 56–83 also Wilson 2012, 559–62. 59  I follow Kamerbeek 1974, ad 902–3 in assuming that ψυχῇ depends on ἐμπαίει. Finglass 2007, ad 903, takes it with σύνηθες.

Gorgias ’ ἀπάτη, Sophocles ’ Electra, and Cognitive Criticism  169 The sight of the lock conjures up an image of Orestes in Chrysothemis’ im­agin­ation. This effect of a visual object is expressed in tactile terms (ἐμπαίει) that conform to later theories of vision as a process of being touched by particles emitted from the object.60 When Chrysothemis actually touches the lock, tears well up in her eyes. The direct translation of the sight of the lock into an image of Orestes, a perceptual metonymy of sorts, connoting the literal as well as metaphorical entanglement of seeing with touching, as well as the marked repetition of ὄμμα, first for the imagined object, then for the real eye,61 all combine to drive home the force that sight has in this scene. The salience of vision in Chrysothemis’ rhesis makes it particularly striking that the Paedagogus goes out of his way to emphasize his status as an eyewitness at the end of his report: ‘Such is my story for you, painful indeed when told, but for those who saw it, the greatest sorrow that I have ever seen.’ (τοιαῦτά σοι ταῦτ᾽ ἐστίν, ὡς μὲν ἐν λόγῳ | ἀλγεινά, τοῖς δ᾽ ἰδοῦσιν, οἵπερ εἴδομεν, | μέγιστα πάντων ὧν ὄπωπ᾽ ἐγὼ κακῶν, 761–3). This is of course a lie. The old teacher’s invocation of his autopsy also sits uneasily with the omniscient style of his narration.62 It has further been claimed that the vehement assertion of his autopsy highlights the deceptive character of the speech.63 At the level of the action onstage, however, the effect is the opposite. The vision to which the Paedagogus lays claim seems somehow to be transferred to his listeners. Clytaemnestra speaks of the πίστα τεκμήρια that the messenger has provided (774).64 Τεκμήρια is the term that is used later in the play for the material, if equally deceptive, evidence of the urn (1109).65 For the audience on stage, it seems, the narration of the Paedagogus is as visual as the material signs and indeed outshines them in its credibility. Under the influence of the messenger speech, Electra considers Orestes as ‘clearly gone’ (τῶν φανερῶς οἰχομένων, 831). Φανερός and its cognates not only denote visibility in general but are used in Electra specifically for the longed-­for appearance of Orestes at Mycenae (171–2; 1261; 1274; 1285). The death of Orestes, as relayed by the Paedagogus, appears to be as visible as the real appearance of Orestes. The old teacher makes his listeners somehow see what he narrates. The quasi-­visible nature of the narrated events prefigures a later critical and rhetorical concept that has received much attention lately. Ἐνάργεια is often defined as the capacity of a speaker or narrator to make the audience or reader (nearly) visualize words.66 It is an intriguing case of ἐνάργεια avant la lettre that 60  For earlier literary texts that anticipate these theories of vision, see Cairns 2011b. 61  Cf. Easterling 1973, 27, Seale 1982, 68. 62  Cf. Barrett 2002, 162–5. On the omniscient style, see also Marshall 2006, 210. 63  Seale 1982, 65. For the tenor of this interpretation, see Reinhardt 1947, 162. In contrast, Wilson 2012, 561 argues that ‘the listeners, both onstage and offstage, feel as if they, too, have seen these events, not merely heard about them “in words” ’. 64  Cf. de Jong 1994. 65  On the urn and its deceptive evidence, see now esp. Billings 2018. 66  Among the vast literature, see Manieri 1998, Otto 2009, Webb 2009. Huitink 2019 demonstrates that the ancient understanding of ἐνάργεια is not limited to sight but also encompasses other senses.

170  Jonas Grethlein the Paedagogus’ (false) claim to autopsy seems to translate into the visibility of his account for his audience and prevails over what Chrysothemis has actually seen. Particularly noteworthy is the occurrence of a cognate of ἐνάργεια itself, the adverb ἐναργῶς, in Chrysothemis’ account of what she has seen at Agamemnon’s grave: ‘Orestes is here with us, know this as you hear it from me, manifestly, just as you see me’ (πάρεστ᾽ Ὀρέστης ἡμίν, ἴσθι τοῦτ᾽ ἐμοῦ | κλύουσ᾽, ἐναργῶς, ὥσπερ εἰσορᾷς ἐμέ, 877–8). Signifying Orestes’ real, bodily presence, ἐναργῶς serves to stress the evidence Chrysothemis is offering. Its association with vision is palp­ able in the explanation of ‘just as you see me’. While qualifying πάρεστ’,67 ἐναργῶς is also linked to an act of speaking: ‘know this as you hear it from me’. The link is stressed by the position of ἐναργῶς far away from the verb it qualifies and right after the reference to hearing. Chrysothemis does not assert that she speaks ἐναργῶς, but the later understanding of ἐνάργεια as the capacity to make things visible through words lies dormant in the conjunction of a speech act and the visibility of its object.68

9.4  Aesthetic illusion and deceit, an uncanny entanglement Cognitive models of reader response have helped to analyse the experiential quality of the messenger speech in Electra. The report of the Paedagogus is artfully engineered to immerse the external as well as internal listeners. The old teacher expounds ‘the fortunes and misfortunes of other people’s affairs and bodies’ (ἐπ’ ἀλλοτρίων τε πραγμάτων καὶ σωμάτων εὐτυχίαις καὶ δυσπραγίαις) in order to instil in his audience ‘an affection of its own, by the medium of word’ (ἴδιόν τι πάθημα διὰ τῶν λόγων). Since the speech is a lie, it brings together aesthetic illusion with deception. The Electra thus stages the entwinement of aesthetic illusion and deceit with which Gorgias fr. B 23 DK plays.69 Whereas Gorgias evokes the semantics of deception to render a comment on aesthetic illusion maximally poignant, Sophocles entwines aesthetic illusion with ethical deception dra­mat­ic­ al­ly. However, the theatre audience is not deceived while they immerse themselves in the action. Their aesthetic illusion is tied up with a deception at the level of the action. The elements that transport Sophocles’ audience are simultaneously a means of deceit, but in the world of the play. A full merging of aesthetic illusion and deception takes place only on stage, where the messenger’s experiential account helps to dupe Clytaemnestra. 67  Cf. Groeneboom 1935, Kells 1973, and Finglass 2007: ad loc. 68 The Electra thus belongs to a literary prehistory of the critical concept of ἐνάργεια, which is trad­ ition­al­ly seen as borrowed from Hellenistic philosophy: Zanker 1981, 308–10, Manieri 1998, 113–22, Otto 2009, 37–62, Bussels 2012, 61–71. 69  For a very different interpretation of the deceit, see Segal 1981, 282: ‘The lie about the Delphic festival at the exact center of the play is a paradigm for the corrupted ritual and civic order.’

Gorgias ’ ἀπάτη, Sophocles ’ Electra, and Cognitive Criticism  171 We have seen that aesthetic illusion and factuality operate at different levels. Whereas the latter concerns the referential status of a text, the former pertains to the recipient’s mode of attention and thus cuts across the line between fact and fiction. In associating the immersion of aesthetic illusion with deceit, however, Electra raises the uncomfortable question of whether this distinction can be maintained. It makes us wonder whether aesthetic illusion and deception, both signified by ἀπάτη, have a common essence or are even more or less the same. In Sophocles’ play, make-­believe is deployed to make the internal audience believe something that is false—­is imagining something that is not there inherently linked to accepting something that is not true? Does aesthetic illusion create a slope on which we inevitably slide into delusion? The hermeneutic stance of the theatre audience is different from that of the internal listeners. The external spectators are not tricked by the messenger, and yet this trickery hinges on the very spell that generates their aesthetic illusion, thus gesturing uncannily towards at least an affinity between the two. This association may baffle us, but the popularity of the thesis that the spectators were taken in so strongly by the messenger’s report as to believe its truth illustrates its appeal even to modern scholars. More importantly, the association of aesthetic illusion with delusion seems to have enjoyed some plausibility in antiquity. Plato is perhaps its most eloquent and influential advocate. In his eyes, poetry is rendered particularly pernicious by its immersive appeal; this comes to the fore when in Republic 10 Socrates notes that only very few are immune to the spell of poetry (605c6–d5).70 In an argument that seems to respond to Helen (fr. B 11.9 DK), he continues to expound on the strong responses that the fates of others elicit and the lasting impact of this on the recipient’s soul. The audience’s indulgence in tears and lamentation, but also in desire and laughter, weakens their ability to control their desires in real life (606a3–d7). This effect, which Plato has Socrates introduce as the greatest danger of poetry (605c6), is different from the peril adumbrated in Electra. Whereas in Electra an experiential narrative is a means of selling a lie, Plato is concerned with the corruption of the soul, which loses its capacity to suppress reprehensible drives. That said, Plato’s elaboration of the harm effected by aesthetic illusion is part of a polemics against poets as liars; in the very passage discussed, caution against poetry is justified by the fact that it lacks truth (608a7).71 The entanglement of aesthetics with ethics which Gorgias pinpoints and which Sophocles teases out dramatically is at the heart of Plato’s criticism of poetry. There are further ancient texts that bring aesthetic illusion into a direct encounter with deceit; but alas, this chapter does not afford space for their

70 See Halliwell 2011, 179–207 for a nuanced reading of Republic 10 that foregrounds Plato’s acknowledgement of poetry’s spell. 71  The idea of poetic lies is prominent in Republic 3. See e.g. 382d1–9.

172  Jonas Grethlein discussion.72 Instead, I would like to close with a comment on the triangulation of ancient literature and ancient criticism with cognitive theory that this chapter has advanced. Both Gorgias and Sophocles reveal the potential that cognitive criticism has for classics. Cognitive accounts of aesthetic experience allow us to see in Gorgias’ reflections on poetry more than pre-­theoretical rumination; they have helped make Gorgias fr. B 23 DK into a robust framework for a new reading of Electra. With the help of enactive models of narrative, I was able to analyse the means by which Sophocles entwines deceit with aesthetic illusion. At the same time, the readiness with which Gorgias, Sophocles, and other ancient authors blend together the immersive appeal of literature with untruthfulness is confusing for us. Why, we wonder, should aesthetic illusion be inherently linked to untruthfulness, why should an immersion in tragedy corrupt our souls? This lingering tension is a helpful reminder of the limits as well as the benefits of cognitive approaches to classics. Experiments play a key role in cognitive research; it is therefore not incidental that most cognitive studies concentrate on the present. The dialogue between classics and cognitive studies can thus be mutually beneficial: on the one hand, classicists can reconsider their material in the light of cognitive research; on the other hand, they are capable of endowing cognitive studies with a historical perspective. This dialogue, however, needs to be conducted with due caution. While the human cognitive apparatus seems to change only very slowly, cognition is not only a biological but also, and at the same time, a cultural activity. The inclination to couple aesthetic illusion with deception is a case in point, as it alerts us to an understanding of aesthetic experience that is different from ours. While the ancient awareness of embodied responses to the spell of narrative chimes with cognitive research, the alignment of aesthetic illusion with ethical delusion reveals an approach that is different from ours.

Acknowledgements I wish to thank Ineke Sluiter and Felix Budelmann as well as the anonymous ­readers for their comments and suggestions, which helped me clarify the argument. This chapter draws on parts of Grethlein 2021, ch. 3.

72  See Grethlein 2022.

10

Seeing Together Joint Attention in Attic Tragedy A. C. Duncan

10.1 Introduction To focus spectators’ attention upon a significant part of a busy stage is a core ­challenge of the theatre.1 Faced with a dense, dynamic, and multi-­sensory presentation, theatre-­goers are continually tasked with establishing provisional and fluid hierarchies of significance to make sense of a production. In today’s theatre, spotlights and other lighting effects often assist spectators visually in this endeavour, but these are relatively recent innovations. It is suggestive that soon after its technical introduction in the eighteenth century, the ‘spotlight’ became a common metaphor in English for delimited attention within and beyond the theatre.2 The widespread theatrical and figurative use of the spotlight in our modern era raises the question of how ancient dramatists focused their audience’s attention without such devices.3 In this chapter, I propose that such spotlight effects were often achieved through processes of social cognition and could be signalled by certain verbal phrases. Such ‘sight invitations’, as I will call them, lead the audience to see together, practising a focused and self-­ conscious form of joint attention at moments of special dramatic importance. In addition to reducing spectators’ ‘attentional load’ by marking certain aspects of the performance as significant, these moments of seeing together often blurred distinctions between the internal (i.e. ‘fictional’) and external (‘real-­life’) audiences. By exploiting the cognitive mechanisms of joint attention, sight invitations enriched the experience of Attic tragedy, involving theatre-­goers more deeply within the epistemic, emotional, and thematic issues raised in the plays.

1  The task of directing audience attention falls, unevenly and in different ways, upon playwrights, directors, and actors alike. On attention in theatre generally, see Hamilton 2018; on actors and attention in particular, see Blair 2008, 61–2. 2  Oxford English Dictionary s.v. ‘spotlight, n.’ 1.b; cf. the earlier ‘limelight, n.’ 3 For a complementary but rather different approach to this question, see the discussions of Meineck 2011 and 2018b, especially 96–8. A. C. Duncan, Seeing Together: Joint Attention in Attic Tragedy In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0010

174  A. C. Duncan The ancient Greek theatre had a busy stage. Even in the most austere conception of Attic drama, the totality of performers, costumes, properties, movements, and sounds inevitably overwhelms theatre-­goers’ cognitive capacity for detail.4 Unable to process the whole in full, ancient spectators, like their modern counterparts, were compelled to become selective in their attention. This focus must be intelligently managed by playmakers, especially those working within a genre such as tragedy that demands more than desultory engagement from its audience for its fullest effect. To follow the narrative and thematic essentials of a developing play, theatre-­goers must be guided through a series of crucial junctures and events. Readers must depend upon words alone, but those in the theatre follow not only the performance onstage but also the collective gasps, head-­turns, and other social behaviours that cascade across the audience. Group psychology, that is to say, shapes theatrical attention in ways that literary theorists, both ancient and modern, have tended to overlook. Recent scholarship on theatrical cognition has explored methods by which both ancient and modern theatre-­makers help focus individual spectators’ attention, but the significance of collective—­and, more precisely, joint—­ attention in the theatre has, until very recently, eluded study.5 Joint attention is a basic, robustly studied feature of human social behaviour that may be distinguished from the somewhat wider concept of ‘shared’ attention.6 To attend jointly to something, it is not sufficient for multiple subjects to attend to the same object at the same time, as individuals across a hemisphere might regard the same moon. Rather, for conditions of joint attention to obtain, two or more individuals must attend to an object in a way that is ‘mutually manifest’ to each participant, as when those gathered around a fire gaze at its flame.7 In practising joint attention, subjects not only are aware of others attending to the same object but also believe those others to be reciprocally aware of the subjects’ own attention. Such an ‘intersubjective’ state both entails and facilitates some level of coordination, however slight or implicit, among participants. It also serves as a dynamic basis for joint activities of various sorts, from simple expressions of emotion to complex deliberations. Archetypally a visual phenomenon, joint 4  In cognitivist terms, theatrical performance presents a high perceptual load. On focused attention in such a context, see Lavie, Beck, and Konstantinou 2014. 5  For instance, Budelmann and Van Emde Boas 2020, 76–9 consider ‘asymmetrical’ joint attention between the audience and the Greek tragic messenger, taking an approach broadly complementary to this chapter’s. On theatrical attention in general, see especially Blair 2006, 168–9 and 2008, 61–2, McConachie 2008, 23–32, Tribble 2011, 35–40, McConachie 2013, 40–1; regarding Shakespearean theatre, see Lyne 2014. 6  Theorizing digital media, Shteynberg 2015, 581 offers a definition of ‘shared attention’ he believes more accurate, efficient, and generally applicable than joint attention as classically defined by Baron-­ Cohen 1995. Although many of Shteynberg’s observations about shared attention have application to Attic drama (especially in our increasingly electronic age), here I retain the established, narrow definition. 7  Sperber and Wilson 1995 (first edition 1986) introduced this compact, alliterative phrase.

Seeing Together: Joint Attention in Attic Tragedy  175 attention has been observed or hypothesized across the other senses as well.8 Joint attention has been studied most extensively in relationship to developmental psych­ology and autism spectrum disorders, but as practised among mature neurotypical subjects, studies suggest joint attention has broad implications for Theory of Mind, common knowledge, cooperative action, emotional response, and even language itself.9 The theatrical ramifications of joint attention are wide-­ranging. Applicable to performance in general, joint attention is especially pertinent to the conditions of ancient drama, when plays were performed in broad daylight, enclosed within a roughly semi-­circular seating area where spectators simultaneously observed one another as they watched the performance onstage.10 Although it has become something of a truism to note, it is nevertheless telling that the Greeks called this place, so favourable to joint attention, the theatron, the ‘seeing-­place’, and its attendees theatai, ‘spectators’. These words reflected and shaped ancient cultural attitudes about dramatic experience and reveal that Greek theatre-­goers recognized themselves as an important, even integral, part of the spectacle. In theatres of the ancient Mediterranean, far more than those of today, vision was manifestly collective and attention, correspondingly, joint. Although our understanding of its neurological mechanisms remains inchoate, empirical studies demonstrate that joint attention promotes information processing, improves memory, and generally enhances what might be called ‘engagement’.11 Laboratory findings are not always applicable ‘in the wild’, nor are the dynamics of something as complex as visual attention unaffected by culture or convention.12 Still, by adhering to a minimal definition—­namely, mutually manifest attention—­the concept of joint attention provides a useful framework for considering cognitive processes per­tin­ ent to the ancient theatre.13 8  On joint attention (henceforth ‘JA’ in the notes) and communication, see Clark 1996, 50–67. On JA and music, see Cochrane 2009 and Zangwill 2012; on JA and touch, see Batero 2016. 9  On JA, autism, and Theory of Mind, see Baron-­Cohen 1995. JA has many intentional effects, from advancing discourse (see Clark 1996) to anticipating collective remembrance (see Seeman 2016). Eilan et al. 2005 provide several chapters on the broader relevance of JA. On group attention and emotional intensification, see Shteynberg et al. 2014. On JA and literary studies, see especially Tobin 2008, Currie 2010, 86–108, Polvinen 2013, and Lively 2016. 10  Rehm 2002, 37–44 observes that the theatre was similar in both form and function to other archetypal locations of joint activity in Athens, such as the pnyx or bouleutêrion. 11  Seeman 2011 offers a useful overview of core issues regarding JA. On JA and information processing, see especially Kim and Mundy 2012, 1–2. On JA and working memory, see Gregory and Jackson 2017. On the ‘amplification’ of shared experiences entailing JA, see Clark et al. 2014. For an overview of the potential consequences of the related concept of ‘shared attention’, see Shteynberg 2015. 12  Gibson 1979 issues the classic warning of difficulties encountered when assessing cognition ‘in the wild’, although Hutchins 1995 revisits the issue more optimistically. On the difficulties of applying laboratory models of attention to theatrical experience, see Hamilton 2018, 216–18. 13  It is generally assumed that the modern minds, which form the empirical basis of current neuroscientific research, are sufficiently close to those of the ancient Mediterranean: see, for instance, Larson 2016, xiii. But as culture undeniably determines (social) vision, caution must be exercised when extrapolating laboratory findings not only onto the stage but also onto ancient minds. On the composite nature of tragic vision, see Noel 2019b.

176  A. C. Duncan Work in the cognitive sciences typically seeks to identify and isolate specific consequences of joint attention, but here I am concerned with joint attention’s composite effects. In particular, I aim to explore ways playwrights might use joint attention to involve spectators more deeply within the dramatic experience. I use the word ‘involve’ since it captures the social and affective consequences of joint attention central to the claims of this chapter.14 Socially, by definition, joint attention places theatre-­goers within intersubjective networks based upon their common focus, involving individuals within established or emergent communities of viewers. Affectively, joint attention fosters conditions in which theatre-­goers feel themselves personally involved in issues dramatized by the play: entangled in passively witnessing a coup, for instance, or groping through the fog of an am­bigu­ous report. Aligning (or contrasting) the mental experiences of spectators not only with their fellow theatre-­goers but also with those of the minds represented onstage, joint attention’s ramifications affect the phenomenology of theatrical experience in ways impossible to disentangle.15 Although joint attention is operative to some degree during the entire theatrical experience, its scope may be narrowed for special effect. Theatre-­makers create such ‘spotlight’ moments through diverse means, the majority of which fall outside actors’ lines—­and therefore, for the study of ancient drama, (mostly) beyond analytical reach. Some methods, however, are verbally encoded. It is a certain set of these expressions, which I will be calling ‘sight invitations’, that this chapter draws upon to frame and illustrate joint attention’s special effects on Attic drama. Let sight invitations be defined as capaciously addressed verbal utterances, the ostensible goal of which is to direct visual attention towards a discrete target within space. By ‘capaciously addressed’ in a theatrical context, I mean that even when an individual or group onstage is the ostensible recipient of the invitation, the utterance leaves open the possibility of involving further hearers (including, I will suggest, theatre-­goers) who consider themselves addressed and are in this way ‘hailed’ by the utterance.16 Although relatively simple in its expression, this definition applies to such a large and varied set of phrases in Greek drama that it confounds systematic study. Many methodological obstacles stand in the way of establishing a clear boundary for the set: not only are numerous lexemes involved, most obviously so-­called ‘verbs of seeing’ (e.g. horan, skopein, leussein, etc.) and their compound forms, but these verbs may also be couched in a variety of syntactical constructions, including imperatives (e.g. ‘Look!’), interrogatives (‘Don’t you see?’), and hortatory subjunctives (‘Let’s see . . .’). Beyond or in conjunction 14  I also owe the term to Hedreen 2007, whose study of Athenian eye-­cups has many inter­sub­ject­ive resonances. 15  On theatrical phenomenology, see especially States 1985; for a phenomenological approach to the space of Attic tragedy, see Weiss 2020. 16  On ‘hailing’ in Attic drama, see Jacobson 2011, 16.

Seeing Together: Joint Attention in Attic Tragedy  177 with such phrases, deictic expressions also direct visual attention; however, much like seeing verbs, demonstratives cannot always be determined as having visual significance unambiguously.17 Despite the lexical and grammatical variety of sight invitations, unity may be found in what speech-­act theorists call the expression’s illocutionary force—­that is, the speaker’s perceived intention in speaking.18 Sight invitations openly direct the visual attention of an indefinitely-­sized audience, and are therefore a primary means of focusing joint attention in the theatre. Although the ability to locate a common target in theatrical space is essential for sight invitations’ ability to focus spectators’ joint attention, the object itself can be immaterial. To return to this chapter’s opening metaphor, I am concerned here with the spotlight itself rather than the objects it illuminates. Indeed, sight invitations may even focus joint attention on invisible objects. Take for example the well-­known soliloquy in Act II of Shakespeare’s Macbeth, in which the Scottish chief asks, ‘Is this a dagger which I see before me?’ Although Macbeth is alone onstage, the speaker’s capaciously addressed question invites hearers to attend visually to what (in most productions, at any rate) is a physically absent target, fixed in dramatic space only by the actor’s gaze and gesture. Macbeth’s sight invitation involves theatre-­goers powerfully in the first of several hallucinations in the drama, and the scene is of crucial dramatic importance. Putting differences in dramatic conventions between Elizabethan and fifth-­century Attic theatre aside, the fame of this example from Macbeth not only demonstrates the impact sight invitations can have on audiences, it also hints at their particular usefulness in calling attention to characters’ altered mental states—­ a point to which we shall return. Rather than attempt a systematic study of sight invitations in Attic tragedy, here I offer close readings of a few well-­known and exemplary scenes, insights from which may be generally applied. First, a speech from Aeschylus’s Libation Bearers introduces sight invitations in their fifth-­century dramatic context, demonstrating how a series of ambiguously addressed directives gradually involves theatrical audiences within intersubjective networks that seem to include those in the play. Next, two scenes from Sophocles’ Women of Trachis show how sight invitations may entangle theatre-­goers not only within a community of spectators practising joint attention but also in the drama’s core thematic concerns of know­ ledge and vision. Third, two passages from Euripides’ Bacchae reveal the ironies and epistemic limits of joint attention in cases when, as with Macbeth’s ‘air-­drawn dagger’, the speaker’s vision differs significantly from his or her addressees’. In  scenes bookending the fall of the Theban royal house, sight invitations

17  In Roman drama, the Latin demonstrative ecce is far more frequent than imperative forms such as vide(te), etc., and almost supersedes them entirely as a sight invitation. 18  Austin 1965, 98–9, introduces and defines the concept of illocutionary utterances, which are expanded by Searle 1969; see my discussion below, Section 10.2.

178  A. C. Duncan underscore the ‘doubled’ or ‘blended’ nature of theatrical vision, as it becomes clear that attention which is ostensibly joint is, in fact, practised upon radically different premises. Yet, even when its epistemic and social benefits are called violently into question, joint attention retains its affective power. Its resolution underscores the emotions of a particularly horrific recognition scene. Altogether, examples from across Athens’ three major playwrights illustrate the diverse ways joint attention involved audiences in fifth-­ century tragedy; commonalities between these scenes, furthermore, suggest how sight invitations first developed to exploit spectators’ cognition in theatre had an important afterlife as a literary trope capable of conveying a characteristically tragic combination of vision, intersubjectivity, and extreme emotion.

10.2  Sight invitations as speech acts and catalysts for joint attention By what mechanisms does an invitation to ‘Look!’, uttered by a fictional character in performance, direct theatre-­goers’ collective attention? The communicational dynamics are hardly simple, as the meanings of the script quickly multiply in per­ form­ance. One may begin, however, by asking how the words themselves are received by the theatrical audience. In contrast to their contemporary Old Comedians, who revelled in flaunting the convention, Attic tragedians were reluctant to break the ‘fourth wall’ separating actors and audience to address theatre-­goers expressly as theatre-­goers.19 The question of actors’ asides and other forms of direct address in fifth-­century tragedy is relevant to the present discussion, but special cases of more or less overt references to the performance context should not distract from the broad and imaginative collusion always taking place between theatre-­goer, performer, and playwright, upon which theatrical communication ultimately rests.20 Audience address does not need not to be direct in order to be effective, and playwrights have many tools at their disposal to steer their spectators’ gaze. One 19  Taplin 1977, 394–5 summarizes the situation: ‘there is not a single place in surviving tragedy where the world of the play expressly acknowledged the world of the audience’, but ‘all the words of every tragedy are meant for the ears of the audience’ (emphasis original). Semiotic accounts of theatrical communication, such as that of Segre 1980, ingeniously place audience address from the playwright within a structuralist framework. But the simplicity and flexibility of cognitivist models, and in particular ‘conceptual blending’ as championed by Fauconnier and Turner 2002, have begun to call some received distinctions into question: see McConachie 2008. 20  On asides and similar practices in Attic tragedy, see Mauduit 2015 and Paré-­Rey 2015 more generally. In Anglophone circles, the classic work on the topic remains that of Bain 1977, whom Chapman 1983, Dedoussi 1995, and others follow in using the phrase ‘dramatic illusion’ to describe the maintenance of pretence. Since theatre-­goers are in fact willing and active co-­creators of the fiction (a truth all of these authors recognize), dramatic collusion is the more accurate expression.

Seeing Together: Joint Attention in Attic Tragedy  179 such tool is to place theatre-­goers in communicative positions that mirror or extend those of characters onstage, an alignment of actor and audience that occurs whenever characters themselves become ‘internal’ spectators within the scene. At such moments, certain phrases may exploit the overlapping visual-­ spatial circumstances of theatre-­goers and fictional characters, conflating play-­ internal and -external audiences in ways that—­ at least from a practical or behavioural perspective—­invite theatre-­goers to participate in (and even partly identify with) visual communities in the fiction of the play. To help analyse and explain how sight invitations verbally involve spectators, I  adopt terms and concepts from the speech-­ act theory first framed by John  L.  Austin and subsequently developed by John Searle.21 For Austin and Searle, invitations, together with commands and requests, constitute a category of speech act called ‘directives’. Linguists today differ on the category’s precise boundaries, but for our purposes it will be sufficient to follow Searle in defining directive utterances as ‘attempts . . . by the speaker to get the hearer to do something’.22 In order to function properly, directives must be configured to match salient circumstances of communication, known more technically as the pragmatics of the discourse. An important piece of pragmatic information for any directive is its addressee: those who not only (over)hear the directive but also are intended to hear the directive and understand it to apply to themselves. In theatrical communication, addressees may be determined verbally by the script (via grammatical number, vocative phrases, etc.) or through various non-­verbal cues in performance (e.g. gestures, intonation, etc.). However, ambiguities of language and gesture often leave the identity of the addressee(s) incompletely determined. Ancient theatre-­goers, not unlike modern readers, relied on processes of elim­in­ ation to determine, sometimes gradually, a directive’s addressee. In such circumstances, pragmatic evidence that renders a directive inappropriate (what Austin labels ‘infelicitous’ and Searle ‘defective’) to an otherwise viable addressee offers precious assistance in narrowing a directive’s scope. In particular, two of Austin’s so-­called ‘felicity conditions’ provide useful distinctions for interpreting the communicative dynamics of sight invitations within dramatic performance. As systematized by Searle, these conditions are defined as follows: 21  Outside linguistics circles, Austin 1965 and Searle 1969 remain classic works. Searle 1969, 78, cautions against applying speech-­act theory to such ‘parasitic’ forms of language as ‘play-­acting’, but it is the very parasitism of sight invitations (i.e. their ability to communicate at two different levels, simultaneously) that is of interest here. I am not suggesting theatre-­goers consider themselves directly addressed by the fictional character’s speech act; rather, I propose that a strategic and sufficiently close overlap of pragmatic conditions between fictional and real addressees enables theatre-­ goers to im­agina­tive­ly identify with (or, in the terms of Fauconnier and Turner 2002, 267, ‘selectively project’ themselves as) ‘play-­internal’ spectators. 22  Searle 1976, 11.

180  A. C. Duncan the preparatory condition: The Speaker believes the Hearer is able to do the Act; and the essential condition: The Speaker’s utterance counts as an attempt to get the Hearer to do the Act.23

Putting aside for a moment the knotty question of who is the ‘Speaker’ in theatrical communication, the preparatory condition dictates that the vast majority of tragic commands (e.g. ‘Stop!’, ‘Help!’, ‘Give!’, ‘Dance!’, etc.) are defective when addressed to theatre-­goers who, by well-­established convention, do not interrupt stage action. By contrast, the act of seeing not only is non-­disruptive but also was understood to be the paradigmatic act of Greek theatre-­goers (theatai, or ‘spectators’, as mentioned above). Sight invitations, alongside similar commands (‘Listen!’, ‘Consider!’, etc.) constitute a special set of directives which do not fail the preparatory condition when addressed to spectators. Although it may at first seem banal, the ‘essential condition’ is in fact also useful for framing the level(s) at which playwright, actor, and character communicate as ‘Speaker’ with the dramatic audience. Scenes from Aristophanic drama illustrate how phrases uttered in the fifth-­century theatre might be interpreted as communications originating from the character (generally), playwright (in the parabasis, especially), or even actor onstage—­sometimes in rapid, unmarked succession.24 Differences in generic conventions notwithstanding, Old Comedy reveals the complexity of identity inherent in theatrical communication as well as the capacity of fifth-­century Athenian audiences to attribute statements to a var­iety of ‘Speakers’.25 Directives which satisfy both preparatory and essential conditions when addressed to theatre-­goers have the unique potential to serve as  verbal conduits for overlapping speech acts. Operating simultaneously at several communicative levels within a shared theatrical space, sight invitations blur boundaries between fictional and real addressees to such an extent that subjects’ visual activity may be said to be practically joint even as the conventional mimetic boundary of the fourth wall is maintained. Speech-­act theory provides one account for how sight invitations might verbally communicate across the fourth wall, but this is not the only mechanism by which such expressions focus theatre-­goers’ attention. Even without words guiding them, audiences routinely follow the gaze and gesture of figures onstage, directing their collective attention in ways difficult to distinguish from joint attention as practised outside the theatrical context. In this process the composite or ‘blended’ figure of the embodied actor-­character is a crucial common node, linking networks of 23  Following Searle 1969, 60–71; cf. Austin 1965, 136–7. 24  Cf. Ar. Thesm. 1015 and 1060, with Sommerstein’s (1994) notes ad loc. These passages exemplify how the layered identities of speaker, addressee, and location could destabilize the pragmatics of theatrical communication. 25  Within, but significantly also beyond, Old Comedy, attributions of tragic characters’ statements and opinions to their respective playwrights suggests a low threshold for authorial communication.

Seeing Together: Joint Attention in Attic Tragedy  181 real and fictive viewing subjects by participating in simultaneous (although theoretically separate) ‘mutually manifest’ attention within each group.26 In ancient drama, the fusion of actor and character was both mediated and marked by an outsized mask—­a visage which, although a patently artificial means of signifying the fictional world of the play, nevertheless perfectly tracked the direction of the performer’s gaze underneath.27 Peter Meineck has labelled the Greek dramatic mask a ‘mind tool’ fashioned to direct onlookers’ visually acute, foveal, gaze.28 If indeed the mask is so important for fostering focused joint attention in the theatre, then sight invitations, which function almost as spoken stage directions that determine the orientation of the performers’ heads, are particularly powerful (and verbally encoded) catalysts for shared attention between actor and audience. Whether through purely verbal or through multi-­modal means, then, from at least the time of Aeschylus sight invitations were used to activate focused joint attention in the Attic theatre, involving audiences at moments of special dramatic and thematic interest. It remains, now, to explore the dynamics of these expressions within the context of their works.

10.3  Capacious invitations and conflated addressees in Libation Bearers In a climactic scene from Libation Bearers, a string of sight invitations surreptitiously involves theatre-­goers within the intersubjective visual community of the play. Following a tense series of events culminating in the murder of his mother Clytaemnestra and her lover Aegisthus, Orestes emerges out of the palace and into the public eye to exhibit the bodies of the slain. Besides their corpses, Orestes exhibits a large piece of fabric which he claims had fatally ensnared his father. The speech he offers at this crucial juncture is remarkable for not only its prominent use of sight invitations but also the ambiguity of their addressee. Orestes proclaims: Look at the twin tyranny of the land, the father-­killing destroyers of the house… Look again, those hearing of these evils,             980 26  On theatre as a blended space, see the chapters by Michael Carroll and Hanna Gołąb in this volume. On the blend of actor and character, see Fauconnier and Turner 2002, 266–7 and McConachie 2008, 44. 27  Masks restrict peripheral vision in particular: see Vervain and Wiles 2009, 270 and Meineck 2018b, 131. Those who experienced dancing in a dramatic chorus—­no small share of the Attic audience, by some accounts—­would have had personal and embodied experience of the mask’s narrow, frontal scope. 28  Meineck 2018b, 96–8 and 2019.

182  A. C. Duncan At the device, my wretched father’s bonds, The fetters of his coupled hands and feet. Spread it out and, standing around in a circle, Show the man’s coverings, so that father may see— Not my father, but the one beholding all these things, Helios (the Sun)—the unholy work of my mother, so that he may at some point stand as my witness, that I have justly pursued this murder, my mother’s . . .

985

ἴδεσθε χώρας τὴν διπλῆν τυραννίδα πατροκτόνους τε δωμάτων πορθήτορας… ἴδεσθε δ᾽ αὖτε, τῶνδ᾽ ἐπήκοοι κακῶν,980 τὸ μηχάνημα, δεσμὸν ἀθλίῳ πατρί, πέδας τε χειροῖν καὶ ποδοῖν ξυνωρίδος. ἐκτείνατ᾽ αὐτὸ καὶ κύκλῳ παρασταδὸν στέγαστρον ἀνδρὸς δείξαθ᾽, ὡς ἴδῃ πατήρ, οὐχ οὑμός, ἀλλ᾽ ὁ πάντ᾽ ἐποπτεύων τάδε985 Ἥλιος, ἄναγνα μητρὸς ἔργα τῆς ἐμῆς, ὡς ἂν παρῇ μοι μάρτυς ἐν δίκῃ ποτὲ ὡς τόνδ᾽ ἐγὼ μετῆλθον ἐνδίκως φόνον τὸν μητρός. . .  (Aesch. Cho. 973–4, 980–9)29

Orestes’ presentation illustrates the forensic value of joint attention. Like an expert orator, Orestes uses a highly visual discourse to make post-­factum witnesses of his audience, bolstering his credibility and uniting his audience through their shared subjectivity.30 Uniquely theatrical powers of joint attention, however, are also at work as Orestes’ repeated sight invitations gradually bridge the mimetic gap between stage characters and theatrical audience. Through accumulating referential ambiguities, theatre-­goers become entangled not only within the intersubjective visual community of the mythical scene but within its politics as well. As Orestes emerges from the palace, he utters a plural, but otherwise pragmatically bald, sight invitation: ‘Look!’ (ἴδεσθε, 973). ‘Who’, both theatre-­goer and reader might well ask, ‘is being invited to look?’ It is the performer’s prerogative to specify the addressee through gesture or in­ton­ ation; and yet, perhaps more than any other example from extant tragedy, this scene benefits from exploiting the pragmatic ambiguity of Orestes’ sight 29  Following Page 1972. 30  On such ‘visual discourse’ in Attic oratory, see O’Connell 2016, esp. 113–18. For a cognitive approach to the verbal imagery of this scene, see Duncan 2021. Orestes establishes what Clark 1996 calls ‘common ground’ with his audience, a concept discussed further below.

Seeing Together: Joint Attention in Attic Tragedy  183 invitation. Several lines later, Orestes reiterates, ‘Look, again’ (ἴδεσθε δ᾽ αὖτε, 980), now directed towards a verbally explicit, although rather tautological, addressee: ‘those hearing of these evils’ (τῶνδ᾽ ἐπήκοοι κακῶν, 980). This curious phrase has sparked much debate over the staging of the scene and the pragmatic limits of Orestes’ intended audience.31 But even if one settles upon a plausible addressee, such as the Chorus or a broader (staged or imagined) Argive community, the flatly plural command remains open to incorporating further viewing subjects, as indeed it soon does. Instructing his attendants to unfurl the cloth, Orestes next calls upon an extraordinary spectator: ‘so that father may see— / not mine, but the one who sees all these things / the Sun’ (985–6). Feinting initially towards Agamemnon’s underworld shade, Orestes pivots upwards in an appeal to an all-­seeing figure which, in this quasi–­juridical context, one might well expect to be ‘father’ Zeus. In unexpected enjambement, however, Orestes names the viewer par excellence of early Greek poetry, the sun god Helios, whose remarkable vision is, itself, visible in the form of light.32 The sun’s presence in the sky, like the blended figure of the actor-­character, presents a material anchor that helps bridge the mimetic divide between mythical Mycenae and fifth-­century Athens. Moreover, as a viewer himself, Helios serves as a crucial node connecting intersubjective visual communities both real and imagined. And yet, by appealing to Helios’ transcendent spectatorship in this roundabout way, Orestes hints at the presence of other supernatural onlookers, such as Zeus, who see while remaining themselves unseen—­a position not unlike that of theatre-­goers or, as soon becomes a matter of some urgency, the Erinyes. During his remarkable exposition of the bodies and the cloth, Orestes confounds the limits of his addressee to such an extent that, when he marks his departure as a suppliant to Delphi with a final capacious sight invitation, ‘Watch me’ (ὁρᾶτέ μ᾽, 1034), theatre-­goers might question whether they themselves are being asked to bear witness. Ironically, it is at this very moment that Orestes’ own vision becomes idiosyncratic and hallucinatory. He claims to see the terrifying figures of the Erinyes clearly before him (σαφῶς, 1054; αἵδε, 1057), a vision the Chorus dismiss as mental disturbance. Orestes replies ‘You do not see these women, but I do’ (ὑμεῖς μὲν οὐχ ὁρᾶτε τάσδ᾽, ἐγὼ δ᾽ ὁρῶ, 1061), the linked 31  Orestes’s capacious entreaties and the awkwardness of addressing a chorus of slave women as ἐπήκοοι, a term more naturally applied to the enfranchised male citizenry of a democratic polis, have led some to suppose a crowd of supernumerary Argives onstage for a grand finale. Although Taplin 1977, 357–8, Garvie 1986, ad 973–1076, and McCall 1990, 25 all dismiss such grandeur, Taplin allows for ‘an invisible audience of citizens’—a role for which theatre-­goers might feel some natural affinity. 32  For Garvie 1986, ad 984–6, the reference to a father is ‘initially ambiguous’. On Helios as a paradigmatic spectator, see Blundell et al. 2013, 35n29 and my discussion (with notes) of Soph. Trach. 99–102 below. On the Sun and the ‘notion of vision’s visibility’ widely held across the ancient Greek and Roman worlds, see Bielfeldt 2016, 122–6. On sight and the ancient senses, see Squire 2016 and, on reflexive vision, Grethlein 2016.

184  A. C. Duncan pronouns and particles marking this sudden schism within the intersubjective visual community. The chorus’ inability to see the Furies disrupts the established visual alignment between Orestes, play-­internal characters, and theatre-­goers, a dis­sol­ution of joint attention that underscores Orestes’ cognitive alterity. But as in the dagger scene of Macbeth, through gesture and gaze (note deictic τάσδ᾽, 1061) the actor playing Orestes may enter into new intersubjective communion with the audience as they, too, visualize the Erinyes onstage. Exploiting the theatrical potential of vague and capacious addressees, this series of directives from Libation Bearers provides an orienting example of how sight invitations involve theatre-­goers in moments of marked and dramatically significant joint attention. When theatre-­goers’ identity as spectators overlaps significantly with that of figures onstage whose visual space they share, distinctions between play-­internal and -external viewing communities become blurred. Aided by common spatial orientation and shared reference points such as the panoptic Sun, blended actor-­characters and involved theatre-­goers participate in meaningful visual communion even across the theatre’s conventional mimetic divide. Orestes’ sight invitations do more than leverage the forensic political power of joint attention: they engage theatre-­goers within the political and thematic structures of the play.

10.4  Joint attention and joint knowledge in Women of Trachis Although sight invitations occur in every extant Attic tragedy, Sophoclean drama, and Women of Trachis in particular, employs the set to an unmatched degree. This frequency reflects in part the complex, profound, and often thematically central connections between sight and knowledge across Sophocles’ dramatic oeuvre.33 Indeed, as mentioned above, the remarkable semantic overlap between sight and knowledge in Greek often makes it difficult to determine whether a given ‘sight invitation’ is principally visual in its significance. A verb such as σκοπεῖν, for instance, might in various contexts mean ‘behold’, ‘consider’, or both sim­ul­tan­ eous­ly. Attention is practised in any event, and Sophocles exploits the slippage between jointly seeing and jointly knowing to substantial dramatic effect. As examples from Women of Trachis will illustrate, sight invitations (as I will continue to call them) involve spectators not only within the intersubjective networks of the play but in characters’ epistemological quandaries as well. When deployed at moments of special visual ambiguity or obscurity in the play, sight invitations may call attention to the epistemic gaps of individual perception, highlighting 33  On vision in Sophocles, see Seale 1982. On vision and knowledge in Greek tragedy generally, see Thumiger 2013.

Seeing Together: Joint Attention in Attic Tragedy  185 ways one’s own vision, just as another’s verbal report, may be unsubstantiated and unreliable unless practised in the context of robust joint attention. Since vision in Attic drama is fundamentally mediated by characters’ verbal reports, this process has metatheatrical resonance as well, as sight invitations engage characters and theatre-­goers in simultaneous, parallel processes of interpretation. The unstable relationship between sight and knowledge is a recurrent theme in Women of Trachis.34 Deianeira, the wife of Heracles, introduces the motif early in her prologue. Recalling the fight between Heracles and the river god Achelaos, the victor of which would claim her as wife, Deianeira apologizes that she cannot narrate an event which she could not even bring herself to view or recall: ‘The manner of the struggle I cannot say, for I don’t know it—­whoever was sitting there not terrified of the sight, he could tell you.’35 Deianeira’s apology articulates the affective constraints placed on individual vision, as the trauma of the event either prevented her perception at the time or has subsequently blocked its recollection. Thinking beyond her individual experience, however, Deianeira gestures hypothetically towards an indefinite (ὅστις), impassive (ἀταρβής), and almost theatrical (cf. θακῶν, ‘seated’) spectator. In the tragic universe of the play that her prologue begins to establish, public acts are framed as being open to any number of real or imagined viewers, but with little sense of ‘mutually manifest’ awareness between these subjects. The unsettling combination of limited individual vision and looming panoptic presence is further developed when the eponymous Chorus take the stage. Like Orestes in Libation Bearers, the women of Trachis invoke Helios, here not to serve as fellow witness, but rather to join their search for the missing Heracles, whose belated return their mistress Deianeira anxiously awaits. The Chorus ask Helios to disclose the hero’s location, singing ‘Speak, you who are most powerful with respect to the eye’, but their prayer elicits no response.36 In contrast to the confidence Orestes places in joint attention in Libation Bearers, in Women of Trachis, the Sun god’s silence underscores the pervading sense of limited and disconnected vision within the play. Vision is simultaneously fundamental and flawed across many of Sophocles’ works, but as dramatized in Women of Trachis, the epistemological challenge posed by vision is essentially one of community con­ firm­ation. Given the unreliability of any individual’s senses, true knowledge must be predicated upon networks of viewers and authorities. In such a context, sight invitations become calls for corroboration as much as coordination, and joint attention serves as a crucial check against the errancy of private perception and interpretation.

34  On the play’s epistemological concerns, see Lawrence 1978, Roselli 1982, and Heiden 1989. 35 Soph. Trach. 21–3, καὶ τρόπον μὲν ἂν πόνων | οὐκ ἂν διείποιμ’· οὐ γὰρ οἶδ’· ἀλλ’ ὅστις ἦν | θακῶν ἀταρβὴς τῆς θέας, ὅδ’ ἂν λέγοι, following Lloyd-­Jones and Wilson’s 1990 text here and throughout. 36  ὦ κρατιστεύων κατ᾽ ὄμμα, Soph. Trach. 102.

186  A. C. Duncan The play’s first sight invitations occur in its first stasimon, at the end of a paean to Apollo and Artemis. Transitioning away from divine praise, the chorus-­ women take a sudden, Dionysian turn, adopting the personae of ­maenads before turning to address Deianeira. Both transitions are punctuated by sight invitations: Ch. (Sung) Look, it arouses me—­ Euoi!—the ivy, spinning me around in Bacchic contest! Io, io, Paean! (Spoken or chanted) See, see, dear lady! You may look on these things directly before your eyes, in all clarity. De.  I see, dear women, and it has not escaped my regard, the sight of this band. Χο.  ἰδού μ᾽ ἀναταράσσει ἐὐοὶ, ὁ κισσὸς ἄρτι Βακχίαν ὑποστρέφων ἅμιλλαν ἰὼ ἰὼ Παιάν· ἴδε ἴδ᾽, ὦ φίλα γύναι· τάδ᾽ ἀντίπρῳρα δή σοι βλέπειν πάρεστ᾽ ἐναργῆ. Δη.  ὁρῶ, φίλαι γυναῖκες, οὐδέ μ᾽ ὄμματος φρουρὰν παρῆλθε, τόνδε μὴ λεύσσειν στόλον. (217–26)

The remarkable self-­reference to choral song and dance in this passage has received much attention, but scholars have largely passed over the sight vocabulary that involves theatre-­goers and both frames and enables the reflexive poetics of the scene.37 The chorus-­women begin by singing ‘Look!’ (ἰδού, 217), a forerunner to a rapid series of sight invitations that, in their overall effect, sustain focused joint attention throughout the scene.38 Echoes of this initial ἰδού can be heard five lines later in the singular active imperative, ἴδε, uttered as the choral lyrics draw to a close in brief dimeters.39 Like the indeclinable interjection ἰδού (and like

37  See especially Henrichs 1995. 38  Ancient texts did not mark tonality, leaving the precise meaning of the word unclear. On the accent of ἰδού, cf. Hdn. 1.417.27. Most modern editors join Lloyd-­Jones and Wilson in printing an oxytone rather than circumflex accent (i.e. ἰδού, not ἰδοῦ). The two forms are related, ἰδού being the fossilization of the earlier aorist middle imperative, ἰδοῦ (cf. French voilà, etc.): see Nordgren 2015, 20–2. By the fifth century, the interjection had become divorced from the verb’s original visual meaning (cf. e.g. S. Aj. 860, LSJ s.v. ἰδοῦ II), but its etymology remained obvious. In melodic performance, a firm aural distinction between ἰδού and ἰδοῦ might have been difficult to maintain: for modern comparisons, see List 1961 and Wong and Diehl 2002. At any rate, the linguistic ambiguity is compounded by the general polysemy of choral lyric and choral self-­presentation, on which see Peponi 2015. 39  The hiatus preserved in the manuscripts’ ἴδε ἴδ᾽ at 222 has given editors pause, leading some to print only a single command. Davies 1991 ad loc. entertains, but does not endorse, Schroeder’s emend­ation, ἴδοῦ ἴδ᾽. The precise text would colour the meaning and reception of this directive in context, but is ultimately immaterial to my larger argument, since any of these readings would meet my definition of a sight invitation.

Seeing Together: Joint Attention in Attic Tragedy  187 Orestes’ initial directive in Libation Bearers 978, ἴδεσθε), the addressee of ἴδε is initially ambiguous, inviting theatre-­goers to share in the subjective position of a play-­internal viewer, just as the Chorus, themselves, project awareness of the Dionysian festival context. A vocative phrase, ‘dear woman’ (ὦ φίλα γύναι), soon specifies that it is Deianeira who is addressed, reasserting the dramatic space as Trachis. Yet even as the addressee is pinned down, the intended target of her gaze is left undetermined. Towards what should Deianeira (and formerly hailed theatrical spectators) now look? Fulfilling this expectation, a deictic pronoun, ‘these things’ (τάδ’, 223) opens the next verse, but this vague referent, if left undetermined by gesture or intonation, introduces new ambiguities. The chorus-­leader is curiously emphatic about the self-­evident nature of this visual target, which is described as being ‘directly before your eyes’ (ἀντίπρῳρα . . . βλέπειν, 223) and ‘in all clarity’ (ἐναργῆ, 224). It is ironic, then, that the pronoun’s referent has proven so difficult for modern scholars to determine. While some interpret the deictic as anaphoric, others prefer a cataphoric referent, most obviously the arrival of the messenger Lichas and his company.40 In the event that verbal uncertainty is maintained in per­form­ance, however, this ambiguous visual target reveals another important consequence of theatrical sight invitations. Once a directive to ‘Look!’ is uttered and a visual target is needed to complete the meaning of the scene, a collaborative and dynamic effort begins among viewers to settle upon a commonly recognized object for their joint attention. This collective, provisional process of spectatorship is comparable to what Herbert Clark identifies as a fundamental communicational imperative to establish ‘common ground’—the mutually recognized basis for continuing conversation.41 Although theatre is a remarkably complex and sophisticated form of communication, it relies upon the basic need to establish clear and mutually recognized ‘common ground’ to support the ongoing imaginative collusion between actors and audience. Clark’s terminology is recent, but examples from our earliest texts (most notably Cassandra in Agamemnon, but scores of other cases as well, especially from Old Comedy) make clear that Athenian playwrights were fas­cin­ ated by communication breakdowns resulting from such unstable common ground as vague pronouns, metaphorical expressions, and the like. Ambiguities may lead to productive ironies and strategic misunderstandings in a carefully scripted exchange, but when presented haphazardly to the audience they wreak

40  Jebb 1892, 37 and Easterling 1982, ad loc. treat τάδ’ anaphorically, referring back to what Jebb calls ‘the good tidings (180ff.) of which their [sc. the chorus women’s] minds are full’—verbal reports that will be instantiated with the arrival of Lichas and his train. Davies 1991 and Kamerbeek 1970 suggest instead that τάδ’ (or στόλον) is the subject of παρῆλθε, 226, with Davies citing Taplin’s 1977, 174 discussion of this scene as a case where an approaching entry is announced by the concluding chorus. 41  See Clark 1996, 50–67.

188  A. C. Duncan havoc upon a play’s intelligibility. It is a quality not only of a good theatre-­maker, then, but also of a good theatre-­goer, to limit and provisionally determine chaotic uncertainties as they occur through appealing to our collective communicative need for common ground. When important ambiguities are left unresolved in the theatre, they are often settled, practically and provisionally, through the collective response of the audience—­a process in which joint attention has an important role. Driven to establish common ground and to appeal to the wisdom of the crowd, individual theatre-­goers seek consensus through the audience’s socially scaffolded, cognitively distributed interpretation of the stage. During a live dramatic performance in which open verbal deliberation between spectators is not appropriate, this process is particularly dependent upon joint attention, as individual spectators quiet­ly consult the reactions of their fellow theatre-­goers. Sight invitations create intersubjective communities not only through capacious addressees but through involving theatre-­ goers in collective deliberative processes of interpretation as well. These early scenes from Women of Trachis reveal ways sight invitations involve theatre-­goers in the interpretative challenges of vision dramatized in the play. As the tragedy goes on, however, the constant and collective presence of the chorus (and theatrical audience) gradually re-­establishes collective vision as a means of knowledge and authority. As in Libation Bearers, sight invitations mark and enhance the dramatic climax of Women of Trachis, drawing attention once again to the spectacle of the mortally wounded body. In place of the corpses of Clytaemnestra and Aegisthus however, it is Heracles, unconscious but still alive, who is carried in upon a stretcher to be received by his adult son, Hyllus. Even covered, Heracles’ body becomes the visual focal point of the scene and through strategic use of an opaque shroud, the play reveals the body strategically in time and space to enhance its dramatic effect. As the scene begins, Hyllus guides and focalizes the audience’s vision. Overwhelmed by even the veiled outline of his father’s dying body, Hyllus cannot bear to keep silent while ‘looking upon this evil’ (κακὸν τόδε  λεύσσων, 992). Linking the broad semantics of κακός to his father’s physically present (τόδε) but visually obscured body, Hyllus’ words infuse the scenic tableau with a macabre sense of dread.42 Awakened by his son’s voice, in an emotive agony Heracles insists that ‘no one ever can say that he saw this man behave thus before’, a claim that establishes the hero’s earlier authority over not only his body but its spectatorship as well.43 Recalling the words of Deianeira’s prologue near the tragedy’s end, Heracles invokes a hypothetical visual subject (οὐδ᾽ ἂν εἷς, 1072) to ­underscore the individual and indefinite nature of vision within the play. 42  On the ‘enormous interpretive range’ of κακός, see Sluiter 2008, 4. 43 Soph. Trach. 1072–3, καὶ τόδ οὐδ’ ἂν εἷς ποτε | τόνδ’ ἄνδρα φαίη πρόσθ’ ἰδεῖν δεδρακότα.

Seeing Together: Joint Attention in Attic Tragedy  189 Having established the unprecedented importance of its spectacle, Heracles invites Hyllus to inspect his body: And now, coming close, stand near your father, consider on account of what sort of misfortunes I suffer in this way. For I will show these things out from under my coverings. καὶ νῦν προσελθὼν στῆθι πλησίον πατρός, σκέψαι δ᾽ ὁποίας ταῦτα συμφορᾶς ὕπο πέπονθα· δείξω γὰρ τάδ᾽ ἐκ καλυμμάτων. (1076–8)

At first it seems that Hyllus, alone, will be privy to his father’s unprecedented suffering. The imperatives στῆθι and σκέψαι extend a string of singular directives, stretching back many verses, addressed narrowly to Hyllus.44 Heracles’ earlier reluctance to show his pain, Hyllus’ position near his father onstage, and the established expectation of disconnected vision in the play might lead theatre-­ goers to expect their own view of the suffering hero to be limited, perhaps simply to a narrative account of Hyllus’ gaze. But this is not the case, and Heracles’ intended visual audience suddenly expands. Removing his coverings (ἐκ καλυμμάτων, 1078), Heracles employs a new set of sight invitations in rapid succession to reveal the gruesome spectacle of his tormented body. Look! Gaze, everyone, at my wracked body, See me wretched, how pitifully I suffer! ἰδού, θεᾶσθε πάντες ἄθλιον δέμας, ὁρᾶτε τὸν δύστηνον, ὡς οἰκτρῶς ἔχω. (1079–80)

The plural imperatives ‘gaze’ (θεᾶσθε) and ‘see’ (ὁρᾶτε), together with the vocative ‘everyone’ (πάντες), invite an extremely capacious viewership. Heracles uses three verbs of seeing and three deictics in just five lines—­a remarkable burst of dir­ect­ ives capable of demanding attention to this climactic scene even from the most distracted audiences. Indeed, the staging shares much in common with Orestes’ speech in Libation Bearers: like Orestes, Heracles uses sight invitations to claim for himself the memorializing and validating effects of joint attention and, like  the verbally explicit but practically vague addressee ‘overhearers’ (ἐπήκοοι, Libation Bearers 980), by addressing ‘everyone’ (πάντες) Heracles offers a sly nod 44 Namely γενοῦ, 1064; δός 1066; ἴθι, τόλμησον, and οἴκτιρον, 1070.

190  A. C. Duncan towards the theatre-­goers. Unlike Orestes’ exhibition of the fabric, however, the sudden removal of Heracles’ shroud suddenly converts individual imaginings into a collective event of mutually manifest vision—­a coup de théâtre that appears to have been a favourite feature of Sophocles’ stagecraft.45 Throughout Women at Trachis, knowledge based upon individual perception or verbal report repeatedly proves tragically limited in its scope. Miscommunications and partial observations caused his suffering, but in a final act of exhibition and self-­determination, Heracles insists upon making his tormented body the object of a radically inclusive viewership. The hero’s capacious audience does not hear about his wounds second-­hand or endure the horrific sight each alone, but through focused and communal joint attention, each and all bear shared witness to the unprecedented spectacle of Heracles’ suffering. Even so, the body in pain is seen in public only after it is first privately imagined in the mind. Like the fabric unfurled in Libation Bearers, before it is removed the shroud provides not only a common target for joint attention in theatrical space but also a blank screen upon which each viewer may project a private image of horror. Even as Sophocles exploits the social cognition of the ancient theatre, Women of Trachis remains very much also a theatre of the individual mind.

10.5  Dramatizing the limits of joint attention in Bacchae One further example from fifth-­century tragedy sheds light on the sophistication with which Attic playwrights employed joint attention when conceptualizing and dramatizing the visual dynamics of their theatre. Euripides’ Bacchae has long been understood to offer metatheatrical commentary on dramatic per­form­ance.46 Central to the play and its interpretation is a delirium which Dionysus, as god  of theatre, places upon the Theban king Pentheus and the queen mother Agave. While in the god’s thrall, these characters view people and things as having dual identities that partially and temporarily eclipse one another, a way of ‘doubled’ (or, recalling Fauconnier and Turner’s term, ‘blended’) vision with obvious parallels to theatrical spectatorship.47 What has passed unnoticed in these visually oriented and metatheatrical analyses are the ways in which Euripides employs sight invitations to dramatize the epistemic limits of such doubled or blended vision. In Libation Bearers and Women of Trachis, joint attention serves to corroborate and codify individual experience. But as a means of building 45  Cf. Soph. Aj. 866–1048, and especially 1003ff., when Teucer at last uncovers Ajax’s body. On the presentation of the corpse in Athenian tragedy and life, see Rehm 1994, 11–42. 46  On the play’s generic self-­reflection, see especially Winnington-­Ingram 1948, Foley 1980, Zeitlin 1996, and Segal 1997. 47  On vision in the play, see especially Gregory 1985, Seaford 1987, Segal 1997, Thumiger 2007b and 2013, 234–8.

Seeing Together: Joint Attention in Attic Tragedy  191 community and establishing truth, shared spectatorship fails spectacularly in Bacchae, as a divinely coordinated series of events underscores ultimately insuperable gaps between individual mental experiences. Yet even as the epistemological benefits of joint attention are called into question, its affective powers are boldly reasserted in a recognition scene in which sight invitations, once again, cast a spotlight on a tragically wounded body. Euripides bookends the fall of the house of Cadmus with two moments of ­doubled vision. Midway through the play, Pentheus, enthralled by Dionysus, sets out with the god as his guide to spy upon the Theban bacchantes who have gathered in the wilderness of Mt. Cithaeron. As he is escorted offstage, Pentheus reports seeing two suns, two cities, and even a bull’s face upon his companion (lines 918–21)—peculiar descriptions which signal that Pentheus’ vision has become idiosyncratic and hallucinatory. Although these lines have been much discussed, it is worth noting here that seeing two suns upends the established dramatic trope, observed already in both Libation Bearers and Women of Trachis, of referring to the sun as the supreme and constant spectator of tragic events, a corner­stone of theatrical joint attention. Seeing two suns, then, does more than mark Pentheus’ own delirium, it also calls into question the intersubjective networks of viewing underpinning dramatic communication. In response to his companion’s peculiar descriptions, Dionysus replies, ‘Now you are seeing what you must see’ (νῦν δ᾽ ὁρᾷς ἃ χρή σ᾽ ὁρᾶν, 924), a cryptic phrase that emphasizes the  god’s power (and, by extension, that of his theatre) to modulate vision. Dionysus articulates a higher, normative order of vision (χρή denotes both necessity and duty), established not through intersubjective networks but through personal obligation and divine fiat.48 The god addresses Pentheus alone, but his words also have a performative function, realizing an alternate, imaginary, and, to all but Pentheus, invisible world inaccessible to joint attention. But seeing as one ‘ought’ in Bacchae does little to mitigate the epistemic peril of individual perception. Soon after Pentheus’ departure from the city, a messenger reports how Dionysus exposed his protégé to his Theban followers in the wilderness, where the king was suddenly ‘seen by, more than he saw, the maenads’ (ὤφθη δὲ μᾶλλον ἢ κατεῖδε μαινάδας, 1075). In an inversion of his voyeuristic fantasy, Pentheus becomes himself the object of vision. Indeed, his own perception counts for little as his mother, Agave, believing her son to be a lion, leads her fellow bacchantes in his ritual dismemberment. The messenger reports that at this moment Agave’s eyes were rolling and her thoughts, demented (διαστρόφους | κόρας ἑλίσσουσ᾽, οὐ φρονοῦσ᾽ ἃ χρὴ φρονεῖν, 1122–3)—the play’s first signal that, like that of her son, Agave’s vision has been affected by the god.

48  I follow Diggle’s 1981 Greek text.

192  A. C. Duncan Confirming the messenger’s breathless report, Agave returns to the stage, still delirious and brandishing Pentheus’ grisly head aloft. In an extended and disturbing musical exchange, the eponymous Chorus (composed not of Theban women, but a group of Asian bacchantes, loyal to Dionysus) toy with the queen mother, stoking her false sense of accomplishment with strategic, euphemistic enquiries about her ‘hunt’. After a moment of foreboding silence, the chorus-­leader speaks, goading Agave to display her ‘victorious prize’ (νικηφόρον, 1200). In turn, the queen mother invites the people of Thebes to come and look upon her ‘catch’: Ch. Show then, poor woman, your victorious prize to the townspeople, the one you have come back carrying. Ag. Inhabitants of the fair-­towered city of the Theban land, come so that you may see the catch… Χο.  δεῖξόν νυν, ὦ τάλαινα, σὴν νικηφόρον ἀστοῖσιν ἄγραν ἥν φέρουσ᾽ ἐλήλυθας. Αγ.  ὦ καλλίπυργον ἄστυ Θηβαίας χθονὸς ναίοντες, ἔλθεθ᾽ ὡς ἴδητε τήνδ᾽ ἄγραν . . .  (1200–3)49

As observed during Orestes’ return to the stage and Lichas’ arrival, visual directives are employed at a moment of scenic transition to reframe the dynamics of inter­ sub­ject­ive vision. But whereas sight invitations in Libation Bearers and Women of Trachis hold out the promise of joint attention’s epistemic and social benefits, in Bacchae they call attention to the ultimately individual subjectivity of vision as it occurs in the mind’s eye. The gulf between what different subjects ‘see’, even as they jointly attend ostensibly to the same target in space, structures the following, final, scene. The Chorus and Agave never disagree over terms used to refer to the head she holds in her hand, but polysemy of both language and the theatre make this apparent common ground the unstable basis of deeply ironic communication. Agave’s father Cadmus soon arrives onstage and instructs servants to place the assemblage of his grandson’s remains before the palace walls. This shapeless horror lies in marked contrast to the head Agave brandishes aloft, a focal point on the stage Cadmus calls an ‘unhappy sight’ (ὄψιν οὐκ εὐδαίμονα, 1232).50 Boasting her hunting prowess, Agave invites her father to share both in her pride and in her

49  I take ὡς ἴδητε to extend the illocutionary force of the command ἔλθεθ᾽, making the entire phrase a sight invitation: compare ἴθ᾽, ἐκκάλυψον, ὡς ἴδω at Soph. Aj. 1003. Note that her addressee, ὦ . . . ναίοντες, cannot apply to transient foreign women of the chorus and therefore invites Attic theatre-­goers, as so often, to identify and compare themselves with the citizens of Thebes: see Zeitlin 1990. 50  How Pentheus’ disjointed remains are displayed is unclear. Dodds 1960, ad 1216–19 and Roux 1970, 200 place them in a veil or shroud (linceul), like the bodies of Heracles or Ajax in Sophocles. Taplin 1978, 74 prefers an open bier.

Seeing Together: Joint Attention in Attic Tragedy  193 way of seeing, proclaiming that she has returned, ‘as you see, carrying this prize’ (ὡς ὁρᾷς, τάδε | λαβοῦσα τἀριστεῖα, 1238–9). The ensuing dialogue between father and daughter maintains this emphasis on vision (cf. 1244 ἰδεῖν, 1257 ὄψιν, 1258 ἴδῃ). All the while, Cadmus, rather than correcting his daughter, offers a series of ironic, glancing blows (‘lovely is the sacrifice’, καλὸν τὸ θῦμα, 1246; ‘to a banquet’, ἐπὶ δαῖτα, 1247) which underscore the epistemic failure of their apparent joint attention. At last, Cadmus tells his daughter, ‘turn your face to this sky’ (ἐς τόνδ᾽ αἰθέρ᾽ ὄμμα σὸν μέθες, 1264), beginning a process that aligns Agave’s vision, once again, with his own. This remarkable ‘psychotherapy’ scene, too, has been much discussed, but appreciation of its dramatic power is further enhanced by recognizing joint attention’s contribution to Agave’s rehabilitation. By gazing upwards—­ and, crucially, away from all onlookers—­Agave is released from the tethers of the intersubjective networks sustaining her delirium. The social discontinuity and visual pause resets the queen mother’s vision so that when Agave finally lowers her gaze, her father’s next sight invitation, ‘Look at it properly; the effort of doing so is slight’ (σκέψαι νυν ὀρθῶς‧ βραχὺς ὁ μόχθος εἰσιδεῖν, 1279), lands with devastating emotional impact. It is not only Agave who experiences a flood of emotions at the simultaneous recognition of her son and her own culpability; theatre-­goers, too, are released from the tension of mentally sustaining contradictory visions of the scene during an extended, grotesquely defective period of joint attention. ‘Properly’ or otherwise, once again characters and audience meaningfully ‘see together’. The devastation of the house of Cadmus in Bacchae is framed by moments when Pentheus and Agave first slip away from, and then tragically back into, interpretative common ground within networks of play-­internal and -external viewers. The various thematic ambivalences of Bacchae hinge upon two competing hierarchies of sight: one based upon intersubjective joint attention, the other, upon divinely inspired, yet essentially private, vision. Each way of seeing has its own claim to truth. Cadmus’ assertion that Agave sees ‘correctly’ (ὀρθῶς, 1279) when recognizing her son is normative, based upon his own (and others’) apparent experience. And yet, Cadmus’ pronouncement is curiously parallel to Dionysus’ earlier statement that, in seeing two suns and two Thebes, Pentheus, too, was seeing as he ‘must’ or ‘ought’ (χρή, 927). As in Libation Bearers and Women at Trachis, collective viewership provides a tool for establishing intersubjective truths in Bacchae. And yet, the delirium which is at the heart of this remarkably metatheatrical tragedy reminds spectators of the fundamentally blended experience of their own theatrical vision. Theatre-­ goers, too, participate in scripted exchanges (not unlike those between Pentheus and Dionysus, or between Agave and the Chorus) carefully designed to maintain an illusion. The difference, of course, is that theatre-­goers are not only willing participants but also part of an intersubjective network whose creative, imaginative collusion is mutually manifest to all. In a process that is arguably far more

194  A. C. Duncan important than establishing objective truths, joint attention in the theatre creates and sustains the intersubjective fictions that are the bedrock of dramatic mimesis.

10.6 Conclusion To conduct the minds of one’s audience successfully across a busy stage and complex plot has been recognized as the mark of a good playwright since the classical era.51 Although attention plays a critical role in these processes, its function in ancient drama has to date been poorly theorized, perhaps because attending the stage is so different from attending the page. The phenomenological disparities between seeing and reading Greek tragedy are numerous, but the collective vision of the theatre, and the joint attention it entails, are of particular importance. At moments when spectators’ joint attention is narrowed into a focused ‘spotlight’, the act of markedly and self-­consciously ‘seeing together’ highlights and further activates theatre-­goers’ social cognition. As catalysts for focused joint attention, sight invitations provide valuable evidence for our understanding of social cognition in the ancient theatre. The close readings offered in this chapter demonstrate some of the ways Attic tragedians focused joint attention to involve their audiences more deeply in the performance. While I have highlighted in particular the social and affective consequences of such joint attention, productive work remains to be done on the connections between social cognition and—­among other aspects of the ancient theatre—­chorality, cultural memory, civic participation, and aesthetics. Transcending their theatrical origins, however, sight invitations went on to have a significant cultural afterlife in Athens. As something of a coda, I call attention to a noteworthy fourth-­century echo of tragic sight invitations in prose: the story of Leontius as recounted in Book Four of Plato’s Republic.52 Leontius, we are told by Socrates, was once walking alone outside Athens’ city walls when he unexpectedly came upon the corpses of executed prisoners. Suddenly in the presence of the abject dead (a familiar tragic situation), Leontius is torn between averting his gaze or indulging his morbid fascination. He first covers his head with fabric (παρακαλύπτοιτο, 440a1) but, overcome by desire, rushes towards the corpses with his eyes wide open (διελκύσας τοὺς ὀφθαλμούς, 440a2). Doing so, he calls out: ‘Look, you wretches, have your fill of the fine spectacle’ (ἰδοὺ ὑμῖν . . . ὦ κακοδαίμονες, ἐμπλήσθητε τοῦ καλοῦ θεάματος, 440a3–4).53 51 Arist. Poet. 1455a21–55b23) gives illustrative examples, positive and negative, for would-­be playwrights. Ar. Ran. 908–35 satirizes Aeschylus’ supposedly mismanaged stagecraft. 52 Pl. Resp. 439e–40a. On the episode and its connections to tragedy, see Liebert 2013 and 2017, 161–70. On the Republic and tragic mimesis, Halliwell 1984, Halliwell 1993, and several of the essays included in Destrée and Herrmann 2011. 53  My text and translation follow Slings 2003.

Seeing Together: Joint Attention in Attic Tragedy  195 The tragic resonances of this passage are numerous. Indeed, although used in its original Platonic context to illustrate the divided nature of the human soul, today the curious case of Leontius has become most emblematic of so-­called ‘tragic pleasure’, the paradoxical satisfaction that comes from contemplating or viewing things which are typically considered disturbing and disgusting. Leontius is ‘tragic’ not simply in his psychological circumstance, however, but in his use of language as well. Leontius’ words bear all the marks of a tragic sight invitation, from the initial interjection, ‘Look’ (ἰδού), to the plural imperative that directs the gaze towards a grisly, yet aestheticized, target (compare καλοῦ to κακόν, Soph. Trach. 992 and Aj. 1003). Above all, Leontius speaks to a vague addressee, ‘you wretches’ (κακοδαίμονες), a word that plausibly refers back to his eyes, but also capaciously addresses a broad and layered audience (i.e. his, Socrates’ and Plato’s). Like theatrical spectators, those hailed by these words are prompted to assess their own lurid subjectivity and that of their imagined peers as they place themselves in Leontius’ affective state. Although brief, this passage from the Republic has become crucial evidence for understanding ancient aesthetics—­ especially those of tragedy. Identifying Leontius’ words not merely as tragic, but as a sight invitation in particular, lends further cross-­generic texture to the scene.54 Like Orestes, Heracles, and Agave, Leontius invites an indefinite spectatorship to memorialize, confirm, and af­fect­ ive­ly share in his vision at a moment of extreme emotional disturbance. As the Republic is elsewhere critical of the theatre’s effects on society’s moral development, it is wonderfully ironic that Plato himself turns to a verbal trope developed to exploit tragedy’s intersubjective stage in order to illustrate the pluralistic nature of our individual souls.

Acknowledgements This chapter is itself the result of many eyes looking over a common object. I am especially grateful for initial feedback from fellow conference attendees in Leiden and early direction from David J. Jacobson, Anne-­Sophie Noel, and colleagues in Chapel Hill. Anonymous readers for Oxford University Press saved me from many embarrassing errors late in the process. The volume’s editors, Ineke Sluiter and Felix Budelmann, offered insightful guidance every step of the way. All remaining errors and omissions are, of course, my own.

54  On tragic subtext and parody in Plato, see Nightingale 1995, 60–92.

11

Generic Expectations and the Interpretation of Attic Tragedy Some Preliminary Questions and Cognitive Considerations Seth L. Schein

11.1 Introduction In this chapter I consider the usefulness of cognitive work on genre for the literary interpretation of Attic tragedy. In what ways, if at all, might cognitive studies lead to a better understanding of the existence of generic expectations on the part of ancient and modern audiences and readers, and of the fulfilment and disappointment of these expectations? I take it as axiomatic that such disappointment could engage ancient audiences and readers, as it engages modern ones, by challenging them, as it challenges us, to achieve interpretative and ethical clarity in the face of complex and contradictory generic features that render these dramas ‘difficult’. The use of the word ‘genre’ to mean a literary class or kind goes back only to the early nineteenth century,1 but Aristotle’s Poetics and many archaic and fifth-­ century texts speak to the production, performance, and reception, on specific occasions, of distinct kinds of epic, melic, and dramatic works. Throughout the classical period, the presence of audiences at an annual civic event defined as a competition in tragedy, where they saw plays by authors and performances by actors who specialized in that kind of poetry, reflects the continuity of tragedy as a genre. The conventional language, metre, style, dramatic action, and discourses of works staged in the tragic competition, as well as the elite social status of the characters, defined and strengthened the boundaries of this genre and the ex­pect­ ations associated with it. We know of repeated efforts to police these boundaries, from the fining of Phrynichus to the rejection of Euripides’ first Hippolytus to Aristophanes’ repeated negative representations and mockery of Euripides. The fact that Aristotle could define tragedy as he does and categorize its varieties is itself a further sign of its status as a genre. 1 OED s.v. Seth L. Schein, Generic Expectations and the Interpretation of Attic Tragedy: Some Preliminary Questions and Cognitive Considerations In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0011

GENERIC EXPECTATIONS AND INTERPRETING ATTIC TRAGEDY  197 Cognitive approaches to genre are also concerned with categorization, but unlike literary scholars, who treat genres as properties of particular texts, cognitive theorists locate genres in the mind and see them as manifestations of the general propensity of human minds to categorize. Cognitivists, however, do not always consider in whose mind the genres exist—­the author’s, the viewer’s, the reader’s, or all of these simultaneously. Nor do they always sufficiently allow for differences among members of an audience or readers of a text that is said to belong to a particular genre. Instead, cognitivist scholars speak of what, in their view, is more generally human. For example, George Lakoff, in Women, Fire, and Dangerous Things. What Categories Reveal about the Human Mind (1987), says, ‘Without the ability to categorize, we could not function at all, either in the phys­ ic­al world or in our social and intellectual lives. An understanding of how we categorize is central to any understanding of how we think and how we function, and therefore central to an understanding of what makes us human.’2 Lakoff built his theory of ‘Idealized Cognitive Models’, or ICMs, on Eleanor Rosch’s work in the 1970s on ‘prototype effects’ in categorization, especially her demonstration that people generate categories from prototypical examples rather than abstract rules or qualities, and that they understand categories with reference to such examples. ‘Prototypical’ or ‘central’ examples are representative of a given category, while other examples, in Rosch’s terms, are ‘marginal’ or ‘peripheral’.3 Extrapolating from Rosch’s research, Lakoff argued that ICMs ‘could be propositional, image-­schematic, metonymic, and/or metaphoric and could define category structure, central members, and principles of extension’ as the basis of human thought.4 Rosch’s prototype theory, however, seems more helpful for students of literary genre than Lakoff ’s elaboration of it, not least because Rosch leaves room for historical and cultural differences. For example, nowadays scholars think of the Oresteia, especially Libation Bearers, as the prototype of later Orestes-­Electra tragedies, including Sophocles’ Electra and Euripides’ Electra and Orestes. But might a young viewer of Orestes in 408 bce, perhaps seeing dramas in the Theatre of Dionysus for the first time, have been so engaged by Euripides’ play as a reflection of contemporary Athenian history, society, and culture,5 that for him or her it would be or would become the ‘prototypical’ dramatization of the myth, while older viewers might have considered it merely an eccentric extension (or development) of a different prototype with which they were familiar? Might Orestes, in which only the intervention of Apollo averts destruction and self-­destruction, have seemed similarly prototypical in the re-­performance of 340 bce, in a

2  Lakoff 1987, 6. The categorization of literary genre is, of course, only one among many kinds of categorization by which the mind organizes knowledge and experience. See Rotstein 2010, 3–4. 3  See Rosch 1975, 1983; Rosch 1978.    4  Sinding 2002, 181. 5  Cf. Reinhardt 1960; Schein 1975.

198  Seth L. Schein historical situation similar to that of 408, insofar as the survival of Athens was threatened? For a theatre audience (unlike an audience listening to a performance of Homer), would a tragic prototype be expected to age, become ‘out of date’, and change? Furthermore, since Greek audiences would have included individuals familiar with non-­dramatic versions of the Orestes myth, including those of Homer and Stesichoros, and perhaps even with visual representations such as the diptych by the Dokimasia Painter, which might be earlier than the Oresteia,6 would these non-­dramatic representations have affected what seemed theatrically prototypical? What do modern audiences and readers, in our own historical context, consider the prototypical Orestes-­Electra tragedy, and why? Some cognitive linguists and literary scholars understand literary genres as ‘category concepts’ and have followed Lakoff in applying prototype theory to aspects of form and meaning in language, with particular emphasis on metaphorical, lexical, and syntactic constructions; others focus more broadly on literary genres, following literary historians in identifying genres and classifying works according to their places in historical series or sequences. They characterize generic features in terms of Lakoff ’s ICMs, but develop histories of the novel and of other genres in terms of ‘prototypes’ and ‘extensions’ that would allow for the development of schematic models.7 Andrea Rotstein, in the introduction to her book on Greek iambos, argues that cognitive categories refer not to ‘entities in the real world’ but ‘to mental representations of abstract entities: conceptual groupings of poetic compositions’. These verbal constructs ‘generate semantic meaning and are thus open to interpretation’,8 and by ‘studying ancient literary genres in their historical contexts’, we can, Rotstein argues, ‘recover the historical structures with which the ancients organized their poetic experience’.9 She identifies such historical structures with Lakoff ’s ICMs, and in her view, because these structures are socially and culturally constituted on the basis of individuals’ real experience, they can be seen as an improvement over earlier scholars’ ahistorical, essentialist constructions of genre and of particular genres. Nevertheless, when applied to the interpretation of Attic tragedies in performance or as literary texts, the notion of genres as category concepts that are also historical structures does not, I think, help us to understand adequately how genre matters, because it does not address the generic ex­pect­ ations of audiences and readers and their fulfilment or disappointment. In a recent essay on the role of expectations in the workings of the brain, Lauren Y. Atlas and Tor D. Wager survey a wide range of detailed experimental evidence, with an emphasis on the brain’s anticipation and experience of pain and pleasure.10 They conclude that the fulfilment and disappointment of expectations 6  Boston Museum of Fine Arts, 63.1246, calyx krater. See Vermeule 1966. 7  See Taylor 2011; cf. Labov 2004. 8  Rotstein 2010, 4. 9  Rotstein 2010, 4–5. 10  Atlas and Wager 2013.

GENERIC EXPECTATIONS AND INTERPRETING ATTIC TRAGEDY  199 are ‘central to learning, affect, and emotion, and the coordination of adaptive behavior’, and they emphasize in particular the ‘fundamental importance’ of expectations ‘in driving emotion’.11 Atlas and Wager conclude that ‘the emerging study of brain mechanisms that underlie expectancy belongs to many fields’, but unfortunately they say nothing about the representation of emotions in dramatic literature and the responses of audiences and readers in various cultural and historical settings. In the case of Attic tragedy, where scholars since Aristotle have recognized the fundamental importance of emotions both in the texts themselves and in audiences’ and readers’ responses, cognitive study of the fulfilment and disappointment of expectations may prove at least as illuminating as considering genres as ‘category concepts’.12 The disappointment of expectations seems especially significant when con­ sidered in the light of the Russian Formalist principle of ‘defamiliarization’. Whenever an author ‘defamiliarizes’ a linguistic, stylistic, or metrical phenomenon, an object or manner of description, an embodied representation, or a generic feature, and thereby renders it unexpected, it becomes difficult to appreciate. This difficulty forces a listener, viewer, or reader to confront the phenomenon in a different way or on a different level from what is ‘familiar’ and expected, to linger on it and analyse it in order to understand it. Accordingly, a defamiliarized or unexpected object, narrative, dramatic representation, or generic feature can be elevated and transformed from something ordinary into a work of art, or an element within a work of art can be given special significance. Structuralist theory speaks of ‘marked’ and ‘unmarked’ rather than ‘familiar’ and ‘defamiliarized’, but the interpretative significance of the ‘marked’ is like that of the ‘defamiliarized’: both have to do with the degree and kind of attention elicited from a listener, viewer, or reader through the disappointment of conscious or unconscious expectations. I have argued elsewhere that cognitive neurolinguistic research into ‘Event-­ Related Brain Potentials’ (or ERPs), a class of stimuli that generate distinctive, measurable brain activity in the form of brainwaves correlated with syntactic, semantic, and prosodic processes,13 can throw light on how metrical anomalies in the Homeric hexameter function as ‘defamiliarized’ or ‘marked’ phenomena that elicit a special kind of attention from a listener or even a reader.14 The brainwaves, lasting from .185 to .6 milliseconds, suggest that when stimulated by the unexpected, the brain is challenged to make sense of the disappointment of its ex­pect­ ations. In the same essay, I also suggested that, at the level of narrative, mythological allusions in the Iliad to stories told in the Cypria, which contradict versions of myths found elsewhere in the Iliad, might work in the same way by

11  Atlas and Wager 2013, 374. 12  For an approach to Hamlet in terms of expectation and emotion, see Hogan 2012. 13  Friederici 2004. 14  Schein 2016, 93–115.

200  Seth L. Schein disappointing the expectations of audiences and readers and, in effect, asking them to interpret the unexpected. The disappointment of generic expectations in tragedy must work in a similar way. Unfortunately, I have found no studies that measure the activity of the brain when expectations regarding literary genre are disappointed. Nevertheless, the arguments by Frederick Turner and Ernst Pöppel in their well-­known article, ‘The Neural Lyre: Poetic Meter, the Brain, and Time’, are suggestive.15 Turner and Pöppel note, for example, that ‘the human nervous system seems designed to regis­ter differences. It is habituative, in that it tends to ignore repeated and expected stimuli and respond only to the new and unexpected’; in addition, it is ‘predictive’: ‘the patterns it extrapolates or invents are patterns that involve specific expectations of what will happen next and in the more distant future, ex­pect­ ations which await satisfaction and are tested by the senses’.16 Turner and Pöppel’s arguments, if valid, suggest both that the brain would register deviations from, or violations of, generic norms to which it has become habituated, and that the satisfaction or disappointment of such norms can, so to speak, either make the brain’s predictions come true or falsify them through ‘the new and unexpected’.17 Andy Clark’s studies of the brain’s ‘predictive processing’, its activity as, essentially, a prediction machine that attempts to evaluate and understand incoming sensory inputs in the light of expectations or predictions by constantly constructing and adjusting models of what happens next,18 have recently provided a paradigm for literary scholars, especially Karin Kukkonen, of how to understand the construction and consumption of fictional narratives.19 Readers, Kukkonen argues, construct (usually preconscious) ‘designs’ of which predictions are most probable, and these ‘probability designs’ provide the basis for the satisfaction with which they respond to a wide range of novels.20 Kukkonen focuses exclusively on prose fiction and only rarely refers to dramatic literature, but she discusses readers’ receptions of novels in ways that might be relevant to the fulfilment and disappointment of generic expectations in Attic tragedy: cognitively based ‘predictive templates’ enable the many readers of ‘a single text’, as they might have enabled ancient audiences viewing a single play or a tetralogy, to exploit ‘the strongly pre­dict­ive dimension that is usually accorded to genre’.21 The studies of Clark and Kukkonen, like those of Atlas and Wager and Turner and Pöppel, have affinities with work on the cognition and appreciation of music 15  Turner and Pöppel 1983. 16  Turner and Pöppel 1983, 278–9. 17  Turner and Pöppel 1983, 279. 18  Clark 2013, 2015. 19  See Kukkonen 2014a, 2014b, 2020. Within classics, see Christensen 2018 on the ending of the Odyssey. 20  Kukkonen 2020, 1–2. Tobin 2018, without referring to ‘predictive processing’, develops a different cognitive model of how fictional texts are constructed by authors and understood and enjoyed by readers, which is also based on the fulfilment or thwarting of the brain’s expectations regarding both characters and plots. 21  Kukkonen 2020, 110, 132–3.

GENERIC EXPECTATIONS AND INTERPRETING ATTIC TRAGEDY  201 by Leonard Meyer and Daniel Levitin, among others.22 These authors demonstrate how anticipation is essential to a listener’s enjoyment of the tensions and resolutions in a melody or larger musical structure, even though this anticipation is not always simple nor always directly satisfied. They show how non-­verbal music can make sense and arouse emotional responses through formal, auditory means, including unexpected and ‘difficult’ effects that cause a listener to be puzzled, curious, even disturbed, until a resolution is provided.

11.2  Generic expectations in Attic tragedy I turn now to three questions: (1) how are generic expectations fulfilled or disappointed in Attic tragedy, and to what effect? (2) can we fruitfully address in­ter­ pret­ative problems raised by the genre as a whole or by individual plays in terms of the fulfilment or disappointment of such expectations? (3) does a cognitive approach to genre, especially through ‘predictive processing’ and ‘probability designs’, contribute usefully to answering the first two questions? One major problem is that we really do not know what the expectations of ancient audiences were, their ‘predictive templates’, when they went to the theatre or read the text of an Attic tragedy. Nor do we know what constituted fulfilment or disappointment of whatever expectations they may have had. For example, in terms of plot, it is unclear (at least to me) what was perceived as normal and what constituted a deviation from the norm that would have provoked or invited interpretation. Take the nostos-plot as it occurs in Aeschylus’ Persae and Agamemnon, Sophocles’ Trachiniae, and Euripides’ Heracles. In all four plays, the return home turns out to be associated with death and disaster. Would an audience have expected a ‘happy ending’ and been forced to come to terms with catastrophe and the disappointment of their ‘probability design’, or did their familiarity with nostos-plays as a category mean that they expected or predicted a negative turn of the plot and would have been surprised if it did not occur? The latter seems more likely, and the catastrophe probably was the dominant element of the whole plot and unlikely to have been a surprise. Nevertheless, when, in the final lines of Philoctetes, the Chorus speak of ‘having prayed to the sea-­nymphs / to come as saviours of (our) return home’ (νύμφαις ἁλίαισιν ἐπευξάμενοι | νόστου σωτῆρας ἱκέσθαι, 1470–1), would an audience, whatever they made of νόστος (a ‘return home’ to Troy? A post-­war ‘return home’ to Skyros?), have anticipated something destructive? If so, would they have thought not of Philoctetes’ easy trip home mentioned by Nestor to Telemachus at Odyssey 3.190, but of the destruction of the Greek fleet at Cape Caphereus in southeast Euboea, when it was sailing home,

22  Meyer 1956, Levitin 2006. I thank Mark Griffith for these references.

202  Seth L. Schein which is mentioned in the summary of the Cyclic Nostoi and which Hellenistic and later authors treated as the occasion when Philoctetes was blown off course to Italy, where he settled permanently?23 If a nostos plot, in and of itself, did suggest inevitable destruction to a fifth-­ century audience, would, for example, the audience of Oedipus the King have recognized this sort of nostos in the play’s backstory? I specify ‘a fifth-­century audience’, because for later audiences the story of a return home to his native city and family made by a formerly abandoned baby, who is recognized by tokens as indigenous and legitimate and ends by marrying a local girl, was a basic plot of New Comedy. The power of the OT may stem in part from its use of this comic plot in the most negative way.24 As far as we know, however, this kind of plot does not become a ‘comic plot’ until the late fourth century, when it involves characters who are lower on the social scale than the noble characters of Oedipus the King, characters whom Aristotle would have termed φαῦλοι rather than σπουδαῖοι, like the main figures of fifth- and fourth-­century tragedy. It may be useful to compare our access to the relatively clear-­cut and restrictive generic expectations and predictive processing of fifth-­century audiences and readers with our access to those of medieval, renaissance, modern, and postmodern audiences and readers, for whom poet-­playwrights could and did write both comedies and tragedies (and history plays, pastorals, and satires). Playwrights in these later periods could exploit their audiences’ generic expectations not only by straightforwardly fulfilling or disappointing them but by combining elements of multiple genres in a single play, in ways which went beyond those audiences’ probability designs and helped to make that play problematic or difficult. In Athens, audiences had relatively clear generic expectations of whatever play they were seeing: they knew, for example, that in a tetralogy by a tragic poet, the first three plays would be tragedies, and these tragedies would be followed by a satyr play that might complicate the interpretation of the tetralogy as a whole (the Oresteia, for example, ended with Proteus, set in Egypt); that a play by a tragic poet would be a tragedy and a play by a comic poet, a comedy, even though it might contain paratragic passages. In Elizabethan and Jacobean London, however, where the same poets wrote both tragedies and comedies, the audiences were not always as confident as Athenian audiences had been as to just what they should expect on any particular occasion, and for them a play could more easily be both tragic and comic.25

23  Schein 2013, 43. 24  See Lattimore 1958, 82–4. The OT also operates with and plays against two complementary, ‘happy-­ending’ story-­patterns familiar from folk tales: that of the lost one found and that of a young man who comes to a strange land, wins a contest or accomplishes a difficult task, marries the princess of that land, and becomes its king. 25  See Pollard 2015; 2017, 2, 12–13, 55, 180, 221.

GENERIC EXPECTATIONS AND INTERPRETING ATTIC TRAGEDY  203 Take, for example, Shakespeare’s Troilus and Cressida. As Colin Burrow writes, that play is famously difficult ‘to pin down. The title page of the 1609 Quarto calls it a history. The [first] Folio of 1623 calls it a tragedy. The epistle that was prefixed to the second state of the Quarto refers to it as a comedy. As though to confirm its generic amphibiousness it is placed . . . in the first Folio between the histories and the tragedies.’26 One might add that the prologue presents the play as epic, claiming that it ‘Leaps o’er the vaunt and firstlings of those broils [sc. of the Trojan War] / Beginning in the middle, starting thence away / To what might be digested in a play’ (prologue 27–9). Furthermore, the diction and discourse throughout the play of Thersites, and occasionally of other characters, are conventionally sa­tir­ ic­al. This inclusion of elements belonging to five genres in a single work was possible in the Renaissance, because authors, actors, audiences, and readers had in their minds a shared familiarity not only with the differences among genres but with their permeability.27 An even richer Shakespearean example of a play of mixed genres is Cymbeline (1610), which scholars have often characterized (in Polonius’ words) as ‘Tragical-­ Comical-­Historical-­Pastoral’ (Hamlet 2.2.180),28 but which actually includes elem­ents of tragedy, tragicomedy, comedy, satire, English history play, Roman history play, and dramatic romance, as well as a calumny-­of-­women plot known from medieval romances and folk tales, from the Apocryphal story of Susanna and the Elders, and from numerous sixteenth-­ century lawsuits in English courts.29 We can, I think, be confident that different members of Shakespeare’s original audience were familiar in different ways, and to different degrees, with at least some of the complex background to Cymbeline, but it is unclear how they understood or how we should interpret the play’s extraordinary combination of multiple and sometimes contradictory generic elements and markers.

11.3  Disappointment of generic expectations: some case studies The remainder of this chapter explores the disappointment of generic ex­pect­ ations in Sophocles’ Philoctetes and Euripides’ Orestes and Alcestis. Even though, as I have said, it is not easy to know precisely the generic expectations and cognitive templates of ancient audiences and readers of fifth-­century tragedies and how 26  Burrow 2006, xxi; cf. Bevington 1998, 3–6. 27  Such permeability would have complicated the predictive processing associated with the interpretation of a generically simpler play. 28  E.g. Felperin 1972, 197; Hunt 1990, 42; Bate and Rasmussen 2007, 2240. 29  Wayne 2017, 3–30. The story of Susanna and the Elders is apocryphal for Protestant churches, including the Church of England during Shakespeare’s lifetime, but is considered a canonical part (Chapter 13) of the Book of Daniel by Catholic and Eastern Orthodox churches. This chapter does not appear in the Hebrew Bible but is included in the Septuagint.

204  Seth L. Schein these expectations and templates changed over time, critical attention to their disruption or disappointment would have led readers or viewers then, as it leads readers and viewers now, into the central themes and interpretative problems of the plays. It seems safe to say that many, if not most, fifth-­century theatre-­goers or readers would have known, at least in general terms, the traditional myths involving the characters in a given play; some of the ways in which these myths had been presented in earlier dramas; and (perhaps at an unconscious level) the folk tales or folk-­tale patterns that lay behind (or beneath) these myths and dramatic representations. This knowledge would have given them a set of expectations and enabled them to place the play they were seeing or reading in one or more cat­ egor­ies, as would their familiarity with epic and melic treatments of the myth and with contemporary ‘novellas’ of the kind delineated by Sophie Trenkner in The Greek Novella in the Classical Period.30 There would, in other words, have been a basis in probability designs for their responses to the fulfilment or disappointment of their expectations. Philoctetes (409 bce) and Orestes (408 bce) are useful dramas on which to base a discussion of the fulfilment and disappointment of generic expectations, because we know more than is usually the case about their epic and dramatic antecedents. In particular, we have details of Aeschylus’ and especially Euripides’ earlier Philoctetes-plays from Dio Chrysostom’s 52nd and 59th Orations,31 as well as Aeschylus’ Libation Bearers, Euripides’ Electra, and Sophocles’ Electra, which dramatize the same part of the story of the house of Atreus as does Orestes. This material can help us to see how both Philoctetes and Orestes depart strikingly from previous dramatic representations and traditional mythology, challenging both ancient and modern audiences and readers to interpret their innovations. Unfortunately, we cannot know the cognitive templates with which members of Sophocles’ and Euripides’ audiences entered the Theatre of Dionysus in 409 and 408 bce, or precisely how the manifest changes to traditional mythology and departures from previous dramatic representations might have disappointed their expectations and affected their interpretations of Philoctetes and Orestes. For example, Sophocles presents Lemnos as savage and uninhabited, even though it had been inhabited in Aeschylus’ and Euripides’ Philoctetes-­ plays (and all Athenians knew that it had always been inhabited and for nearly one hundred years had been an Athenian ally). Concomitantly, since the island is uninhabited, the Chorus consists of soldier-­sailors from Neoptolemus’ ship rather than indigenous Lemnians, as in the plays of Aeschylus and Euripides. Perhaps most significantly, Sophocles introduces Neoptolemus into the myth, with all his

30  Trenkner 1957. 31 Aeschylus’ Philoctetes probably dates from the first third of the fifth century; Euripides’ was produced in 431 bce, in the same tetralogy as Medea. For other fifth- and fourth-­century Attic tragedies, comedies, and perhaps two satyr ­plays featuring Philoctetes, see Schein 2013, 3–4.

GENERIC EXPECTATIONS AND INTERPRETING ATTIC TRAGEDY  205 un­savoury mythological associations and Achillean baggage, in place of Diomedes, who had helped Odysseus to steal Philoctetes’ bow in Euripides’ play. It is difficult to say exactly what fifth-­century and later audiences would have made of these dramatic innovations and disruptions of their cognitive models, but clearly the representation of the island as harsh and uninhabited would have heightened their sense of Philoctetes’ isolation and suffering. The introduction of Neoptolemus not only provided a provocative example of character development and change of mind, both of which are rare in surviving Greek literature,32 but would have created a contrast between innocence and experience in the realm of politics and allowed Sophocles to generate a critique of traditional conceptions of nobility and heroism, of ends and means, that seems to go beyond anything that might have transpired in the Philoctetes-­plays of Aeschylus and Euripides (though it is, of course, impossible to know for certain what ethical complexity these plays may or may not have contained). Orestes disappoints audience expectations and probability designs in a similarly radical manner. The play takes place after Orestes’ murder of his mother and Aegisthus, rather than climaxing in those killings; with the Oresteia in mind, it represents Orestes as having violated existing homicide law in avenging his father, rather than making the law emerge in the aftermath of the vengeance. Furthermore, despite the Argive setting, Euripides gives Orestes an unmistakable Athenian flavour, especially by the repeated use of the term ‘Eumenides’ in reference to the Erinyes (38, 321, 836, 1650), a use not found elsewhere in Euripides, and by the representation of the Argive assembly (866–954) and its speakers’ values and policies in language that unmistakably suggests the Athenian assembly and the actions and values of contemporary Athenian political leaders.33 What would Euripides’ audience have made of these references to Athenian institutions and settings, and how would Sophocles’ audience have understood the more explicit references in Philoctetes, as when, in the final lines of the prologue, Odysseus invokes the aid of ‘Victory Athena the City Goddess’ (134) in his intrigue against Philoctetes, anachronistically endowing the goddess with cult titles that a fifth-­century Athenian audience would have recognized as referring to their own civic religion; or when, at 1327–8, Neoptolemus describes the snake that bit Philoctetes as ‘indwelling serpent’ (οἰκουρῶν ὄφις), using a quasi-­technical term that designated the sacred serpent dwelling in the Erechtheum (cf. Ar. Lys. 759, Hdt. 8.41.2, Paus. 1.24.7)?34 Would it have seemed that Athens was somehow partly responsible for Philoctetes’ sufferings or being enlisted in Odysseus’ intrigue against him? How could Sophocles’ audience not have felt, by the time 32  See Gibert 1995, 18–20, 143–58. 33  See Di Benedetto 1965, 174–5; De Romilly 1995, 143–5. For analogous, but more complex, allusions to Athenian political leaders in Philoctetes, see Jameson 1956, 219; Bowie 1997, 56–61; Schein 2013, 137–8 on lines 96–9, and 189–90 on lines 385–8. 34  Schein 2013, 11–12.

206  Seth L. Schein they left the theatre, that the unexpected introduction of Neoptolemus in place of Diomedes gave his dramatic version of the story remarkable dramatic and ethical complexity in comparison with earlier Philoctetes-­plays? Similarly, what would Euripides’ audience have made of the unexpected characterization of Orestes as mad, sick, and opportunistically brutal, of Menelaus as cowardly and self-­serving, and of Tyndareus, the spokesman for law and order, as hypocritically ready to take the law into his own hands? Would they have interpreted the friendship of Orestes, Pylades, and Electra, grounded as it is in a series of shared, murderous actions, as subverting φιλία, the main institution and value that in other Euripidean plays stands in opposition to such actions?35 Finally, how would the audiences of both plays have understood the deus ex machina endings, in which divine intervention so shockingly reverses the expected outcome of the plot on the human level? Might they have predicted such an ending in a Euripidean drama, but not in one by Sophocles?36 Would at least some members of Euripides’ audience have been tempted to think, like some modern audiences and readers, that the ending of Orestes, with Apollo assigning all the characters their traditional mythological destinies that have nothing to do with the dramatic action, makes it a kind of theatre of the absurd? On the other hand, how would Sophocles’ audience have understood the rapid changes in the dénouement, in which Philoctetes first decides to reject Neoptolemus’ offer of rescue from the island, healing, and heroic glory, if this means he must go to Troy to help win the war; then accepts Neoptolemus’ offer to ‘save’ him by bringing him home; and finally yields to the persuasion of Heracles, his friend and heroic model and Zeus’ spokesman, to go to Troy to help sack the city and ‘make [his] life glorious after and through these labours’ (1422)? Would the audience’s probability designs have allowed or encouraged them to understand this ‘happy ending’ as truly ‘happy’? Would it have mattered that Odysseus, who is rejected by Philoctetes, defeated by Neoptolemus, and exposed in his final appearance as a coward, nevertheless gets what he aimed at all along, the presence of Philoctetes and his bow at Troy? Would the allusion to Neoptolemus’ future impiety during the sack of the city (1440–4) have in some way tainted the happy ending? These difficult questions all spring from the play’s disappointment of expectations and nonconformity with predictive templates. Ought we to conclude that although it was part of the tetralogy that won first prize in the tragic competition, Philoctetes 35 The hypothesis of the play by Aristophanes of Byzantium describes it as ‘worst . . . in (its) characters; for they all are “bad” (or “contemptible”?), except Pylades’ (τὸ δὲ δρᾶμα . . . χείριστον . . . τοῖσ ἤθεσιν· πλὴν γὰρ Πυλάδου πάντες φαῦλοί εἰσιν), a strange comment considering that Pylades is represented as no less murderous and criminal than Orestes and Electra. This scholarly judgement, however, dating from the late third or early second century, may not reflect the expectations and interpretations of Euripides’ audience in 408 or of the audience at the play’s reperformance in 340. 36  Heracles in Philoctetes is the only deus ex machina in the surviving plays of Sophocles, though there may have been similar divine epiphanies at the end of Athamas (with Heracles as the god) and of Peleus (with Thetis).

GENERIC EXPECTATIONS AND INTERPRETING ATTIC TRAGEDY  207 is, in modern terms, a problem play that would have generated divided responses among ancient audiences and readers as it does among modern ones, challenging them to rethink their own ethical and political values and the moral choices they are called on to make in their own lives? Alcestis differs from Philoctetes and Orestes in its fulfilment and disappointment of generic expectations, because it was pro-­satyric, if we can believe the hypothesis attributed to Aristophanes of Byzantium, and thus belonged to two ­different, if related, dramatic genres. But as Donald Mastronarde asks, what would ‘pro-­satyric’ have meant to Euripides’ fifth-­century audience, if the term was known to them? Was Alcestis unique in their experience, or had they seen other tetralogies in which the fourth play lacked satyrs? Would they have known in advance, from the proagôn or in some other way, that Alcestis was to be such a play, or did they find that out only when the Chorus of Pheraean elders entered the orchestra instead of a chorus of satyrs?37 Would they have predicted the play’s happy ending, owing to their familiarity with the myth of Alcestis and Admetus or with the folk tales of one spouse who dies for the other and of the defeat of Death?38 What would they have made of Apollo’s prophecy to Death in lines 68–9, that Heracles, ‘a guest in this house of Admetus / will take this woman away from you by force’ (ὃς δὴ ξενωθεὶς τοῖσδ’ ἐν Ἀδμήτου δόμοις | βίαι γυναῖκα τήνδε σ’ ἐξαιρήσεται)? Had Heracles, to their knowledge, figured in other versions of the myth? How would his prophecy affect or not affect their understanding of Admetus’ actions, sufferings, and laments for his dead wife in the course of the play, and their moral evaluation of Admetus’ betrayal of Alcestis at the end, when he surrenders to Heracles’ importuning and, contrary to what he so extravagantly promised Alcestis as she lay dying (328–33, 348–56), takes the veiled woman into his house? Finally, for us as scholars and interpreters, is the revelation that Alcestis is still alive really comparable, as is often claimed, to the ending of Shakespeare’s The Winter’s Tale, in which the statue of Hermione turns out to be alive? Ought we to think of Alcestis as a ‘romance’, that is, in Frank Kermode’s words, as ‘exhibiting the action of magical and moral laws in a version of human life so selective as to obscure, for the special purpose of concentrating attention on these laws, the fact that in reality their force is intermittent and only fitfully glimpsed’?39 To be sure, the term ‘romance’ did not exist in the fifth century bce, but would the audience’s familiarity with the plot of the Odyssey have given them the sense of a particular kind of literature of which Alcestis was a dramatic example? The Aristophanic hypothesis of the play calls its conclusion (καταστροφήν) ‘rather comic’ (κωμικωτέραν), but the play, in Greek terms, is clearly not a comedy. How, then, would audiences and readers have received the ‘happy ending’, which must be 37  Mastronarde 1999–2000, 35; 2010, 57. 39  Kermode 1954, liv–­lv.

38  See Lesky 1925.

208  Seth L. Schein mainly what Aristophanes had in mind, since he also says that the ‘dramatic action’ (δρᾶμα) is ‘rather satyric’ (σατυρικώτερον) because it ends (καταστρέφει) in ‘pleasure and joy, contrary to the tragic’?40 Could and would they have understood the essentially arbitrary restoration of Alcestis to her undeserving husband and their symbolic re-­wedding, as ‘formal element[s] corresponding to, and in some valuable way illuminating, diurnal forces that are intermittent and rarely visible’, which is how Frank Kermode describes the elements of Romance?41 Is the ending of Alcestis comparable to the endings of Euripides’ Helen, Ion, and Iphigeneia in Tauris, which, with a nod to Shakespeare, are often called ‘romances’?42 Or is it more fraught morally than the endings of these three plays? How would audiences’ and readers’ impressions of Admetus throughout the play, as he interacts (or fails to interact) with the dying Alcestis, with his father, with Heracles, and with the Chorus, affect their final judgement of him and of the drama as a whole? If we are going to think of Shakespeare, should we consider Alcestis to be less a romance, like The Winter’s Tale or The Tempest, and more a problem play, like All’s Well That Ends Well or Measure for Measure, where the marriages forced on some of the main characters seem so conspicuously arbitrary and precarious? Finally, how does Aristotle’s notion of a tragedy’s ‘proper pleasure’ apply, if it does apply, to the disappointment of expectations in Alcestis (and for that matter, in Philoctetes), and how might attention to the audience’s pre­dict­ive processing throw light on the moral complexity of Attic tragedies that are also problem plays?

11.4 Conclusion Cognitive studies of genre that focus on the mind’s process of categorization constitute one approach to questions that arise in the interpretation of Attic tragedy. Studies that focus on the disappointment of generic expectations offer a second approach, aimed at clarifying how particular plays claim their audiences’ and readers’ attention by subverting such expectations. Cognitive studies involving categorization are basically descriptive and normative. Like other traditional approaches to genre, they strive to situate a work among other works within a

40 The Aristophanic hypothesis of Orestes similarly calls the dramatic action of that play κωμικωτέραν, adding that the same is true of its διασκευή (‘construction’) and that, as already noted (above, n. 35), it is ‘worst in its characters’, all of whom except Pylades are φαῦλοι. Cf. Aristotle, Poetics 1454a28–­9, on Menelaus in Orestes as an example of a character who is ‘unnecessarily base’, which may mean ‘unnecessarily base’ for a tragedy rather than ‘unnecessarily base’ given the demands of its plot. In his comment on the καταστροφή of Alcestis as ‘rather comic’, could Aristophanes also be thinking of its characters, e.g. Apollo arguing with Death and the drunken Heracles? See Griffith 2015, 150n14. 41  Kermode 1954, lv. 42  See e.g. Whitman 1974. Knox 1979a, 265, speaking of Ion as a ‘prototype of comedy’, compares its ‘hair’s-­breadth escape from catastrophe’ with the elements of ‘recognition and escape’ in Iphigeneia in Tauris and Helen, but prefers to call these plays ‘Euripidean Comedy’ rather than romance.

GENERIC EXPECTATIONS AND INTERPRETING ATTIC TRAGEDY  209 larger domain, and they focus on that which is typical rather than that which is distinctive. On the other hand, cognitive studies attentive to the disappointment of generic expectations focus on what is different and distinctive, on how a specific tragedy stands out from other tragedies and engages the minds of audiences and readers in a unique or special way that goes beyond their normal predictive processing. The present chapter has focused on the disappointment of generic expectations, but both approaches are equally valid cognitive operations. Furthermore, because viewers’ or readers’ heightened responses to the disappointment of expectations depend, in part, on the categorization that teaches them what to expect, the two approaches should be understood as complementary pathways into the interpretation both of Attic tragedy as a genre and of individual plays. There is a similar complementarity on a different plane between the two approaches to Attic tragedy discussed in this chapter. The necessarily speculative, traditionally humanistic approach, which focuses on complication of cognitive templates and the fulfilment or disappointment of generic expectations, can be seen as asking questions and establishing lines of research for experimental scientists to pursue, while the self-­avowedly rigorous, scientific approach through categorization can be seen as providing the basis for detailed literary, aesthetic, cultural, and ethical interpretation of specific works, which is the proper domain of humanists. Neither a humanistic nor a scientific approach is ancillary to, or less worthy than, the other, and only ‘mutual aid’ through collaboration between scientists and humanists will find the most productive way forward for cognitive literary studies.

Acknowledgements I would like to thank the editors of this volume, Felix Budelmann and Ineke Sluiter, for advice, encouragement, and patience as I worked on this chapter. I am also grateful to Evert van Emde Boas and Nancy Felson for references to relevant cognitive scholarship, to an anonymous reader for sympathetic, constructive criticism of an early draft, and to Mark Griffith for helpful comments on the penultimate version.

12

Situated Cognition Sophocles, Milgram, and the Disobedient Hero Bob Corthals and Ineke Sluiter

12.1 Introduction World literature forms an invaluable archive of human cognition1—and maybe even more so of metacognition, the ways in which, in different settings, periods, places, and cultures, people have conceived of their own and other people’s cognitive processes and functions. Since literature offers culturally and historically specific representations of human behaviour, it provides rich material for an understanding of ‘situated cognition’. The study of situated cognition belongs in the context of the development of the 4E approach (embodied, embedded, extended, and enactive cognition).2 We will use the term ‘situated cognition’ to refer to all forms of cognitive interaction between agents and the circumstances (in the widest sense of the word) in which they find themselves.3 The situatedness of cognition is best explored through a comparative approach,4 and that is what this chapter aims to do. It offers a comparison of three models of human behaviour, one from a fifth-­century-­b ce literary text, the ­others from research projects from the 1960s/1970s and the first decade of the twenty-­ first century, respectively. The combination of literary and scholarly sources may seem surprising, but in fact the creative design and analysis of psychological experiments will turn out to belong to that same archive of ­ human (meta)cognition. In our three cases, each distinctive metacognitive model was exploited within its own context to explain a basically rather similar 1  Cave 2016, 20; on literary works and situated cognition: Cave 2016, 20–1 (with reflection on the opposition between ‘universalism’ and ‘the specific cognitive positioning of the work as a unique artifact’); Cave 2017. Morgan 2017 sketches the historical development of the situated approach and its many different forms, which, however, all share a commitment to ‘approaching both body and en­vir­ on­ment as integral parts of intelligent human activity’ (220). 2  See the introduction to this volume by Felix Budelmann; on 4E cognition also Menary 2010. 3 Shapiro and Spaulding 2021 (in their section  2.1) capture all four E’s under the heading of ‘embodied cognition’, which is also perfectly defensible. We prefer the term ‘situated’, because it fits our emphasis on ecological validity as an important criterion of cognitive approaches. For a situational approach to problem-­solving (based on the social, cultural, and material aspects of the situation), see Kirsh 2009. 4  Brown 2018. Bob Corthals and Ineke Sluiter, Situated Cognition: Sophocles, Milgram, and the Disobedient Hero In: Minds on Stage: Greek Tragedy and Cognition. Edited by: Felix Budelmann and Ineke Sluiter, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780192888938.003.0012

SITUATED COGNITION: THE DISOBEDIENT HERO  211 situation—­ the prerequisite for a fruitful comparison: a scenario of conflict between an individual’s moral values and the tasks he or she has in some way agreed to perform. The scenario is not just one of internal conflict: in each case, there is interaction with an authority figure. First, in Sophocles’ tragedy Philoctetes (409 bce), we see the young Neoptolemus put under great pressure to follow Odysseus’ orders and deceive Philoctetes so as to make him come to Troy. The second scenario was created in the 1960s by social psychologist Stanley Milgram in his infamous ‘experiments in obedience to authority’: Milgram equally put his subjects under pressure to transgress their moral sensibilities by exposing them to the instructions of an authority figure. His subjects administered what they believed to be real electric shocks to a person they thought was, like themselves, taking part in the experiment as a subject; many proved willing even to proceed to a level of punishment that would have been lethal had the shocks been real. Sophocles and Milgram had particular ideas and concepts in mind that they believed relevant to the behaviour and cognitions of their ‘subjects’ when those subjects found themselves in the stressful situations they had so carefully constructed for them. They also advertised, to their audience and scholarly readership respectively, those conceptual frames, and thereby steered the interpretation of the behaviour that was elicited. Sophocles, of course, was also in a position to have his characters deploy these concepts themselves. These explicitly advertised intellectual frameworks are forms of situated metacognition. For Sophocles, as we will see (section  12.2), the relevant intellectual context was that of the sophists and their interest in the origins and basic principles of human society, manifest in particular in the tension between phusis—­Neoptolemus’ natural disposition as the son of Achilles—­and cultural and conventional factors such as (rhetorical) argument, persuasion, and deceit. For Milgram (section  12.3), the leading hypothesis of ‘obedience to authority’ was generated by his desire to explain how the Holocaust could have happened. Would the presence of an authority figure and a form of delegated responsibility make people (too) obedient? In each case, the creator of the situation was exploring what at the time were seen as basic motivations and incentives of human behaviour, and so they were offering acceptable frameworks for understanding. It is not as if, for Sophocles and Milgram, the concepts used by the other, even across so many centuries, would have been incomprehensible. In each case, however, the respective preferred mode of ex­plan­ation had greater contextual salience (and this is what makes it situated metacognition). Sophocles’ model was related to what we might call the anthropological interests of his time, Milgram’s belonged to social psychology. In the early twenty-­first century, cognitive linguist Herbert Clark reinterpreted the Milgram transcripts in ‘cognitive’ terms, in an attempt to reach an even more basic level of interpretation (section  12.4).5 Clark analysed the conversational 5 ‘Cognitive linguist’ is part of Clark’s self-­ definition; it does not entail that Milgram’s (or Sophocles’) observations are not also ‘cognitive’ avant la lettre.

212  Bob Corthals and Ineke Sluiter behaviour of Milgram’s subjects, particularly their interactions with the authority figure of the ‘experimenter’, and found evidence of the influence of one of the most foundational principles underlying human behaviour: social coordination and collaboration. This enabled him to reframe the results of the Milgram experiments as (normal) ‘joint commitments’ gone wrong. Is this in fact a more basic level of understanding? Indeed, we can demonstrate that Clark’s analysis also works in the case of Neoptolemus (section  12.5). But then, in a way, so does Milgram’s framing in terms of obedience to authority, which may be equally ‘basic’ in human societies. Nonetheless, it is possible to explain what happens in the two scenarios in Sophocles and Milgram, which had been based on very different conceptual models, by means of a cognitive model claiming to transcend cultural-­historical differences in that it attends to basic principles of human interaction. However, further exploration shows that this is not just a story of one-­ directional intellectual progress. How would the three cognitive models account, not only for the phenomenon of obedience, which was most salient to Milgram,6 and which also seems to be at work early in the Sophoclean play, but also for the disobedience that was equally apparent in many of Milgram’s ­subjects and in Neoptolemus (section  12.6)? This was a problem that had stumped Milgram, and that clearly was not on Clark’s agenda either. Recent studies of the Milgram experiments suggest a solution based on ideas that Sophocles would have had little difficulty recognizing: an interplay between personality and circumstances. In fact, then, and in conclusion (section 12.7), the three models are not mutually exclusive at all, and neither does the latest one either explain everything or force us to discard earlier ideas altogether. The historical and cultural archive, exemplified both by literature and creative scientific experiments, can help to demonstrate an important phenomenon of situated cognition: the differential situ­ated salience of competing (but all relevant) elements in human cognition to agents in that archive. This then is the outline of the argument. Now for the details.

12.2  Neoptolemus in Sophocles’ Philoctetes: a hero under pressure Sophocles’ Philoctetes is a fascinating exploration of some fundamental issues of  human survival, social bonding, and the development of human society. Its eponymous hero has been deserted on an uninhabited island by the Greek army 6  Milgram 1974, 5 is explicit on the fact that Milgram focuses on the ‘obedient’, rather than the ‘defiant’ subjects.

SITUATED COGNITION: THE DISOBEDIENT HERO  213 on its way to Troy on account of the unbearable stench of a foul wound to his foot.7 However, many years later the Greeks have discovered that they cannot take Troy without Philoctetes and/or his bow,8 which he had inherited from his friend Heracles. They dispatch Odysseus, accompanied by the young warrior Neoptolemus, who is the son of the recently fallen great hero Achilles.9 Neoptolemus is put under considerable pressure to use deceit in order to make Philoctetes rejoin the Greek army before Troy. Odysseus makes him overcome his aversion to engage in this kind of un-­Achillean behaviour (when Neoptolemus would have much preferred violence, v. 90), and we witness Neoptolemus’ efforts, when confronted with Philoctetes, to stick to the plan. This is no easy task, since Philoctetes, whose struggle for survival in the desolate wilderness without human companions is emphasized time and again, reaches out to Neoptolemus and his companions (the chorus). These in turn are affected by pity (oiktos, e.g. v. 507 (chorus), 965 (Neoptolemus))10 and a desire to help Philoctetes, whose old-­ fashioned heroic ethos cannot but remind Neoptolemus of his father Achilles, and who certainly adopts a paternal attitude towards the young man.11 When Philoctetes feels the onset of a pain attack, he entrusts Neoptolemus with his sole means of survival: the bow. Upon Philoctetes’ recovery, Neoptolemus’ inner conflict causes him, first, to confess the treacherous plan to his new friend, and later, in spite of the demands of Odysseus, to return Philoctetes’ bow to him.12 Moreover, when after a last attempt he cannot persuade Philoctetes to join the Greek army, he agrees to take him home, knowing full well the cost to his own eternal glory, since he cannot conquer Troy without Philoctetes. It is only the appearance of a deus ex machina, Heracles himself, that can save the traditional outcome of the myth: Neoptolemus and Philoctetes are ordered to go to Troy, where they will win great honour. The play has often been understood in the terms it offers rather conspicuously itself, highly topical ones in Sophocles’ time, when they were associated with the intellectual concerns of the so-­called sophists.13 These itinerant intellectuals, who offered education for money, had developed various theories of social 7  The fact that the island is uninhabited is crucial to the thematic of the play; Sophocles added this innovation to the traditional myth (Rose 1976; 1992, 280). 8  There is some ambiguity about this issue, which will not concern us here; see Schein 2013 ad S. Ph. 113 (only the bow mentioned by Odysseus); 191–200 (Philoctetes plus bow necessary according to Neoptolemus); 1329–47 (Neoptolemus talks about bow plus Philoctetes and mentions Helenus’ oracle). 9  The role of Neoptolemus is the second Sophoclean innovation that drives the action of the play and defines its problematic. 10  On pity in the Philoctetes (eleos, oiktos), see Prauscello 2010. 11  See esp. Avery 1965, 285f., also for Philoctetes’ frequent use of the forms of address παῖ and τέκνον; Roisman 1997 for the ‘struggle between Philoctetes and Odysseus over the paternity of Neoptolemus’ (128): will he be a man of action or of deceitful words? 12  We are omitting some of the back and forth here, which does not affect our argument. 13  The best analysis of the links of S. Ph. with contemporary intellectual culture, in particular with the intellectual milieu of the sophists, is still Rose 1976 (revised in Rose 1992, 266–330).

214  Bob Corthals and Ineke Sluiter anthropology on the origins and development of human society and civilization. They were also interested in the politics and ideology of language use, rhetoric, ethics, and epistemology. Well-­ known topics are the possibility of finding counter-­arguments to any argument and of speaking well on any proposition, irrespective of its morality. Some of these concerns are prominent in the play, and discussed by the characters. This is true in particular for the tension between natural, inherited character and persuasive speech (φύσις and λόγος); between violence and persuasion (βία and πειθώ); or between trickery and truth (δόλος and ἀλήθεια). In add­ition, we also observe how the older, hierarchically superior, and authoritative Odysseus leans on the younger man to follow his lead, and how Neoptolemus gradually gives in. The opening scene between the son of Achilles and Odysseus bristles with signal words flagging these concerns. The sheer density of their use leaves no doubt as to what intellectual concerns are evoked here. Here is a significant interchange: NE Son of Laertius, things which it distresses me to hear spoken of / are things which I hate to do! / It is my nature (ephun) to do nothing by treacherous plotting (ek . . . tekhnês); / that is my nature, and it was also my father’s (houkphusas eme) nature. / But I am ready to take the man by force (pros bian) / and not by cunning (doloisin) . . . / (96) OD Son of a noble father (esthlou patros pai), I too when I was young / had a tongue (glôssan) that was inactive but an arm (kheira) that was active; / but when I come to put it to the proof I see / that it is the tongue (tên glôssan), not actions (targa), that rules in all things for mortals. / NE Then what are you telling me to say except lies (pseudê)? / OD I am telling you to take Philoctetes by a trick (dolôi). / NE But why must I take him by a trick (en dolôi) rather than by persuasion (peisant’)? /OD He will never obey (pithêtai), and you could not take him by force (pros bian) / . . . (108) NE Do you not think it disgraceful (aiskhron) / to tell lies (pseudê)? / OD Not if the lie brings us salvation! / NE With what kind of a face will one be able to utter such words? / OD When you are doing something to gain advantage (kerdos), it is wrong to hesitate. (tr. Lloyd-­Jones, adapted) NE ἐγὼ μὲν οὓς ἂν τῶν λόγων ἀλγῶ κλύων, / Λαερτίου παῖ, τούσδε καὶ πράσσειν στυγῶ· / ἔφυν γὰρ οὐδὲν ἐκ τέχνης πράσσειν κακῆς, / οὔτ’ αὐτὸς οὔθ’, ὥς φασιν, οὑκφύσας ἐμέ. / ἀλλ’ εἰμ’ ἑτοῖμος πρὸς βίαν τὸν ἀνδρ’ ἄγειν / καὶ μὴ δόλοισιν· . . . / (96) OD ἐσθλοῦ πατρὸς παῖ, καὐτὸς ὢν νέος ποτὲ / γλῶσσαν μὲν ἀργὸν, χεῖρα δ’ εἶχον ἐργάτιν· / νῦν δ’ εἰς ἔλεγχον ἐξιὼν ὁρῶ βροτοῖς / τὴν γλῶσσαν, οὐχὶ τἄργα, πάνθ’ ἡγουμένην. / NE τί οὖν μ’ ἄνωγας ἄλλο πλὴν ψευδῆ λέγειν; / OD λέγω σ’ ἐγὼ δόλῳ Φιλοκτήτην λαβεῖν. / ΝΕ τί δ’ ἐν δόλῳ / δεῖ μᾶλλον ἢ πείσαντ’ ἄγειν; / OD οὐ μὴ πίθηται· πρὸς βίαν δ’ οὐκ ἂν λάβοις. / . . . (108) ΝΕ οὐκ αἰσχρὸν ἡγῇ / δῆτα τὸ ψευδῆ λέγειν; / OD οὐκ, εἰ τὸ σωθῆναί γε τὸ ψεῦδος φέρει. / ΝΕ πῶς οὖν βλέπων τις ταῦτα τολμήσει λακεῖν; / OD ὅταν τι δρᾷς εἰς κέρδος, οὐκ ὀκνεῖν πρέπει. (86–111)

SITUATED COGNITION: THE DISOBEDIENT HERO  215 The orders of Odysseus to resort to clever speech, artifice, deceit (tekhnê, dolos, peithô) do not fit Neoptolemus’ self-­image, which rests entirely on his natural disposition, his φύσις, as the son of Achilles, marked by phu-words (ἔφυν, οὑκφύσας). Odysseus acknowledges the importance of Neoptolemus’ descent (ἐσθλοῦ πατρὸς παῖ), and excuses Neoptolemus’ preference for violence (βία) as a sentiment of younger people shared at one time by himself. He then first replaces Neoptolemus’ blunt framing of the plan as ‘lies’ with the idea of δόλος (tricking), excludes the alternative options of persuasion and violence, and removes the shame from resorting to lying (a term on which Neoptolemus insists) by associating it with salvation and advantage. In the end Neoptolemus accepts to play the part assigned to him.14 Odysseus removes himself from the action, since he must not be recognized by his arch-­enemy Philoctetes. In the rest of the play, this framework for understanding human behaviour in terms of the tension between natural disposition and cultural forces resurfaces.15 Philoctetes is overjoyed to see the son of ‘his dearest’ Achilles (242). When he wakes up from the horrible attack of his illness and finds Neoptolemus and his bow still there, he praises his steadfastness and naturally attributes it to the younger man’s phusis (874–5 ‘for your natural disposition is noble, my child, and you are born from noble parents’ (εὐγενὴς γὰρ ἡ φύσις κἀξ εὐγενῶν,/ὦ τέκνον, ἡ σή). Neoptolemus is increasingly uneasy with the deceit, and explains his discomfort as the logical result of betraying or ‘leaving behind’ one’s disposition (τὴν αὑτοῦ φύσιν . . . λιπών, 902–­3). ‘No’, says Philoctetes, ‘you’re not doing anything out of line with the one who fathered you’ (ἔξω τοῦ φυτεύσαντος, 904).16 But when he realizes that Neoptolemus is indeed going to betray him he calls out against ‘the son born of Achilles’ (940). ‘Give [my bow] back’, he says, ‘be your own self ’ (ἐν σαυτῷ γενοῦ, 950), i.e. behave according to your phusis. When Neoptolemus ul­tim­ate­ly not only returns the bow, but also gives up his attempt to persuade Philoctetes, the latter says: ‘Child, you have shown the phusis from which you’ve sprung: not from Sisyphus as father, but from Achilles’ (τὴν φύσιν δ’ ἔδειξας, ὦ τέκνον,/ἐξ ἧς ἔβλαστες, οὐχὶ Σισύφου πατρός,/ἀλλ’ ἐξ Ἀχιλλέως, 1310–11).17 The metacognitions activated and advertised in the opening scene remain in place elsewhere in the play. In describing behaviour that flies in the face of one’s phusis, characters speak of resorting to the (wrongful, deceitful) use of words, tricks and contrivances, and deceit: Neoptolemus expresses his fear of being 14  On the metatheatrical aspects of S. Ph., see Lada-­Richards 2009. She observes that at a decisive moment in the interaction with Philoctetes, Neoptolemus will allow his phusis ‘[to hijack] the “character” he is supposed to project’ (56). 15  See Blundell 1988 for the sustained importance of Neoptolemus’ phusis in the play. 16  ἔξω is a logical follow-­up from λιπών. 17  ‘Sisyphus as father’: Philoctetes is here referencing the rumour that Odysseus was fathered by arch-­villain Sisyphus, rather than Laertius/Laertes (cf. Schein 2013 ad 417; 1311). Cf. also Ph. 1284, when Philoctetes is still in despair and calls Neoptolemus ‘most hateful son of the best father’, ἀρίστου πατρὸς ἔχθιστος γεγώς.

216  Bob Corthals and Ineke Sluiter found inferior, κακός, through the use of disgraceful speech (λέγων αἴσχιστ’ ἐπῶν, 909). Philoctetes calls him out for the same behaviour as ‘you most hateful contrivance of sly villainy’, using a τέχνη-term (ὦ . . . πανουργίας/δεινῆς τέχνημ’ ἔχθιστον, 927–8). And he adds: ‘How you deceived me!’ (οἷ’ ἠπάτηκας, 929). The natural tools of an Odysseus in his interactions with his environment (wily deceit, sly words) are unbearable for someone like Philoctetes when exposed to them, and cause moral quandaries for someone like Neoptolemus when using them against another human being, since they are at odds with his natural disposition as the son of his father. We will note again that Neoptolemus would have had no trouble (at least that’s what he himself thought at the beginning of the play) in using violence against Philoctetes. This alternative is not put to the test. In sum, then, in this play Sophocles explores a particular way to understand (the morality of) human behaviour under stressful circumstances, by invoking a powerful contemporary interpretative framework: his situated metacognition is passed on to his characters and would have resonated with the audience.

12.3  Milgram’s experiments in obedience to authority In the first part of the Philoctetes, Neoptolemus is pressured by an authority figure to treat another person in a way that conflicts with his own values. This calls to mind the set-­up of some of the most infamous experiments in social psychology: the ones involving obedience to authority by social psychologist Stanley Milgram (1963; 1974). The willingness of subjects to inflict even lethal electrical shocks to people they were supposedly ‘teaching’, has usually been explained in just these terms: obedience to authority and delegated responsibility. Those concepts were of considerable cultural-­historical interest at the time, in the aftermath of the Eichmann trial. In July 1961, Stanley Milgram began a series of experiments to investigate the willingness of ordinary people to follow the orders of an authority figure in performing acts that are increasingly harmful to others.18 In Milgram’s set-­up subjects were led to believe that they were taking part in a learning experiment, using a memory task. When they entered the lab, someone playing an experimenter would explain that this experiment required a ‘learner’ and a ‘teacher’. Through a rigged lottery, the subject would always be assigned the role of ‘teacher’, while the ‘learner’ was another associate of Milgram’s. Within the triangle of experimenter, learner, and teacher, then, there was only one subject each time: the teacher. In most versions of the experiment, teacher and experimenter would be together in one room, while the learner was in an adjacent one. Subjects were ordered by the

18  Milgram 1963 (the experiments were conducted in 1961); 1974.

SITUATED COGNITION: THE DISOBEDIENT HERO  217 experimenter to administer increasingly severe shocks to the learner in the course of supposedly helping him to commit to memory a list of word pairs. Whenever the learner provided a false answer, the teacher was supposed to give feedback in the form of a punitive shock. To this end a fake ‘shock generator’ was used, fitted out with switches that allegedly produced voltage levels ranging from ‘Slight Shock’ (15 volts) all the way up to ‘Danger: Severe Shock’, and ultimately to two final ones simply marked ‘XXX’ (450 volts). In cognitive terms, this machine was used as ‘environmental scaffolding’ for the teacher, an item in the environment supporting the understanding of the situation by the subject.19 At every incorrect answer, the voltage level was to be raised by a single notch. Milgram conducted many versions of this experiment, varying some conditions. But in most versions (from experiment 2 onwards), the teacher would receive feedback from the victim, consisting of either emphatic vocal responses of protest or distress (e.g. a claim of heart trouble, a request to stop) or other signs to the same effect (e.g. pounding on the wall). Worried reactions by the ‘teacher’ would be blocked by the ‘impassive’, ‘stern’ experimenter:20 ‘the experiment requires that you continue’.21 After the 315-­volt shock, the victim remained silent—­the experimenter would instruct the subject to regard silence as an incorrect answer and to continue. In experiment 2, 25/40 subjects continued until the end; in experiment 1, 26/40 subjects had done so without vocal feedback—­a surprisingly small difference. The subjects were observed to suffer extreme stress and showed various nervous reactions, sighing, sweating, nervous laughter, etc. The experiment lasted until either the experimenter put a stop to the session after the thirtieth and most severe shock level had been reached or the subject at some point broke off the experiment unilaterally by refusing to continue the administration of shocks. In the two years following the original experiment, repeated editions that tested the possible influence of such variables as setting, racial or cultural background, and gender, on the whole yielded somewhat similar results.22 Physical proximity of subject and learner substantially reduced the willingness of the subject to continue.23

19  For environmental scaffolding, see Andy Clark 1997. 20  Milgram 1963, 373. 21  1963, 374; other stock phrases used as prods were: ‘please continue/please go on’, ‘it is absolutely essential that you continue’, ‘you have no other choice: you must go on’, all of this in a ‘firm, but not impolite’ tone. 22  Increasing attention has been paid to the 18 different conditions Milgram created in various iterations of the experiment. In many cases, the transcripts were analysed anew with attention to different factors. See e.g. Kaposi 2020 for a rhetorical analysis of different proximity conditions, or Gibson 2019. And cf. sections 12.4 and 12.6 below. Replication has itself been an ethical dilemma; see Burger 2009 and Miller 2009. 23  Milgram 1974, 32–5. In such cases the bond between learner and ‘teacher’ would compete with the commitment of the ‘teacher’ to the experimenter (or the experiment as such). Cf. the influence of witnessing Philoctetes’ suffering on Neoptolemus. Conversely, if the experimenter was physically removed from the ‘teacher’, obedience dropped sharply (to 9/40, Milgram 1974, 59).

218  Bob Corthals and Ineke Sluiter Milgram notes a number of points relevant to the ‘situated cognition’ (not his term) of his subject, for instance, that they trust the usefulness of their participation, or the importance of the ‘background authority’ of Yale, where the first experiments took place. He also notes that the subjects realize that they have entered the experiment voluntarily. The subject therefore ‘perceives himself under obligation to aid the experimenter. He has made a commitment, and to disrupt the experiment is a repudiation of this initial promise of aid’.24 Another issue is that the subject finds himself torn between the competing demands of two persons, of which he can help only one.25 These points will be important for the re­analy­sis of the experiments by Herbert Clark (section 12.4). The publication of Milgram’s study in 1963 caused a tremendous stir. The fact that some two-­thirds of the participants in the initial experiment obeyed the experimenter’s command all the way to the administration of the most severe—­in fact lethal—­shock seemed to indicate that under authoritative pressure ordinary people could let down their moral guard and commit atrocities.26 This was highly suggestive at a time when people were trying to come to grips with the events of World War II. Notably, it seemed to accord perfectly with Hannah Arendt’s idea of the ‘banality of evil’, her famous 1963 summation of the Eichmann trial in 1961.27 The desire to explain how the Holocaust could ever have happened had obviously and explicitly informed Milgram’s clearly advertised metacognition-­ cum-­hypothesis of ‘obedience to authority’ in the first place. Here too, then, we have an example of situated metacognition, or, to adapt the title of Hutchins’s 1995 book: an example of ‘metacognition in the wild’.28 Milgram hypothesized that the willingness of the obedient subjects could be explained by the presence of an authority, and thus to a form of delegated responsibility.29 ‘[T]he individual who is commanded by a legitimate authority ordinarily obeys’ (1963, 372). The

24  Milgram 1963, 376; cf. Milgram 1974, 64, 149. 25  Milgram 1963, 376; 1974, 24. 26  Milgram has come under a lot of criticism for the way in which he reached this high a percentage of ‘obedient’ subjects; for one example, see Perry 2013, who points out that over half the subjects initially were not inclined to go along with the experiment. The pressures they were put under went further than just ‘authority’, and seemed more like bullying. 27  Arendt 1963; her report of the trial had first been published in five parts in The New Yorker. Modern editions of the book are based on the expanded and revised 1964 version. Milgram (1974, 6) refers to her work. 28  Morgan 2017 discusses Hutchins 1995 (Cognition in the Wild) among several earlier positions on situated cognition. This chapter argues that metacognitions are just as dependent as primary cognition on context and culture. Kirsh 2009 discusses situated problem-­solving, arguing that ‘problems’ are not a natural category, but that they are themselves a product of construction and definition. Metacognitions are themselves an example of situated problem-­solving. 29  Milgram 1963 appeared in the Journal of Abnormal and Social Psychology [our emphasis] and called attention to ‘destructive’ obedience. He says in so many words that ‘[o]bedience, as a de­ter­min­ ant of behavior, is of particular relevance to our time’ (1963, 371) referring to World War II. He regards obedience as an ‘impulse overriding training in ethics, sympathy, and moral conduct’. On deferred responsibility and the link with World War II, see also Milgram 1974, xii, 145. Ultimately, though, it was Milgram’s ambition to explain ‘obedience to authority’ as normal human behaviour (1974, 188).

SITUATED COGNITION: THE DISOBEDIENT HERO  219 context of explaining the Holocaust was a shared interpretative framework for Milgram and his contemporaries. This is evidenced by the remark made by the wife of one of the ‘teachers’ on hearing about the experiment: ‘You can call yourself Eichmann’.30

12.4  Herbert Clark and the natural tendency to cooperate Milgram had set up his experiments and interpreted their results in specific cultural and historical terms. In the early years of this century, Herbert Clark made an effort to explain Milgram’s data in terms designed to transcend such specificity and bring them under the umbrella of a universal and basic phenomenon of human cognition.31 That phenomenon is the primacy of social coordination and human collaboration, based on a view of human nature that attempts to redress excessive emphasis on competition. Human beings are, or are equipped with, ‘interaction engines’.32 They are highly motivated to be cooperative and to engage in what are called ‘joint commitments’. Through a minute linguistic analysis of the communicative exchanges between experimenter and subjects, Clark shows that, in these interactions too, a relationship of cooperation and joint commitment was constructed in the initial phases of the experiments. Note that this approach means that Clark elides the hierarchical relationship between experimenter and ‘teacher’ and thus the whole issue of authority. His interest is in basic human communication in line with some of the universalizing tendencies of the cognitive turn. In that sense, his cognitive model is also an example of situated metacognition. Interaction and communication are key to Clark. Since the basis of communication is, precisely, exchange, Clark’s unit of analysis is not the individual speech act, but rather what he calls a ‘projective pair’, an adjacency pair consisting of a move (called ‘proposal’) by one speaker, and the counter-­move (called ‘uptake’) of the interlocutor.33 A joint commitment is established through repeated positive interactions on joint courses of action. Negotiating a joint activity, then, is usually a piecemeal exercise requiring multiple joint commitments to particulars that emerge incrementally and hierarchically (the latter in the sense that each separate joint action serves the higher goal of carrying out the main joint project). Clark uses the joint project of putting together a TV stand as an example (the more commonly available cognitive model would probably be an IKEA Billy

30  Milgram 1974, 54. 31  Clark 2006, updated in Clark and Henetz 2014. 32  Levinson 2006, referring to an ensemble of cognitive abilities and motivational predispositions which underlie human communication and action (44). 33  Clark 2006, esp. 131–3.

220  Bob Corthals and Ineke Sluiter bookcase these days):34 two people agree to construe the thing together, there is a division of labour, and then there are all the individual tasks and subtasks which are coordinated in projective pairs: ‘now we have to do x’—‘ok’,//‘if you hold this, I’ll find x’—[takes hold of shelf],//‘this seems to be right’— ‘[assenting grunt]’, etc. It would be socially unacceptable if one partner, by refusing a logical next step, in practice would ruin the whole stack of joint actions and the joint commitment to the project as such. Clark’s analysis does not directly target the usefulness of the customary interpretative framework (which may still be relevant), but rather it lays bare this underlying structure of engagement in a joint action, which involves an ongoing process of establishing, negotiating, and maintaining joint commitments. In the Milgram experiments there are a great many positive interactions and commitments before the subjects experience moral unease. A subject voluntarily enters the experiment together with others at Yale laboratory, he assents to a div­ ision of labour through a lottery to divide the roles of teacher, learner, and experimenter for a memory task with word pairs, the procedures for that task are established, and then there is a long series of executions of the memory task proper. At each iteration this involves exchanging word pairs and providing feedback through (initially) light shocks.35 In the mind of the subject, refusal to take the next step would not just be the simple rejection of one item, but would undermine the whole joint project, and thereby negate all the individual small steps of commitment-­building that went before. The subject might feel that he would be ruining the whole experiment. By Clark’s calculation, by the time the experiment had reached word pair 14, the subject had agreed and acted on no less than 99 smaller joint actions, supporting the overall joint project of taking part in a Yale experiment.36 Subjects would be wary of unilaterally breaking off the joint commitment. Instead, most of them would either attempt to negotiate a joint exit with the experimenter (blocked effectively by the programmed refusals of the latter), or try to reframe the implications of the task at hand (e.g. deference of responsibility, denial of the impact of the shocks)—or they would simply keep cooperating, thus keeping intact the long string of already completed joint actions. Joint commitment and joint actions can primarily derail in two different ways: exploitation, i.e. (over-)using the fact that a partner has ceded a certain amount of control in a joint activity; and overcommitment, caused by the general difficulty or even the sheer impossibility of renegotiating prior commitments. While these two factors may increasingly make participants want to back out of the joint commitment, the gradualness of their emergence also means that participants keep 34  Clark 2006 and Clark and Henetz 2014 borrow the example of the TV stand from an unpublished paper by Julie Heiser and Barbara Tversky. We only paraphrase the gist of their analysis. 35  Clark 2006, 145. 36 Clark 2006, 145: the number is not easy to verify, but there are certainly over 70 joint commitments.

SITUATED COGNITION: THE DISOBEDIENT HERO  221 feeling bound to those joint commitments, well beyond moral or other objections. What makes the experiments so wrenching, then, is not so much the conflict between obedience to authority and moral considerations, but between the latter and our basic human tendency to form commitments.37 We should note that Clark’s analysis is reductionist by choice, in itself a perfectly legitimate scientific approach. However, in comparing agreeing to continue building a TV stand to administering shocks to a fellow human being with a focus on social co­ord­in­ ation and collaboration, one is bound to lose something along the way.

12.5  Cooperation in the Philoctetes Nevertheless, if Clark is right about having tapped into a more basic level of human interaction by focusing on issues of joint commitment and joint action, it should be possible to analyse the exchanges in Philoctetes in the terms Clark has offered.38 And indeed, this is highly feasible. We have shown earlier how the prologue establishes and advertises the situated metacognition of phusis and tekhnê. However, it can also be analysed as the mutual and shared establishment of a plan and a role division, in which these first steps (taken onstage, but also reported from the time when the expedition set out from Troy) commit Neoptolemus to what will be a series of joint and coordinated actions. Neoptolemus has agreed to come with Odysseus and assist him in bringing Philoctetes to Troy—­the overarching joint commitment (cf. ‘we are going to put together a Billy bookcase’). At the opening of the play, Odysseus asks him to identify Philoctetes’ cave, a task easily accomplished by Neoptolemus, constituting another step in the joint action, 26–7. Odysseus had promised (24–5) to then set out the rest of the plan, so that Neoptolemus might listen and Odysseus might do the explaining (role division), and the plan would proceed as a joint project (koina) from the two of them (dual form).39 When the first task has been fulfilled, their communication continues with projective pairs, consisting of more 37  In fairness to Milgram, it should be observed that he himself had anticipated a number of Clark’s points, see above section  12.3, p. 219. Earlier, Orne 1962 had signalled the fact that psychological experiments constitute a special form of social interaction known as ‘taking part in an experiment’, which means there are well-­defined roles and the shared assumption that ‘a legitimate purpose will be served by the subject’s behavior’ (777). In the Milgram experiments, the subject (not unreasonably) assumes that the (Yale) experimenter knows what both of them are doing. Cf. e.g. ‘Surely you’ve considered the ethics of this thing’ (quoted in Clark 2006). This is testimony to the initial acceptance of authority. 38  As we will see, this is even the case if Clark’s model does not represent the ultimate basic layer of human interaction. In classical studies, too, more attention is being paid to cooperative values, rather than just to the traditional competitive ones. See Elmer’s study on consensus-­building in Homer (Elmer 2013), and Lambert 2011. Moreover, Clark’s work is also related to so-­called ‘conversation analysis’ (CA; for a brief overview, cf. Chevalier 2011), and this too is applied to classical works, see e.g. Van Emde Boas 2017. 39  Ph. 24–5: ὡς τἀπίλοιπα τῶν λόγων σὺ μὲν κλύῃς,/ἐγὼ δὲ φράζω, κοινὰ δ’ ἐξ ἀμφοῖν ἴῃ.

222  Bob Corthals and Ineke Sluiter directions and questions from Odysseus and Neoptolemus’ positive uptakes. For example, ‘look to see whether . . .’ (30)/‘I see . . .’ (31).40 After this first joint ex­plor­ ation of Philoctetes’ abode, with the younger man doing the actual inspecting, Neoptolemus initiates the next step by inviting Odysseus to proceed with further instructions (‘if there is anything you want, speak again and tell me’, 49). It is interesting to note that Odysseus delivers his instructions as a necessity, repeatedly using the impersonal δεῖ, ‘it is necessary that you’.41 After the first δεῖ sentence, exhorting Neoptolemus to fulfil his task, there is a positive uptake in Neoptolemus’ ‘what are your orders then?’ (τί δῆτ’ ἄνωγας;, 54). Odysseus does not hesitate to exert (social) pressure, even when Neoptolemus has not yet indicated that he may not be willing to cooperate: ‘if you fail to do this, you will hurt all the Greeks; for if his bow is not taken, you will not be able to conquer the Trojan plain’.42 Neoptolemus tries to negotiate (86ff., discussed in section 12.2), referring to his natural, dispositional preference for violence over deceit, but feels the weight of the earlier joint commitment: ‘however, since I was sent here as your collaborator, I shrink from being called a traitor’ (πεμφθείς γε μέντοι σοὶ ξυνεργάτης ὀκνῶ/προδότης καλεῖσθαι, 93–4). At this point there may already be some overcommitment. Odysseus has evoked the weight of social pressure. Neoptolemus realizes that the army is aware of his commitment and would disapprove of any failure of his to honour it. As we have seen in section 12.2, a negotiation over the framing of the task follows, with Neoptolemus expressing his aversion to lies, and bringing up his concerns about engaging in despicable forms of action. In the end, though, Neoptolemus once again repeatedly formulates consent. Initially his phrasing is highly reluctant and impersonal: ‘If that’s the situation, it may be necessary for it [the bow] to be captured’ (θηρατέ’ οὖν γίγνοιτ’ ἄν, εἴπερ ὧδ’ ἔχει, 116);43 but soon he sounds more confident: ‘so be it, I’ll do it!’(120), ‘[I’ll remember what to do], now that I’ve once and for all consented’ (122).44 Right from the prologue, then, we can see some of the overcommitment that Clark’s model identifies as a risk inherent in our natural tendency to form joint commitments. The other risk is exploitation. Both of these come to the fore when Neoptolemus forms a bond with Philoctetes that competes with Neoptolemus’ 40  Other pairs, after Neoptolemus’ successful completion of his first tasks, e.g. in v. 28–9, 32–3, 34–6. Each question/order and reaction cements the joint commitment. 41  Ph. 50, 54, and cf. the impersonal κλεπτέον in v. 57. See Schein 2013 ad loc. This impersonal use is comparable to the formulaic responses by Milgram’s experimenter. Odysseus uses impersonal constructions also with Philoctetes, e.g. 993–4 OD ἡ δ’ ὁδὸς πορευτέα./PH οὔ φημ’ OD ἐγὼ δέ φημι. πειστέον τάδε, ‘OD the journey must be made. PH I say no! OD but I say yes. This must be obeyed’. 42  Ph. 66–9: . . . εἰ δ’ ἐργάσῃ/μὴ ταῦτα, λύπην πᾶσιν Ἀργείοις βαλεῖς./εἰ γὰρ τὰ τοῦδε τόξα μὴ ληφθήσεται,/οὐκ ἔστι πέρσαι σοι τὸ Δαρδάνου πέδον. 43  Note that everything is impersonal and indefinite here; see Schein 2013 ad loc. 44  Ph. 120 ἴτω· ποήσω. Schein 2013 ad loc. observes that ἴτω ‘frequently indicates a decisive commitment to a particular course of action, whatever the consequences’. Ph. 122 ἐπείπερ εἰσάπαξ συνῄνεσα.

SITUATED COGNITION: THE DISOBEDIENT HERO  223 commitment to Odysseus and their common plan.45 Philoctetes acts like a parent towards Neoptolemus,46 but his behaviour is also manipulative and exploitative. For instance, when he first asks Neoptolemus to take him along on his ship, he says (477) σοὶ δ’, ἐκλιπόντι τοῦτ’, ὄνειδος οὐ καλόν,/δράσαντι δ’ . . . ‘but for you, when you abandon this, this will be a disgraceful cause for blame, but if you do it (there will be honour)’. The verb ἐκλείπειν ‘to abandon, leave out’ is rather stronger than simply ‘not doing’: it suggests that the object of the verb is something to which Neoptolemus was already somehow committed. Similarly, the way in which Philoctetes manages to get physical help from Neoptolemus personally is manipulative. Neoptolemus had suggested that his men could lift Philoctetes up, to which he replies: ‘thank you, my boy, raise me up as you intend’ (αἰνῶ ταδ’, ὦ παῖ, καί μ’ ἔπαιρ’, ὥσπερ νοεῖς, 889).47 An example of both exploitation and overcommitment comes up with the question of whether or not Neoptolemus actually ever swore to take Philoctetes home.48 The upshot of this is that Clark’s analysis in terms of an overcommitment to and exploitation of established joint commitments turns out to be enlightening also for the behaviour of Neoptolemus in Philoctetes. However, this does not imply that Milgram’s conceptual apparatus should be discarded. His framing of the interactions in terms of obedience to authority can also be shown to have rele­vance for Sophocles’ play and would not have been incomprehensible to Sophocles and his audience. Odysseus is cast as the authority figure from the beginning, as is clear from the fact that the younger man addresses him as ἄναξ ‘lord’ (Ph. 26, also 94), asks for instruction (49), and is sent as (subordinate) helper (ὑπηρέτης, 53); his role is labelled ὑπηρετεῖν (15), and ὑπουργεῖν ‘render service, assist’ (53).49 When towards the end of the play he decides to redress the wrong he has done to Philoctetes, he connects that hamartia to his misplaced obedience to Odysseus and the whole Greek army (1226).50 Even more explicit is what happens when the whole Trojan myth cycle teeters at the brink of destruction, since Neoptolemus 45  On the bond between Neoptolemus and Philoctetes, competing with that between Neoptolemus and Odysseus, see Kyriakou 2012, 152. Cf. the bond that the ‘teacher’ would feel with the innocent learner in the Milgram experiments. 46  For Philoctetes as a father figure and Achilles-­like character, see Schein 2013, 15 with n. 50, with the literature cited there. See also n. 11 above. Schein 2013, 15–16 nuances this picture by pointing out similarities and differences between Philoctetes and Odysseus, and Philoctetes and Achilles (in epic and tragedy). 47  Clearly, this had not been ‘the intention’, and in the next verse (Ph. 890) Philoctetes dismisses the help of Neoptolemus’ men altogether, after which Neoptolemus indeed picks him up himself. 48  See Ussher 1990 on Ph. 941, 1367, and 1398: in all of these passages it is Philoctetes who claims that Neoptolemus has sworn. When the issue had come up between them, Neoptolemus was still on his Odyssean mission and so is constantly and purposely very vague on where he will take Philoctetes (see e.g. Ph. 528–9). But Philoctetes acts as if he had made a good-­faith offer and in the end, Neoptolemus just accepts this overcommitment (‘if you wish, let us go’, 1402). 49 Both spoken by Odysseus; Neoptolemus uses the slightly more egalitarian ξυνεργάτης, 93. Blundell (1988, 138) points out this authority status of Odysseus. 50  Ph. 1226 ἣν [sc. ἁμαρτίαν] σοὶ πιθόμενος τῷ τε σύμπαντι στρατῷ. When Odysseus brings up the power of the army, Neoptolemus now resists the pressure (1243–51) (see above p. 223 on his earlier

224  Bob Corthals and Ineke Sluiter and Philoctetes seem to be on their way home: it is only the intervention by the undoubtedly authoritative Heracles that saves the charter myth of the Greeks and makes the stubborn Philoctetes accept to go to Troy after all: οὐκ ἀπιθήσω τοῖς σοῖς μύθοις, ‘I will not disobey your words’, 1447.51 ‘Obedience to authority’ is itself well-­anchored in the sociocognitive structure of primate life with its issues of status and social hierarchy, and the ensuing dom­ in­ance and submissiveness. This, too, may be recognizable everywhere in human societies. It can lay as strong a claim to being ‘basic’ and ‘cognitive’ as Clark’s analysis can.

12.6  The disobedient hero Milgram was explicit about the fact that his primary interest was in the obedient rather than the defiant subjects.52 Clark followed him in this and explained why it is so hard to extricate oneself from joint commitments. However, how about all those subjects—­in many iterations of the experiment a majority—­who in fact did not ‘obey’, but turned ‘defiant’? One of the most notable results of the Milgram experiments was actually the large between-­subjects variation in behaviour under exactly the same conditions. Milgram had of course observed the very substantial presence of defiant subjects, but could not explain the difference between the responses.53 In this respect, it is actually Sophocles who gives us the most open exploration of how a conflict might play out between one’s (self-­perceived) nat­ ural disposition and the pressures of (conflicting) social commitments, while acknowledging issues of obedience to authority and joint commitments.54 Neoptolemus after all turns into a defiant subject.55 Recent studies of the Milgram experiments suggest different solutions, all based (like Clark’s work) on reanalysis of the transcripts. Gibson, for instance, doubts whether these experiments are really about ‘obedience’, a concept not clearly delineated psychologically, and suggests that the communicative processes are key: it is all about rhetoric and persuasion—­Sophocles’ model in a modern sensitivity to the disapproval of the army, v. 93–4). Nonetheless, Odysseus threatens social retribution on leaving the stage: Ph. 1257–8 OD τῷ δὲ σύμπαντι στρατῷ/λέξω τάδ’ ἐλθών, ὅς σε τιμωρήσεται. 51  Remember Odysseus’ remark that Philoctetes would never obey (οὐ μὴ πίθηται, 103), see section 12.2, p. 215. 52  Milgram 1974, 5. 53  Milgram 1974, 162–3. Level of ‘aggression’, for instance, turned out not to explain the difference. Milgram 1974, chapter 13. 54  While Sophocles offers the most open exploration, he is also happy not to offer one simple key to understanding human behaviour. The action in Philoctetes is steered by many factors: phusis, an or­acle, rhetoric and persuasion, emotions such as pity, commitment to the collective of the army, desire for glory, and obedience to the deus ex machina. 55 See Fulkerson 2013, who sees Neoptolemus as a character rethinking a decision on moral grounds, and takes this as a sign of development.

SITUATED COGNITION: THE DISOBEDIENT HERO  225 guise.56 There is also an echo of Sophocles in the model that has held most appeal to modern researchers: the idea that the subjects’ responses can best be predicted by personal disposition (‘personality’) or by an interplay between personality and circumstances.57 A further step is the complication of the notion of selfhood.58 Sandra Jovchelovitch’s overview of modern psychological theory about selfhood or personhood suggests that selfhood itself most likely comes about in a constant interaction of processes of individuation and socialization. The self maintains a relationship to itself (through a dialogue between I and me), but self-­realization takes place in a sociocultural context that provides templates for selves, in terms of values, ideas, roles, and practices of cultural communication (‘social representations’). They serve as anchors for the ‘self ’.59 Of our three case-­studies, Sophocles, Milgram, and Clark, it is Sophocles who not only factors in the pressures of social interaction (all three authors ac­know­ ledge those) but also leaves room for the individual self (Neoptolemus). Milgram had been unable to identify a well-­defined role for personality, and Clark is focused on universal aspects of human interaction, which elide individual differences. How then does Neoptolemus’ resistance to Odysseus’ plan relate to his ‘personality’? In the light of modern theory of individuation, it is interesting to observe that all Sophoclean characters seem to relate Neoptolemus’ phusis to the simple fact that he is the son of Achilles. However, what does that mean for Neoptolemus’ behaviour? Neoptolemus has never met his own father, but has only heard about him and tries to imitate him. He is construing what behaviour would be appropriate to his phusis in the terms in which his heroic society (and Sophocles’ fifth-­ century one) envisages the role of the Achillean hero. Odysseus makes his own plan sound like what the son of Achilles is supposed to do.60 Philoctetes, too, uses phusis-rhetoric to commend actions of which he approves (now you are like your father), and to reject other ones (this is not like your father).61 Neoptolemus’ phusis itself, then, becomes the product of social interactions.62 One other point needs to be made about Neoptolemus’ phusis. Neoptolemus’ role in this play was new to the original audience in Athens, but a later episode in

56  Gibson 2019. 57  The literature is vast, but see e.g. de Swaan 2014 (also in the context of thinking about genocidal regimes); Bègue et al. 2015 (claiming personality predicts obedience); Blass 1991 and 2000 (promoting an interactional model). 58  This has not been discussed in the context of Sophocles, Milgram, or Clark. 59 Jovchelovitch forthcoming, in an explicit discussion of the psychological study and use of ancient Greek tragic characters. See also her extensive overview of the relevant literature. 60  See Schein 2013, ad Ph. 50–1. 61  Cf. section 12.2 p. 216. 62  Sophocles does not have his characters theorize the social component to inherited nobility, but he shows how it works. For that purpose it does not need to form part of his explicit (or perhaps even conscious) conceptual framework.

226  Bob Corthals and Ineke Sluiter the Trojan War made him a familiar mythological figure: the Athenian audience would have known that Neoptolemus had behaved like a godless and brutal war criminal after the fall of Troy, killing the king on the altar, ordering human sacrifice, and raping a virgin. Their expectations as to whether he could live up to his father would have been low in this prequel, although his preference for violence over words would sound true.63 Neoptolemus’ tragic double bind at the end of the play before the intervention by Heracles made it impossible for him to live up to his father either way. On the one hand, he could have broken his (over-)commitment to Philoctetes—­hardly Achilles-­like. On the other hand, had he taken Philoctetes home, that would have meant the end of the Greek ambition to capture Troy and attain glory:64 that, too, would have been a thoroughly un-­Achillean decision. Being able to just accept the instructions of Heracles (thus returning to ‘obedience’) saves him at this point.

12.7 Conclusion In this chapter, we have studied a Greek tragedy by Sophocles as part of the cultural archive of human cognition and situated metacognition constituted by world literature. But we have also expanded that notion to include the imaginative designs and analyses of scientific researchers, such as the psychologist Stanley Milgram and the cognitive linguist Herbert Clark. They, too, offer examples of socioculturally situated metacognition, rather than decontextualized universal truths—­this also goes for work done in the specific context of cognition studies.65 There is a definite task for the humanities here. Cognitively inflected in­ter­pret­ ations of human behaviour in different periods, including modern times, are themselves the object of historical study. They need full contextualization, i.e. an awareness of the intellectual agenda of their authors in their own environment, in order to grasp their preference for specific forms of metacognition. Such in­ter­ pret­ative models of human behaviour are not necessarily mutually exclusive: they are different because they provide answers to different questions.66 They have differential situated salience to agents in different contexts and to their understanding of self and others: the explanatory power of certain concepts is deemed greater in a given situation than that of others. Sophocles preferred phusis and deceitful

63  Neoptolemus’ role in the fall of Troy is evoked for the audience towards the end of Heracles’ speech, when he warns the two men to behave piously towards the gods, once they have captured the city, a warning that will go unheeded (Ph. 1441–4). 64  As Neoptolemus realizes: Ph. 1404. 65  For a wry commentary on universal concepts and the rather more messy and embodied ‘bricolage’ of human cognition, see Cave 2020. 66 This historicizing study of metacognitions should remedy the scepticism of critics like Sharrock 2018.

SITUATED COGNITION: THE DISOBEDIENT HERO  227 persuasion,67 Milgram ‘obedience to authority’, and Clark ‘joint commitment’ and ‘human interaction’ for good reasons, as we have shown. Such words and concepts are also ‘repositories of knowledge’, that can be reopened.68 As Morgan says: ‘There is no single mode of situated cognition. But there is a historical archive’ (2017, 222). We have contributed here to a history of reflection on cognition, which is unlikely to have come to a definitive end with the so-­called cognitive turn. This was not a story of unidirectional progress. There is not one final answer, one key to the understanding of all human cognition, since every effort at understanding is itself contextual and situated. However, each new insight and all new experimental data added to the archive of (meta)cognition deepen our understanding. New insights will likely be applicable to older examples of human interaction in the archive. And they are unlikely to be able to explain all such interactions, in whatever time and place, by themselves. While undoubtedly some concepts that have been developed historically are no longer usable, the archive is also stacked and cumulative: theories and concepts can be retooled and refined, but remain of use in later periods. The history of cognitive understanding is path-­dependent, and new ideas are anchored,69 not only in older ones but always also in the condition humaine of being human beings, human bodies in a world of other beings, in a certain historical, social, and cultural context, trying to make the best of it.

Acknowledgements We would like to thank Felix Budelmann and audiences at the Universities of Oxford, Leiden, and Edinburgh, and at the British Academy for incisive comments.

67 But see n. 54 for his willingness to leave the scenario played out in the Philoctetes overdetermined. 68  Morgan 2017, 221. 69  Sluiter 2017.

Bibliography Adler, H. and Gross  S. (2002), ‘Adjusting the Frame: Comments on Cognitivism and Literature’, Poetics Today 23.2: 195–220. Alber, J., Caracciolo, M., Iversen, S., Kukkonen, K., and Nielsen, H.  S. (eds.) (2018), Unnatural and Cognitive Perspectives on Narrative (a Theory Crossover), special issue of Poetics Today 39.3. Alber, J. and Heinze, R. (eds.) (2011), Unnatural Narratives—­Unnatural Narratology, Berlin. Alber, J., Iversen, S., Skov Nielsen, H., and Richardson, B. (2010), ‘Unnatural Narratives, Unnatural Narratology: Beyond Mimetic Models’, Narrative 18: 113–36. Allan, R. J. (2003), The Middle Voice in Ancient Greek: A Study in Polysemy, Amsterdam. Allan, R. J. (2011), ‘The Historical Present in Thucydides: Capturing the Case of αἱρεῖ and λαμβάνει’, in J. Lallot (ed.), The Historical Present in Thucydides: Semantics and Narrative Function, Leiden, 37–63. Allan, R. J. (2020), ‘Narrative Immersion: Some Linguistic and Narratological Aspects’, in J.  Grethlein, L.  Huitink, and A.  Tagliabue (eds.), Experience, Narrative and Literary Criticism in Ancient Greece, Oxford, 15–35. Anderson, M., Cairns, D., and Sprevak, M. (eds.) (2019), Distributed Cognition in Classical Antiquity, Edinburgh. Angelopoulou, A. (2018), ‘Feeling Words: Embodied Metaphors in Seven against Thebes’, in J. Lauwers, H. Schwall, and J. Opsomer (eds.), Psychology and the Classics: A Dialogue of Disciplines, Berlin, 62–76. Apperly, I. A. (2011), Mindreaders: The Cognitive Basis of ‘Theory of Mind’, Hove. Apperly, I.  A. (2012), ‘What is “Theory of Mind”? Concepts, Cognitive Processes and Individual Differences’, Quarterly Journal of Experimental Psychology 65: 825–39. Apperly, I.  A. and Butterworth, S.  A. (2009), ‘Do Humans Have Two Systems to Track Beliefs and Belief-like States?’, Psychological Review 116: 953–70. Arendt, H. (1963), Eichmann in Jerusalem: A Report on the Banality of Evil, New York. Armstrong, P. B. (2013), How Literature Plays with the Brain: The Neuroscience of Reading and Art, Baltimore, MD. Assmann, J. (1995), ‘Collective Memory and Cultural Identity’, New German Critique 65: 125–33. Assmann, J. (2008), ‘Communicative and Cultural Memory’, in A.  Erll and A.  Nünning (eds.), Cultural Memory Studies: An International and Interdisciplinary Handbook, Berlin, 109–18. Atlas, L.  Y. and Wager, T.  D. (2013), ‘Expectancies and Beliefs: Insights from Cognitive Neuroscience’, in K.  N.  Ochsner and S.  M.  Kosslyn (eds.), The Oxford Handbook of Cognitive Neuroscience. Volume 2: The Cutting Edges, Oxford 359–81. Atran, S. (2002), In Gods We Trust: The Evolutionary Landscape of Religion, Oxford. Auster, P. (1988), The Invention of Solitude, New York. Austin, J. L. (1965), How to Do Things with Words, Cambridge. Avery, H. C. (1965), ‘Heracles, Philoctetes, Neoptolemus’, Hermes 93: 279–97. Baddeley, A. (1997), Human Memory: Theory and Practice, Revised edition, Hove.

230 Bibliography Bain, D. (1977), Actor and Audience: A Study of Asides and Related Conventions in Greek Tragedy, Oxford. Bain, D. (1987), ‘Some Reflections on the Illusion in Greek Tragedy’, Bulletin of the Institute of Classical Studies 34: 1–14. Bakewell, G.  W. (2013), Aeschylus’ Suppliant Women: The Tragedy of Immigration, Madison, WI. Bakker, E. (1997), ‘Verbal Aspect and Mimetic Description in Thucydides’, in E.  Bakker (ed.), Grammar as Interpretation: Greek Literature in its Linguistic Contexts, Leiden, 7–54. Bakola, E. (2017), ‘Seeing the Invisible: Interior Spaces and Uncanny Erinyes in Aeschylus’ Oresteia’, in A. Kampakoglou and A. Novokhatko (eds.), Gaze, Vision and Visuality in Greek Literature, Berlin, 163–86. Banks, K. and Chesters, T. (eds.) (2018), Movement in Renaissance Literature: Exploring Kinesic Intelligence, Cham. Baraz, Y. and van den Berg, C. S. (2013), ‘Intertextuality: Introduction’, American Journal of Philology 134: 1–8. Barclay, C.  R. and DeCooke, P.  A. (1988), ‘Ordinary Everyday Memories: Some of the Things of Which Selves Are Made’, in U. Neisser and E. Winograd (eds.), Remembering Reconsidered: Ecological and Traditional Approaches to the Study of Memory, Cambridge, 91–124. Baron-Cohen, S. (1995), Mindblindness: An Essay on Autism and Theory of Mind, Cambridge, MA. Barrett, J. (2002), Staged Narrative: Poetics and the Messenger in Greek Tragedy, Berkeley, CA. Barrett, J. L. (2000), ‘Exploring the Natural Foundations of Religion’, Trends in Cognitive Science 4.1: 29–34. Bartz, J.  A., Tchalova, K., and Fenerci, C. (2016), ‘Reminders of Social Connection Can Attenuate Anthropomorphism: A Replication and Extension of Epley, Akalis, Waytz, and Cacioppo (2008)’, Psychological Science 27.12: 1644–50. Bate, J. and Rasmussen, E. (eds.) (2007), William Shakespeare, Complete Works, Basingstoke. Batero, M. (2016), ‘Tactless Scientists: Ignoring Touch in the Study of Joint Attention’, Philosophical Psychology 29.8: 1200–14. Beach, K. (1993), ‘Becoming a Bartender: The Role of External Memory Cues in a WorkDirected Educational Activity’, Applied Cognitive Psychology 7: 191–204. Bègue, L., Beauvois, J.  L., Courbet, D., Oberlé, D., Lepage, J., and Duke, A.  A. (2015), ‘Personality predicts obedience in a Milgram paradigm’, Journal of Personality 83: 299–306. Belfiore, E. S. (1992), Tragic Pleasures: Aristotle on Plot and Emotion, Princeton, NJ. Belluc, S. and Bénéjam, V. (eds.) (2018), Cognitive Joyce, Basingstoke. Best, S. and Marcus, S. (2009), ‘Surface Reading: An Introduction’, Representations 108: 1–21. Bevington, D. (ed.) (1998), Troilus and Cressida, The Arden Shakespeare, Third Series, Walton-on-Thames. Bianchi, E., Brill, S., and Holmes, B. (eds.) (2019), Antiquities Beyond Humanism, Oxford. Bielfeldt, R. (2016), ‘Sight and Light: Reified Gazes and Looking Artefacts in the Greek Cultural Imagination’, in M. Squire (ed.), Sight and the Ancient Senses, London, 123–42. Billings, J. (2018), ‘Orestes’ Urn in Word and Action’, in M. Mueller and M. Telò (eds.), The Materialities in Greek Tragedy, London, 49–62. Billings, J. and Budelmann, F. (forthcoming), Non-Hermeneutic Reading, special issue of Helios.

Bibliography  231 Blair, R. (2006), ‘Image and Action: Cognitive Neuroscience and Actor-Training’, in B. McCornachie and F. E. Hart (eds.), Performance and Cognition: Theatre Studies and the Cognitive Turn, London, 167–85. Blair, R. (2008), The Actor, Image, and Action: Acting and Cognitive Neuroscience, New York. Blair, R. (2009), ‘Cognitive Neuroscience and Acting: Imagination, Conceptual Blending, and Empathy’, The Drama Review 53.4: 93–103. Blair, R. and Cook, A. (2016), ‘Introduction’, in R.  Blair and A.  Cook (eds.), Theatre, Performance and Cognition: Languages, Bodies and Ecologies, London, 1–15. Blass, T. (1991), ‘Understanding Behavior in the Milgram Obedience Experiment: the Role of Personality, Situations, and their Interactions’, Journal of Personality and Social Psychology 60: 398–413. Blass, T. (ed.) (2000), Obedience to Authority: Current Perspectives on the Milgram Paradigm, London. Blundell, M.  W. (1988), ‘The phusis of Neoptolemus in Sophocles’ Philoctetes’, Greece & Rome 35: 137–48. Blundell, M. W. (1989), Helping Friends and Harming Enemies: A Study in Sophocles and Greek Ethics, Cambridge. Blundell, M.  W. (1990), Sophocles’ Oedipus at Colonus, Focus Classical Library, Newburyport, Mass. Blundell, S., Cairns, D., Craik, E., and Sorkin Rabinowits, N. (2013), ‘Introduction’, Helios 40: 3–37. Bolens, G. (2012), The Style of Gestures: Embodiment and Cognition in Literary Narrative, Baltimore, MD. Bona, G. (1974), ‘Λόγος e ἀλήθεια nell’ encomio di Elena di Gorgia’, Rivista di Filologia e di Istruzione Classica 102: 5–33. Bond, G. W. (1974), ‘Euripides’ Parody of Aeschylus’, Hermathena 118: 1–14. Bonifazi, A. (2018), ‘Embedded Focalization and Free Indirect Speech in Homer as Viewpoint Blending’, in J.  Ready and C.  Tsagalis (eds.), Homer in Performance: Rhapsodes, Narrators, and Characters, Austin, TX, 230–54. Bonifazi, A. (2019), ‘The Forbidden Fruit of Compression in Homer’, in P. Meineck et al. (2019b), 122–38. Bouffartigue, J. (2007), ‘Les statues divines du paganisme: objets artificiels ou surnaturels?’, in C.  Delattre (ed.), Objets sacrés, objets magiques de l’Antiquité au Moyen-âge, Paris, 53–64. Bowie, A.  M. (1997), ‘Tragic Filters for History: Euripides’ Supplices and Sophocles’ Philoctetes’, in C. Pelling (ed.), Greek Tragedy and the Historian, Oxford, 39–62. Boyd, B. (2009), On the Origin of Stories: Evolution, Cognition, and Fiction, Cambridge MA. Boyer, P. (1996), ‘What Makes Anthropomorphism Natural: Intuitive Ontology and Cultural Representations’, Journal of the Royal Anthropological Institute 2: 83–97. Boyer, P. (2001), Religion Explained: The Evolutionary Origins of Religious Thought, New York. Boyer, P. (2018), Minds Make Societies: How Cognition Explains the World Humans Create, New Haven, CT. Boyer, P. and Baumard, N. (2016), ‘The Diversity of Religious Systems Across History: An Evolutionary Cognitive Approach’, in J.  R.  Liddle and T.  K.  Shackleford (eds.), The Oxford Handbook of Evolutionary Psychology and Religion, New York, 1–25. Bracher, M. (2013), Literature and Social Justice, Austin, TX. Bremmer, J.  N. (2013), ‘The Agency of Greek and Roman Statues: From Homer to Constantine’, Opuscula, Annual of the Swedish Institutes at Athens and Rome 6: 7–21. Brock, R. (2013), Greek Political Imagery from Homer to Aristotle, London.

232 Bibliography Brown, M.  F. (2018), ‘Musings on Comparative Directions for Situated Cognition’, Comparative Cognition & Behavior Reviews 13: 21–4. Brown, W.  E. (1965), ‘Sophocles’ Ajax and Homer’s Hector’, The Classical Journal 61: 118–21. Bruhn, M.  J. (2011), ‘Introduction. Exchanging Values: Poetics and Cognitive Science’, Poetics Today 32: 403–60. Bruhn, M. J. and Wehrs, D. R. (eds.) (2014), Cognition, Literature and History, New York. Budelmann, F. (2000), The Language of Sophocles: Communality, Communication and Involvement, Cambridge. Budelmann, F. (2013), ‘Dramatic Illusion and Realism’, in H.  M.  Roisman (ed.), The Encyclopedia of Greek Tragedy, Chichester, 299–301. Budelmann, F. (2018), ‘Lyric Minds’, in F. Budelmann and T. Phillips (eds.), Textual Events: Performance and the Lyric in Early Greece, Oxford, 235–56. Budelmann, F. (2019), ‘Group Minds in Classical Athens? Chorus and Dēmos as Case Studies of Collective Cognition’, in Anderson et al. (2019), 190–208. Budelmann, F. and Easterling, P. (2010), ‘Reading Minds in Greek Tragedy’, Greece and Rome 57: 289–303. Budelmann, F. and LeVen, P. (2014), ‘Timotheus’ Poetics of Blending: A Cognitive Approach to the Language of the New Music’, Classical Philology 109: 191–210. Budelmann, F., Maguire, L., and Teasdale, B. (2013), ‘The Play’s the Thing: Audience Reactions to Greek and Shakespearean tragedy’, Times Literary Supplement 5755: 13–15 (without charts); http://ora.ox.ac.uk/objects/uuid:da99b8a5-1102-4d47-aabb-118ca658722d. Budelmann, F., Maguire, L., and Teasdale, B. (2016), ‘Ambiguity and Audience Response’, Arion 23: 89–114. Budelmann, F. and van Emde Boas, E. (2020), ‘Attending to Tragic Messenger Speeches’, in J. Grethlein, L. Huitink, and A. Tagliabue (eds.), Experience, Narrative, and Criticism in Ancient Greece: Under the Spell of Stories, Oxford, 59–80. Budra, P. V. and Werier, C. (eds.) (2016), Shakespeare and Consciousness, New York. Buhrmann, T., Di Paolo, E. A., and Barandiaran, X. (2013), ‘A Dynamical Systems Account of Sensorimotor Contingencies’, Frontiers in Psychology 4.285. Burger, J.  M. (2009), ‘Replicating Milgram: would people still obey today?’, American Psychologist 64: 1–11. Burian, P. (1974), ‘Pelasgus and Politics in Aeschylus’ Danaid Trilogy’, Wiener Studien 8: 5–14. Burian, P. (2012), ‘Polyphonic Ajax’, in K.  Ormand (ed.), A Companion to Sophocles, Oxford, 69–83. Burke, M. and Troscianko, E. T. (eds.) (2017), Cognitive Literary Science: Dialogues Between Literature and Cognition, New York. Burrow, C. (2006), ‘Introduction’, in R. A. Foakes (ed.), William Shakespeare: Troilus and Cressida, London, xxi–lxiii. Bussels, S. (2012), The Animated Image: Roman Theory on Naturalism, Vividness and Divine Power, Berlin. Buxton, R. (2006), ‘Weapons and Day’s White Horses: The Language of Ajax’, in I. J. F. de Jong and A. Rijksbaron (eds.), Sophocles and the Greek Language, Leiden, 13–23. Cairns, D. (2006), ‘Virtue and Vicissitude: The Paradoxes of the Ajax’, in D.  Cairns and V.  Liapis (eds.), Dionysalexandro: Essays on Aeschylus and His Fellow Tragedians in Honour of Alexander F. Garvie, Swansea, 99–131. Cairns, D. (2011a), ‘Veiling Grief on the Tragic Stage’, in D.  Lacourse Munteanu (ed.), Emotion, Gender, and Genre in Classical Antiquity, London, 15–33.

Bibliography  233 Cairns, D. (2011b), ‘Looks of Love and Loathing: Cultural Models of Vision and Emotion in Ancient Greek Culture’, Métis 9: 37–50. Cairns, D. (2016), ‘Mind, Body, and Metaphor in Ancient Greek Concepts of Emotion’, L’Atelier du Centre de recherches historiques 16. Cairns, D. (2017a), ‘Mind, Metaphor, and Emotion in Euripides (Hippolytus) and Seneca (Phaedra)’, Maia 69: 247–67. Cairns, D. (2017b), ‘Horror, Pity, and the Visual in Ancient Greek Aesthetics’, in D. Cairns and D.  Nelis (eds.), Emotions in the Classical World: Methods, Approaches, and Directions, Stuttgart, 53–78. Cairns, D. (2019), ‘Distributed Cognition and the Classics’, in Anderson et al. (2019), 18–36. Cairns, D. (2020), ‘Phaedra’s Fantasy Other: Phenomenology and the Enactive Mind in Euripides’ Hippolytus’, in M. Liatsi (ed.), Ethics in Ancient Greek Literature: Aspects of Ethical Reasoning from Homer to Aristotle, Trends in Classics Supplement 102, Berlin, 117–28. Cairns, D. (2021), ‘The Dynamics of Emotion in Euripides’ Medea’, Greece and Rome 68: 8–26. Calder, W.  M., III (1988), ‘Vita Aeschyli 9: Miscarriages in the Theatre of Dionysos’, The Classical Quarterly 38: 554–5. Campeggiani, P. (2020), ‘Nec cogitare sed facere: The Paradox of Fiction at the Tribunal of Ancient Poetics’, Theoria 86: 709–26. Canevaro, L.-G. (2019), ‘Materiality and Classics: (Re)Turning to the Material’, The Journal of Hellenic Studies, 139: 222–32. Caracciolo, M. (2014), The Experientiality of Narrative: An Enactivist Approach, Berlin. Caracciolo, M. (2016a), ‘Cognitive Literary Studies and the Status of Interpretation. An Attempt at Conceptual Mapping’, New Literary History 47: 187–207. Caracciolo, M. (2016b), Strange Narrators in Contemporary Fiction: Explorations in Readers’ Engagement with Characters, Lincoln, NB. Carroll, J. (2004), Literary Darwinism: Evolution, Human Nature, and Literature, New York and London. Carroll, J. (2011), Reading Human Nature: Literary Darwinism in Theory and Practice, Albany, NY. Carruthers, P. (2000), ‘The Evolution of Consciousness’, in P. Carruthers and A. Chamberlin (eds.), Evolution and the Human Mind:Modularity, Language and Meta-Cognition, Cambridge, 254–75. Carston, R. (2010), ‘Metaphor: Ad Hoc Concepts, Literal Meaning and Mental Images’, Proceedings of the Aristotelian Society 110: 297–323. Casey, E. S. (2000), Remembering: A Phenomenological Study, 2nd edn, Bloomington, IN. Cave, T. (2016), Thinking with Literature: Towards a Cognitive Criticism, Oxford. Cave, T. (2017), ‘Situated Cognition: The Literary Archive’, Poetics Today 38: 235–53. Cave, T. (2020), ‘Imagining the Emergence of the Human: Reflections on Chris Johnson’s Late Work’, Paragraph 43: 45–57. Chapman, G. A. H. (1983), ‘Some Notes on Dramatic Illusion in Aristophanes’, American Journal of Philology 104: 1–23. Chaston, C. (2010), Tragic Props and Cognitive Function: Aspects of the Function of Images in Thinking, Leiden. Chater, N. and Loewenstein, G. (2016), ‘The Under-Appreciated Drive for Sense-Making’, Journal of Economic Behavior and Organization 126: 137–54. Chesi, G. M. and Spiegel, F. (eds.) (2019), Classical Literature and Posthumanism, London. Chesters, T. (2014), ‘Social Cognition: A Literary Perspective’, Paragraph 37: 62–78.

234 Bibliography Chevalier, F. H. G. (2011), ‘Language and Social Interaction: an introduction to conversation analysis’, Nottingham French Studies 50: 1–18. Christensen, J. P. (2018), ‘Human Cognition and Narrative Closure: The Odyssey’s Open End’, in P. Meineck et al. (2019b), 139–55. Clark, A. (1997), Being There: Putting Brain, Body, and World Together Again, Cambridge, MA. Clark, A. (2010), ‘Material Surrogacy and the Supernatural: Reflections on the Role of Artefacts in ‘Off-line’ Cognition’, in Malafouris and Renfrew (2010), 23–8. Clark, A. (2013), ‘Whatever Next? Predictive Brains, Situated Agents, and the Future of Cognitive Science’, Behavioral and Brain Sciences 36: 181–204. Clark, A. (2015), Surfing Uncertainty: Prediction, Action, and the Embodied Mind, Oxford. Clark, A. and Chalmers, D. (1998), ‘The Extended Mind’, Analysis 58: 7–19. Clark, H. H. (1996), Using Language, Cambridge. Clark, H.  H. (2006), ‘Social Actions, Social Commitments’, in Nicholas  J.  Enfield and Stephen C. Levinson (eds.), Roots of Human Sociality: Culture, Cognition, and Human Interaction, Oxford, 126–50. Clark, H.  H. and Henetz, T. (2014), ‘Working Together’, in T.  M.  Holtgraves (ed.), The Oxford Handbook of Language and Social Psychology, Oxford, 85–97. Clark, M.  S., Boothby, E.  J., and Bargh, J.  A. (2014), ‘Shared Experiences are Amplified’, Psychological Science 25.12: 2209–16. Cochrane, T. (2009), ‘Joint Attention to Music’, British Journal of Aesthetics 49: 59–73. Cook, A. (2007), ‘Interplay: The Method and Potential of a Cognitive Scientific Approach to Theatre’, Theatre Journal 59: 579–94. Cook, A. (2010), Shakespearean Neuroplay: Reinvigorating the Study of Dramatic Texts and Performance Through Cognitive Science, New York. Crane, M. T. (2001), Shakespeare’s Brain: Reading with Cognitive Theory, Princeton, NJ. Culler, J. (1997), Literary Theory: A Very Short Introduction, Oxford. Culpeper, J. (2001), Language and Characterisation: People in Plays and Other Texts, Harlow. Culpeper, J. (2002), ‘A Cognitive Stylistic Approach to Characterisation’, in E. Semino and J.  Culpeper (eds.), Cognitive Stylistics: Language and Cognition in Text Analysis, Amsterdam, 251–78. Currie, B. (forthcoming), ‘Recognizing Odysseus, Reading Penelope: The Anagnōrisis in the Twenty-Third Book of the Odyssey’, Journal of Hellenic Studies. Currie, G. (2010), Narratives and Narrators: A Philosophy of Stories, Oxford. Dalfen, J. (1974), Polis und Poiesis: Die Auseinandersetzung mit der Dichtung bei Platon und seinen Zeitgenossen, Munich. Dancygier, B. (2016), ‘Multimodality and Theatre: Material Objects, Bodies and Language’, in R. Blair and A. Cook (eds.), Theatre, Performance and Cognition: Languages, Bodies and Ecologies, London, 21–39. Dancygier, B. and Sweetser, E. (2014), Figurative Language, New York. Davidson, J. F. (1988), ‘Homer and Sophocles’ Electra’, Bulletin of the Institute of Classical Studies 35: 45–72. Davies, M. (1991), Sophocles: Trachiniae, Oxford. Davis, P. (2007), Shakespeare Thinking, London. Dedoussi, C. (1995), ‘Greek Drama and its Spectators’, in A. Griffiths (ed.), Stage Directions: Essays in Ancient Drama in Honour of E. W. Handley, Bulletin of the Institute of Classical Studies Supplement 66, London, 123–32. de Jong, I. J. F. (1994), ‘Πιστὰ τεκμήρια in Soph. El. 774’, Mnemosyne 47: 679–81.

Bibliography  235 de Jong, I. J. F. (2006), ‘Where Narratology Meets Stylistics: The Seven Versions of Ajax’ Madness’, in I. J. F. de Jong and A. Rijksbaron (eds.), Sophocles and the Greek Language, Leiden, 73–93. de Romilly, J. (1995), ‘L’assemblée du people dans l’Oreste d’Euripide’, in J.  de Romilly, Tragédies grecques au fil des ans, Paris, 143–57. Originally in Studi Classici in Onore di Quinto Cataudella, vol. 1. Catania (1972), 237–51. Destrée, P. and Herrmann, F. G. (eds.) (2011), Plato and the Poets, Leiden. de Swaan, Abram (2014), Compartimenten van vernietiging: Over genocidale regimes en hun daders, Amsterdam. Engl. trans. The Killing Compartments: the Mentality of Mass Murder, New Haven, CT-London (2015). De Temmerman, K. and van Emde Boas, E. (eds.) (2018a), Characterization in Ancient Greek Literature, Leiden. De Temmerman, K. and van Emde Boas, E. (2018b), ‘Introduction’, in K. De Temmerman and E.  van Emde Boas (eds.), Characterization in Ancient Greek Literature, Leiden, 1–23. Di Benedetto, V. (1965), Euripidis Orestes, Florence. Dietrich, B. C. (1962), ‘Demeter, Erinys, Artemis’, Hermes 90: 129–48. Diggle, J. (1994), Euripidis Fabulae, Oxford. Dobson, M. (2018), ‘Why Does Orestes Stay Mad?’, in J.  Lauwers, H.  Schwall, and J. Opsomer (eds.), Psychology and the Classics: A Dialogue of Disciplines, Berlin, 158–70. Dodds, E. R. (1960), Euripides: Bacchae, 2nd edn, Oxford. Donoghue, D. (2014), Metaphor, Cambridge, MA. Donohue, A. (1988), Xoana and the Origins of Greek Sculpture, Atlanta, GA. Dosi, A. (1968), ‘La definizione gorgiana della tragedia’, Rendiconti dell’ Istituto Lombardo 102: 35–90. Dressler, M., Shirer, W. R., Konrad, B. N., Muller, N. C. J., Wagner, I., Fernandez, G. S. E., Czisch, M., and Greicius, M. D. (2017), ‘Mnemonic Training Reshapes Brain Networks to Support Superior Memory’, Neuron 93.5: 1227–35. Duncan, A. (2023), ‘Visualising the Stage’, in P.  Burian and J.  Bromberg (edd.), A Companion to Aeschylus, Oxford, 214–229. Duncan, A. (2005), ‘Gendered Interpretations: Two Fourth-Century bce Performances of Sophocles’ Electra’, Helios 32: 55–79. Dunn, M. (1996), Tragedy’s End: Closure and innovation in Euripidean drama, New York. Easterlin, N. (ed.) (2019), Knowledge, Understanding, Well-Being: Cognitive Literary Studies, special issue of Poetics Today 40.3. Easterling, P. E. (1973), ‘Repetition in Sophocles’, Hermes 101: 14–34. Easterling, P. E. (1982), Sophocles: Trachiniae, Cambridge. Easterling, P. E. (1984), ‘The Tragic Homer’, Bulletin of the Institute of Classical Studies 31: 1–8. Easterling, P. E. (1990), ‘Constructing Character in Greek Tragedy’, in C. B. R. Pelling (ed.), Characterization and Individuality in Greek Literature, Oxford, 83–99. Eder, J., Jannidis, F., and Schneider, R. (2010), ‘Characters in Fictional Worlds: An Introduction’, in J.  Eder, F.  Jannidis, and R.  Schneider (eds.), Characters in Fictional Worlds: Understanding Imaginary Beings in Literature, Film, and Other Media, Berlin, 3–65. Edmunds, S. T. (2012), ‘Picturing Homeric Weaving’, in V. Bers, D. Elmer, and L. Muellner (eds.), Donum natalicium digitaliter confectum Gregorio Nagy septuagenario a discipulis collegis familiaribus oblatum, Washington DC. http://nrs.harvard.edu/urn-3:hul. ebook:CHS_Bers_etal_eds.Donum_Natalicium_Gregorio_Nagy.2012 Eilan, N., Hoerl, C., McCormack, T., and Roessler, J. (eds.) (2005), Joint Attention: Communication and Other Minds. Issues in Philosophy and Psychology, Oxford.

236 Bibliography Ellendt, F. (1965), Lexicon Sophocleum, 2nd revised edn, ed. H.  Genthe, Hildesheim. Original edn 1872. Elmer, D.  F. (2013), The Poetics of Consent. Collective Decision Making and the Iliad, Baltimore, MD. Epley, N., Akalis, S., Waytz, A., and Cacioppo, J. T. (2008a), ‘Creating Social Connection through Inferential Reproduction: Loneliness and Perceived Agency in Gadgets, Gods, and Greyhounds’, Psychological Science 19.2: 114–20. Epley, N., Waytz, A., Akalis, S., and Cacioppo, J.  T. (2008b), ‘When we Need a Human: Motivational Determinants of Anthropomorphism’, Social Cognition 26.2: 143–55. Erll, A. (2010), ‘Cultural Memory Studies: An introduction’, in A. Erll and A. Nünning (eds.), Cultural Memory Studies: An International and Interdisciplinary Handbook, Berlin, 1–15. Evans, V. and Green, M. (2006), Cognitive Linguistics: An Introduction, Edinburgh. Falletti, C., Sofia, G., and Jacono, V. (eds.) (2016), Theatre and Cognitive Neuroscience, London. Fauconnier, G. (1997), Mappings in Thought and Language, Cambridge. Fauconnier, G. and Turner, M. (1996), ‘Blending as a Central Process of Grammar’, in A. Goldberg (ed.), Conceptual Structure, Discourse, and Language, Cambridge, 113–30. Fauconnier, G. and Turner, M. (1998), ‘Conceptual Integration Networks’, Cognitive Science 22.2: 133–87. Fauconnier, G. and Turner, M. (2002), The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities, New York. Fauconnier, G. and Turner, M. (2008), ‘Rethinking Metaphor’, in R.  W.  Gibbs (ed.), The Cambridge Handbook of Metaphor and Thought, Cambridge, 53–66. Favorini, A. (2008), Memory in Play: From Aeschylus to Sam Shepard, New York. Felperin, H. (1972), Shakespearian Romance, Princeton, NJ. Felski, R. (2015), The Limits of Critique, Chicago, IL. Ferrari, G. R. F. (2019), ‘Aristotle on Musical Catharsis and the Pleasure of a Good Story’, Phronesis 64: 117–71. Ferrari, P. F. and Rizzolatti, G. (eds.) (2014), ‘Mirror Neurons: Fundamental Discoveries, Theoretical Perspectives and Clinical Implications’, Philosophical Transactions of the Royal Society B 369: 1644. Finglass, P. J. (2007), Sophocles: Electra, Cambridge. Finglass, P. J. (2011), Sophocles: Ajax, Cambridge. Finglass, P.  J. (2012), ‘Ajax’, in A.  Markantonatos (ed.), Brill’s Companion to Sophocles, Leiden, 59–72. Flashar, H. (1956), ‘Die medizinischen Grundlagen der Lehre von der Wirkung der Dichtung in der griechischen Poetik’, Hermes 84: 12–48. Flashar, H. (1958), Der Dialog Ion als Zeugnis platonischer Philosophie, Berlin. Fletcher, J. (2013), ‘Weapons of Friendship: Props in Sophocles’ Philoctetes and Ajax’, in G.  W.  M.  Harrison and V.  Liapis (eds.), Performance in Greek and Roman Theatre, Leiden, 199–216. Fludernik, M. (1996), Towards a ‘Natural’ Narratology, London. Fludernik, M. (2011), ‘Introduction’, in M.  Fludernik (ed.), Beyond Cognitive Metaphor Theory: Perspectives on Literary Metaphor, New York, 1–16. Försterling, F. (2001), Attribution: An Introduction to Theories, Research, and Applications, Hove. Foley, H.  P. (1980), ‘The Masque of Dionysus’, Transactions of the American Philological Association 110: 107–33. Foley, H. P. (2001), Female Acts in Greek Tragedy, Princeton, NJ.

Bibliography  237 Ford, A. (2016), ‘Catharsis, Music, and the Mysteries in Aristotle’, Skene: Journal of Theatre and Drama Studies 2.1: 23–41. Foster, J. K. (2009), Memory: A Very Short Introduction, Oxford. Fraenkel, E. (1950), Agamemnon, Oxford. Freeman, M. (2007), ‘Cognitive Linguistic Approaches to Literary Studies’, in D. Geeraerts and H.  Cuyckens (eds.), The Oxford Handbook of Cognitive Linguistics, Oxford, 1175–202. Freeman, M. (2010), ‘Telling Stories: Memory and Narrative’, in S.  Radstone and B. Schwartz (eds.), Memory: Histories, Theories, Debates, New York, 263–77. Friederici, A.  D. (2004), ‘Event-Related Brain Potential Studies in Language’, Current Neurology and Neuroscience Reports 4: 466–70. Friis Johansen, H. and Whittle, E. W. (1980), Aeschylus: The Suppliants, 3 vols., Copenhagen. Fulkerson, L. (2013), ‘Neoptolemus grows up? “Moral development” and the interpretation of Sophocles’ Philoctetes’, The Cambridge Classical Journal 52: 49–61. Gaifman, M. (2012), Aniconism in Greek Antiquity, Oxford. Gallagher, S. (2001), ‘The Practice of Mind: Theory, Simulation or Primary Interaction?’, Journal of Consciousness Studies 8: 83–108. Gallagher, S. (2005), How the Body Shapes the Mind, Oxford. Gallagher, S. (2009), ‘Two Problems of Intersubjectivity’, Journal of Consciousness Studies 16.6–7: 289–308. Gallagher, S. (2017), Enactivist Interventions: Rethinking the Mind, Oxford. Gallagher, S. (2020), Action and Interaction, Oxford. Gallagher, S. and Hutto, D. D., (2008), ‘Understanding Others through Primary Interaction and Narrative Practice’, in J. Zlatev, T. Racine, C. Sinha, and E. Itkonen (eds.), The Shared Mind: Perspectives on Intersubjectivity, Amsterdam, 17–38. Gallese, V., Fadiga, L., Fogassi, L., and Rizzolatti, G. (1996), ‘Action Recognition in the Premotor Cortex’, Brain 119: 593–609. Garner, S.  B. (2018), Kinesthetic Spectatorship in the Theatre: Phenomenology, Cognition, Movement, Cham. Garvie, A. F. (1986), Aeschylus’ Choephori, Oxford. Garvie, A. F. (1998), Sophocles, Ajax, Warminster. Garvie, A. F. (2006), Aeschylus’ Supplices: Play and Trilogy, 2nd edn, Cambridge. Garzya, A. (1987), ‘Gorgia e l’ apate della tragedia’, in S. Boldrini (ed.), Filologia e forme letterarie: Studi offerti a Francesco Della Corte, Urbino, 245–60. Gazzola, V. and Keysers, C. (2009), ‘The observation and execution of actions share motor and somatosensory voxels in all tested subjects. Single-subject analyses of unsmoothed fMRI Data’, Cerebral Cortex 19: 1239–55. Gell, A. (1998), Art and Agency. An Anthropological Theory, New York and Oxford. Gellie, G. (1988), ‘The Character of Medea’, Bulletin of the Institute of Classical Studies 35.1: 15–22. Gerrig, R.  J. (1993), Experiencing Narrative Worlds: On the Psychological Activities of Reading, New Haven, CT. Gerrig, R.  J. and Prentice, D.  A. (1991), ‘The Representation of Fictional Information’, Psychological Science 2.5: 336–40. Gianotti, G. F. (1975), Per una poetica pindarica, Turin. Gibbs, R. W. (2006), ‘The Psychological Status of Image Schemas’, in B. Hampe (ed.), From Perception to Meaning: Image Schemas in Cognitive Linguistics, Berlin, 113–36. Gibbs, R. W. (ed.) (2008), The Cambridge Handbook of Metaphor and Thought, Cambridge.

238 Bibliography Gibbs, R. W. (2017), ‘Embodied Dynamics in Literary Experience’, in M.  Burke and E.  Troscianko (eds.), Cognitive Literary Science: Dialogues Between Literature and Cognition, New York, 219–38. Gibbs, R. W. and Colston, H. (1995), ‘The Cognitive Psychological Reality of Image Schemas and Their Transformations’, Cognitive Linguistics 6: 347–78. Gibbs, R. W. and Matlock, T. (2008), ‘Metaphor, Imagination, and Simulation: Psycholinguistic Evidence’, in Gibbs (2008), 161–76. Gibert, J. (1995), Change of Mind in Greek Tragedy, Göttingen. Gibson, J. J. (1979), The Ecological Approach to Visual Perception, Boston, MA. Gibson, M. (2008), Objects of the Dead: Mourning and Memory in Everyday Life, Melbourne. Gibson, S. (2019), Arguing, Obeying, and Defying: A Rhetorical Perspective on Stanley Milgram’s Obedience Experiments, Cambridge. Gill, C. (1996), Personality in Greek Epic, Tragedy, and Philosophy, Oxford. Gill, C. (2019), ‘Enactivism and Embodied Cognition in Stoicism and Plato’s Timaeus’, in Anderson et al. (2019), 150–68. Glebkin, V. (2015), ‘Is Conceptual Blending the Key to the Mystery of Human Evolution and Cognition?’, Cognitive Linguistics 26: 95–111. Godley, A. D. (trans.) (1920), Herodotus: The Histories, Cambridge, MA. Goff, B. E. (1999), ‘Try to Make it Real Compared to What? Euripides’ Electra and the Play of Genres’, in D. Sansone, M. J. Cropp, and K. Lee (eds.), Euripides and Tragic Theatre in the Late Fifth Century, special issue of Illinois Classical Studies 24/25, Champaign, IL, 93–105. Goldhill, S. (1984), Language, Sexuality, Narrative, the Oresteia, Cambridge. Goldhill, S. (1986), Reading Greek Tragedy, Cambridge. Goldhill, S. (2002), Sophocles and the Language of Tragedy, Oxford. Gosden, C. (2010), ‘The Death of the Mind’, in Malafouris and Renfrew (2010), 39–46. Gottschall, J. (2008), The Rape of Troy: Evolution, Violence, and the World of Homer, Cambridge. Gould, J. (1978), ‘Dramatic Character and “Human Intelligibility” in Greek Tragedy’, Proceedings of the Cambridge Philological Society 24: 43–67. Gould, J. (2012), ‘Euripides’, in S. Hornblower, A Spawforth, and S. Eidinow (eds.), Oxford Classical Dictionary, 4th edn, Oxford, 551–3. Green, M. C. and Brock, T. C. (2000), ‘The Role of Transportation in the Persuasiveness of Public Narratives’, Journal of Personality and Social Psychology 79.5: 701–21. Gregory, J. (1985), ‘Some Aspects of Seeing in Euripides’ Bacchae’, Greece and Rome 32: 23–31. Gregory, S.  E.  A. and Jackson, M.  C. (2017), ‘Joint Attention Enhances Visual Working Memory’, Journal of Experimental Psychology: Learning, Memory, and Cognition 43.2: 237–49. Grethlein, J. (2015), ‘Is Narrative “the Description of Fictional Mental Functioning”? Heliodorus against Palmer, Zunshine & Co.’, Style 49: 257–84; responses 285–98. Grethlein, J. (2016), ‘Sight and Reflexivity: Theorizing Vision in Greek Vase-Painting’, in M. Squire (ed.), Sight and the Ancient Senses, London, 85–106. Grethlein, J. (2017), Aesthetic Experiences and Classical Antiquity: The Significance of Form in Narratives and Pictures, Cambridge. Grethlein, J. (2021), The Ancient Aesthetics of Deception, Cambridge. Grethlein, J. and Huitink, L. (2017), ‘Homer’s Vividness: An Enactive Approach’, The Journal of Hellenic Studies 137: 67–91. Griffith, M. (1999), Sophocles: Antigone, Cambridge.

Bibliography  239 Griffith, M. (2002), ‘Slaves of Dionysos: Satyrs, Audience, and the Ends of the Oresteia’, Classical Antiquity 21: 195–258. Griffith, M. (2006), ‘The Subject of Desire in Sophocles’ Antigone’, in V.  Pedrick and S. M. Oberhelman (eds.), The Soul of Tragedy: Essays in Athenian Drama, Chicago, IL. Griffith, M. (2015), Greek Satyr Play: Five Studies, Berkeley, CA. Grodal, T. (1999), Moving Pictures: A New Theory of Film Genres, Feelings, and Cognition, Oxford. Grodal, T. (2009), Embodied Visions: Evolution, Emotion, Culture and Film, Oxford. Groeneboom, P. (1935), Sophocles’ Electra, 2nd edn, Groningen. Grünbaum, T. (2007), ‘Action Between Plot and Discourse’, Semiotica 165: 295–314. Guthrie, S. E. (1993), Faces in the Clouds: A New Theory of Religion, Oxford. Gutzwiller, K. (2000), ‘The Tragic Mask of Comedy: Metatheatricality in Menander’, Classical Antiquity 19: 102–37. Halbwachs, M. (1992), On Collective Memory, ed. and trans. L. A. Coset, Chicago, IL. Halbwachs, M. (2011), ‘From The Collective Memory’, in J. K. Olick, V. Vinitzky-Seroussi, and D. Levy (eds.), The Collective Memory Reader. Oxford, 139–49. Trans. F. J. Ditter, Jr., and V. Yazdi Ditter, original publication New York 1980. Hall, E. (2009), ‘Deianeira Deliberates: Precipitate Decision-Making and Trachiniae’, in S. Goldhill and E. Hall (eds.), Sophocles and the Greek Tragic Tradition, Cambridge, 69–96. Halliwell, S. (2002), The Aesthetics of Mimesis: Ancient Texts and Modern Problems, Princeton, NJ. Halliwell, S. (2011), Between Ecstasy and Truth: Interpretations of Greek Poetics from Homer to Longinus, Oxford. Halliwell, S. (1984), ‘Plato and Aristotle on the Denial of Tragedy’, Proceedings of the Cambridge Philological Society 30: 49–71. Hamilton, C. (2002), ‘Conceptual Integration in Christine de Pizan’s City of Ladies’, in E.  Semino and J.  Culpeper (eds.), Cognitive Stylistics Language and Cognition in Text Analysis, Amsterdam, 1–22. Hamilton, C. (2003), ‘A cognitive grammar of “Hospital Barge” by Wilfred Owen’, in G. Steen and J. Gavins (eds.), Cognitive Poetics in Practice, London, 55–66. Hamilton, J. (2018), ‘Attention to Theatrical Performances’, in R. Kemp and B. McConachie (eds.), The Routledge Companion to Theatre, Performance, and Cognitive Science, London, 216–24. Hashimazade, N., Mayles, G. and Black, J. (2017), A Dictionary of Economics, Oxford. Heath, M. (1987), The Poetics of Greek Tragedy, Stanford, CA. Hedreen, G. (2007), ‘Involved Spectatorship in Archaic Greek Art’, Art History 30.2: 217–46. Heersmink, R. (2013), ‘A Taxonomy of Cognitive Artifacts: Function, Information, and Categories’, Review of Philosophy and Psychology 4: 465–81. Heiden, B. (1989), Tragic Rhetoric: An Interpretation of Sophocles’ Trachiniae, Frankfurt am Main. Helms, N.  R. (2012), ‘Conceiving Ambiguity: Dynamic Mindreading in Shakespeare’s Twelfth Night’, Philosophy and Literature 36: 122–35. Helms, N. R. (2019), Cognition, Mindreading, and Shakespeare’s Characters, Basingstoke. Hemenover, S.  H. and Schimmack, U. (2007), ‘That’s Disgusting! . . ., But Very Amusing: Mixed Feelings of Amusement and Disgust’, Cognition and Emotion 21: 1102–13. Henrichs, A. (1993), ‘The Tomb of Aias and the Prospect of Hero Cult in Sophokles’, Classical Antiquity 12: 165–80. Henrichs, A. (1995), ‘Why Should I Dance? Choral Self-Referentiality in Greek Tragedy’, Arion 3.1: 56–111.

240 Bibliography Herman, D. (2008), ‘Narrative Theory and the Intentional Stance’, Partial Answers 6: 233–60. Herman, D. (2009), ‘Storied Minds: Narrative Scaffolding for Folk Psychology’, Journal of Consciousness Studies 16.6–8: 40–68. Herman, D. (2011), ‘Introduction’, in D.  Herman (ed.), The Emergence of Mind: Representations of Consciousness in Narrative Discourse in English, Lincoln, NB, 1–40. Herman, D. (2013), ‘Cognitive Narratology’, in P. Hühn, J. Pier, W. Schmid, and J. Schönert (eds.), The Living Handbook of Narratology, Hamburg. http://www.lhn.uni-hamburg.de. Hertel, P. (2004), ‘Memory for emotional and non-emotional events in depression. A question of habit?’, in D.  Reisberg and P.  Hertel (eds.), Memory and Emotion, New York, 186–216. Hesk, J. (2000), Deception and Democracy in Classical Athens, Cambridge. Hesk, J. (2003), Sophocles, Ajax, London. Hester, D. (1971), ‘Sophocles the Unphilosophical. A Study in the Antigone’, Mnemosyne 24: 11–59. Hickok, G. (2009), ‘Eight Problems for the Mirror Neuron Theory of Action Understanding in Monkeys and Humans’, Journal of Cognitive Neuroscience 21: 1229–43. Hogan, P. C. (2011), Affective Narratology: The Emotional Structure of Stories, Lincoln, NB. Hogan, P.  C. (2012), ‘The Mourning Brain: Attachment, Anticipation, and Hamlet’s Unmanly Grief ’, in I.  Jaen and J.  J.  Simon (eds.), Cognitive Literary Studies: Current Themes and New Directions, Austin, TX, 89–104. Hogan, P. C. (2018), Literature and Emotion, London. Holmes, B. A. (2010), The Symptom and the Subject: The Emergence of the Physical Body in Ancient Greece, Princeton, NJ. Holt, P. (1981), ‘The debate scenes in Sophocles’ Ajax’, The American Journal of Philology 102: 275–88. Hotton, S. and Yoshimi, J. (2010), ‘Extending Dynamical Systems Theory to Model Embodied Cognition’, Cognitive Science 35: 444–79. Hubbard, T.  K. (1985), The Pindaric Mind: A Study of Logical Structure in Early Greek Poetry, Leiden. Huitink, L. (2019), ‘Enargeia, Enactivism and the Ancient Readerly Imagination’, in Anderson et al. (2019), 169–89. Hunt, M. (1990), Shakespeare’s Romance of the Word, Lewisburg, PA. Hutchins, E. (1995), Cognition in the Wild, Cambridge, MA. Hutto, D. D. (2008), Folk Psychological Narratives: The Sociocultural Basis of Understanding Reasons, Cambridge, MA. Hutto, D. D. (2011), ‘Understanding Fictional Minds without Theory of Mind!’, Style 45: 276–82, 415. Hutto, D. D. and Myin, E. (2013), Radicalizing Enactivism: Basic Minds Without Content, Cambridge, MA. Iacoboni, M., Woods, R.  P., Brass, M., Bekkering, H., Mazziotta, J.  C., and Rizzolatti, G. (1999), ‘Cortical Mechanisms of Human Imitation’, Science 286: 2526–8. Iser, W. (1972), Der implizite Leser: Kommunikationsformen des Romans von Bunyan bis Beckett, Munich. Iser, W. (1976), Der Akt des Lesens: Theorie ästhetischer Wirkung, Munich. Jackson, T.  E. (2003), ‘ “Literary Interpretation” and Cognitive Literary Studies’, Poetics Today 24: 191–205. Jacobson, D. J. (2011), Show Business: Deixis in Fifth-Century Athenian Drama, PhD diss. Berkeley, CA. Jameson, M. H. (1956), ‘Sophocles and the Philoctetes’, Classical Philology 51: 217–27.

Bibliography  241 Jannidis, F. (2013), ‘Character’, in P. Hühn, J. Pier, W. Schmid, and J. Schönert (eds.), The Living Handbook of Narratology, Hamburg. http://www.lhn.uni-hamburg.de. Jebb, R. C. (1892), Sophocles: The Plays and Fragments. Vol. 5: Trachiniae, Cambridge, repr. Bristol 2004. Jebb, R. C. (1896), Sophocles: The Plays and Fragments. Vol. 7. Ajax, Cambridge. Jebb, R.  C. (1900), Sophocles: The Plays and Fragments. Vol. 3. Antigone, 3rd edn, Cambridge. Johnson, P. J. (1997), ‘Woman’s Third Face: A Psycho/Social Reconsideration of Sophocles’ Antigone’, Arethusa 30: 369–98. Johnston, S. I. (2008), ‘Animating Statues: A Case Study in Ritual,’ Arethusa 41: 445–78. Joly, R. and Byl, S. (1984), Hippocrates: Du régime, Berlin. Jouanna, J. (1997), ‘Notes sur la scène de la reconnaissance dans les Choéphores d’Eschyle (v. 205–211) et sa parodie dans l’Électre d’Euripide (v. 532–537)’, Cahiers du groupe interdisciplinaire du théâtre antique 10: 69–85. Jovchelovitch, S. (forthcoming), ‘The Psychology of Selfhood Now and Then’, in L. Huitink, V. P. Glăveanu, and I. Sluiter (eds.), Social Psychology and the Ancient World. Methods and Applications, Leiden. Kaibel, G. (1896), Elektra, Leipzig. Kamerbeek, J. C. (1948), ‘Sophocle et Héraclite’, in Studia Varia Carolo Guliemo Vollgraff a discipulis oblata, Amsterdam, 84–98. Kamerbeek, J. C. (1963), The Plays of Sophocles. Vol. 1 The Ajax, Leiden. Kamerbeek, J. C. (1970), The Plays of Sophocles. Vol. 2 The Trachiniae, Leiden. Kamerbeek, J. C. (1974), The Plays of Sophocles. Vol. 5 The Electra, Leiden. Kamerbeek, J. C. (1978), The Plays of Sophocles. Vol. 3. The Antigone, Leiden. Kaposi, D. (2020), ‘Saving a Victim from Himself: The Rhetoric of the Learner’s Presence and Absence in the Milgram Experiments’, British Journal of Social Psychology 59: 900–21. Keen, S. (2007), Empathy and the Novel, New York. Kelley, H.  H. (1967), ‘Attribution Theory in Social Psychology’, Nebraska Symposium on Motivation 15: 192–238. Kelley, H.  H. (1972a), ‘Attribution in Social Interaction’, in E.  E.  Jones, D.  E.  Kanouse, H. H. Kelley, R. E. Nisbett, S. Valins, and B. W. Weiner (eds.), Attribution: Perceiving the Causes of Behavior, Morristown, NJ, 1–26. Kelley, H.  H. (1972b), ‘Causal Schemata and the Attribution Process’. in E.  E.  Jones, D. E. Kanouse, H. H. Kelley, R. E. Nisbett, S. Valins, and B. W. Weiner (eds.), Attribution: Perceiving the Causes of Behavior, Morristown, NJ, 151–74. Kells, J. H. (1973), Sophocles: Electra, Cambridge. Kemp, R. and McConachie, B. (eds.) (2018), The Routledge Companion to Theatre, Performance, and Cognitive Science, London. Kermode, F. (ed.) (1954), The Tempest, The Arden Shakespeare, Second series, Cambridge, MA. Kim, K.  K. and Mundy, P. (2012), ‘Joint Attention, Social-Cognition, and Recognition Memory in Adults’, Frontiers in Human Neuroscience 6.172: 1–11. Kindt, J. C. (2010), ‘Olympia, Statues, and the Cultural Memory of Greece’, SSRN Electronic Journal 10.2139/ssrn.1607322. Kirk, A. (2005), ‘Social and Collective Memory’, in A. Kirk and T. Thatcher (eds.), Memory, Tradition, and Text: Uses of the Past in Early Christianity. Atlanta, 1–24. Kirsh, D. (2009), ‘Problem Solving and Situated Cognition’, in P. Robbins and M. Aydede (eds.), The Cambridge Handbook of Situated Cognition, Cambridge, 264–306. Kitto, H.  D.  F. (1956), Form and Meaning in Drama: A Study of Six Greek Plays and of Hamlet, London.

242 Bibliography Kitto, H. D. F. (1961), Greek Tragedy, London. Knox, B. (1961), ‘The Ajax of Sophocles’, Harvard Studies of Classical Philology 65: 1–37. Knox, B. (1979a), ‘Euripidean Comedy’, in B. Knox, Word and Action: Essays on the Ancient Theater, Baltimore, MD, 250–74. (Originally published in A. Cheuse and R. Koffler (eds.), The Rarer Action: Essays in Honor of Francis Fergusson, New Brunswick, NJ, 1970, 68–96.) Knox, B. (1979b), ‘The Ajax of Sophocles’, in B.  Knox, Word and Action: Essays on the Ancient Theater, Baltimore, MD, 125–60. Knox, B. (1979c), ‘Second Thoughts in Greek Tragedy’, in B. Knox, Word and Action. Essays on the Ancient Theater, Baltimore, MD, 231–49. Kocur, M. (2001), Teatr Antycznej Grecji, Wrocław. Konstan, D. (1999), ‘The Tragic Emotions’, Comparative Drama 33: 1–21. Konstan, D. (2001), Pity Transformed, London. Krull, D.  S., Loy, M.  H.-M., Lin, J., Wang, C.-F., Chen, S., and Zhao, X. (1999), ‘The Fundamental Attribution Error: Correspondtermence Bias in Individualist and Collectivist Cultures’, Personality and Social Psychology Bulletin 25: 1208–19. Kukkonen, K. (2014a), ‘Presence and Prediction: The Embodied Reader’s Cascades of Cognition’, Style 48: 367–84. Kukkonen, K. (2014b), ‘Bayesian Narrative: Probability, Plot, and the Shape of the Fictional orld’, Anglia: Zeitschrift für Englische Philologie 132: 720–39. Kukkonen, K. (2019), 4E Cognition and Eighteenth-Century Fiction: How the Novel Found Its Feet, New York. Kukkonen, K. (2020), Probability Designs: Literature and Predictive Processing, Oxford. Kukkonen, K. and Caracciolo, M. (eds.) (2014a), Cognitive Literary Study: Second Generation Approaches, special issue of Poetics Today 48.3. Kukkonen, K. and Caracciolo, M. (2014b), ‘Introduction: What is the “Second Generation?” ’, Style 48: 261–74. Kukkonen, K. and Nielsen, H.  S. (2018), ‘Fictionality. Cognition and Exceptionality’, Poetics Today 39: 473–94. Kuzmičová, A. (2012), ‘Presence in the Reading of Literary Narrative: A Case for Motor Enactment’, Semiotica 189: 23–48. Kuzmičová, A. (2013), Mental Imagery in the Experience of Literary Narrative: Views from Embodied Cognition, Stockholm. Kyriakou, P. (2012), ‘Philoctetes’, in A. Markantonatos (ed.), Brill’s Companion to Sophocles, Leiden-Boston, 149–66. Labov, W. (2004), ‘The Boundaries of Words and their Meanings’, in B. Aarts, D. Denison, E. Keizer, and G. Popova (eds.), Fuzzy Grammar. A Reader, Oxford, 67–89. Lada, I. (1994), ‘ “Empathic Understanding”: Emotion and Cognition in Classical Dramatic Audience-Response’, Proceedings of the Cambridge Philological Society 39: 94–140. Lada-Richards, I. (2002), ‘The Subjectivity of Greek Performance’, in P. Easterling and E. Hall (eds.), Greek and Roman Actors: Aspects of an Ancient Profession, Cambridge, 395–418. Lada-Richards, I. (2009), ‘ “The Players will tell all”: the dramatist, the actors and the art of acting in Sophocles’ Philoctetes’, in Simon Goldhill and Edith Hall (eds.), Sophocles and the Greek Tragic Tradition, Cambridge, 48–68. Lakoff, G. (1987), Women, Fire, and Dangerous Things: What Categories Reveal about the Human Mind, Chicago. IL. Lakoff, G. and Johnson, M. (1980), Metaphors We Live By, Chicago, IL. Lakoff, G. and Turner, M. (1989), More Than Cool Reason: A Field Guide to Poetic Metaphor, Chicago, IL. Laks, A. and Most, G. W. (eds.) (2016), Early Greek Philosophy, 9 vols., Cambridge, MA.

Bibliography  243 Lambert, S. (ed.) (2011), Sociable Man: Essays on Ancient Greek Social Behaviour in honour of Nick Fisher, Swansea. Lardinois, A. (2006), ‘The Polysemy of Gnomic Expressions and Ajax’ Deception Speech’, in I. J. F. de Jong and A. Rijksbaron (eds.), Sophocles and the Greek Language, Leiden, 213–23. Larson, J. (2016), Understanding Greek Religion, New York. Lattimore, R. (1958), The Poetry of Greek Tragedy, Baltimore, MD. Lauwers, J., Opsomer, J., and Schwall, H. (eds.) (2018), Psychology and the Classics: A Dialogue of Disciplines, Berlin. Lavie, N., Beck, D.  M., and Konstantinou, N. (2014), ‘Blinded by the Load: Attention, Awareness and the Role of Perceptual Load’, Philosophical Transactions of the Royal Society 369.1641: 1–10. Lawrence, S. (1978), ‘The Dramatic Epistemology of Sophocles’ Trachiniae’, Phoenix 32: 288–304. Lawrence, S. (2005), ‘Ancient Ethics, the Heroic Code, and the Morality of Sophocles’ Ajax.’ Greece and Rome 52: 18–33. Lebeck, A. (1971), The Oresteia: A Study in Language and Structure, Washington DC. Lefèvre, E. (2001), Die Unfähigkeit, sich zu erkennen: Sophokles’ Tragödien, Mnemosyne Suppl. 227, Leiden. Lesky, A. (1925), Alkestis, der Mythus und das Drama, SBAW Wien, Phil.-Hist. Kl. 203.2, Vienna. Lesky, A. (1966), ‘Decision and Responsibility in Aeschylus’, Journal of Hellenic Studies 85: 42–55. Levinson, S.  C. (2006), ‘On the Human “Interaction Engine” ’, in N.  J.  Enfield and S.  C.  Levinson (eds.), Roots of Human Sociality: Culture, Cognition, and Human Interaction, Oxford, 39–69. Levitin, D. (2006), This is Your Brain on Music: The Science of a Human Obsession, London. Liebert, R. S. (2013), ‘Pity and Disgust in Plato’s Republic: The Case of Leontius’, Classical Philology 108: 179–201. Liebert, R. S. (2017), Tragic Pleasure from Homer to Plato, Cambridge. List, G. (1961), ‘Speech Melody and Song Melody in Central Thailand’, Ethnomusicology 5: 16–32. Lively, A. (2016), ‘Joint Attention, Semiotic Mediation, and Literary Narrative’, Poetics Today 37: 517–38. Lloyd, G. E. R. (2007), Cognitive Variations: Reflections on the Unity and Diversity of the Human Mind, Oxford. Lloyd, M. (2005), Sophocles: Electra, London. Lloyd, M. (2020), ‘Realism in Euripides’, in A. Markantonatos (ed.), Brill’s Companion to Euripides, vol. 1, Leiden, 605–26. Lloyd-Jones, H. and Wilson N. G. (eds.) (1990), Sophoclis Fabulae, Oxford. Lowe, N. J. (2000), The Classical Plot and the Invention of Western Narrative, Cambridge. Lyne, R. (2011), Shakespeare, Rhetoric and Cognition, Cambridge. Lyne, R. (2014), ‘Shakespeare, Perception and Theory of Mind’, Paragraphus 37: 79–95. Lyne, R. (2016), Memory and Intertextuality in Renaissance Literature, Cambridge. Macé, M. (2013), ‘Ways of Reading, Modes of Being’, New Literary History 44: 213–29. MacLeod, L. (2001), Dolos and dike in Sophokles’ Elektra, Leiden. Maehler, H. (1963), Die Auffassung des Dichterberufs im frühen Griechentum bis zur Zeit Pindars, Göttingen. Mäkelä, M. (2013), ‘Cycles of Narrative Necessity: Suspect Tellers and the Textuality of Fictional Minds’, in L. Bernaerts, D. de Geest, L. Herman, and B. Vervaeck (eds.), Stories and Minds: Cognitive Approaches to Literary Narrative, Lincoln, NB, 129–51.

244 Bibliography Malafouris, L. (2004), ‘The Cognitive Basis of Material Engagement: Where Brain, Body, and Culture Conflate’, in E.  Demarrais, C.  Gosden, and C.  Renfrew (eds.), Rethinking Materiality: The Engagement of Mind with the Material World, Cambridge, 53–62. Malafouris, L. (2013), How Things Shape the Mind: A Theory of Material Engagement, Cambridge, MA. Malafouris, L. and Renfrew, C. (2010), The Cognitive Life of Things: Recasting the Boundaries of the Mind, Oakville, CT. Malle, B.  F. (2004), How the Mind Explains Behavior: Folk Explanations, Meaning and Social Interaction, Cambridge, MA. Maner, J.  K., Dewall, C.  N., Baumeister, R.  F., and Schaller, M. (2007), ‘Does Social Exclusion Motivate Interpersonal Reconnection? Resolving the “Porcupine Problem” ’, Journal of Personality and Social Psychology 92: 42–55. Manieri, A. (1998), L’immagine poetica nella teoria degli antichi: Phantasia ed enargeia, Pisa. March, J. R. (1993), ‘Sophocles’ Ajax: The Death and the Burial of a Hero’, Bulletin of the Institute of Classical Studies 38: 1–36. March, J. R. (2001), Sophocles: Electra, Warminster. Markowitsch, H. J. (1994), ‘The Memory Storehouse’, Trends in Neurosciences 17.12: 513–14. Marshall, C. W. (2006), ‘How to write a messenger speech (Sophocles, Electra 680–763)’, Bulletin of the Institute of Classical Studies 49: 203–21. Martin, G., Iurescia, F., Hof, S., and Sorrentino, G. (eds.) (2020), Pragmatic Approaches to Drama: Studies in Communication on the Ancient Stage, Leiden. Martin, W. (1986), Recent Theories of Narrative, Ithaca, NY. Mastronarde, D. J. (1999–2000), ‘Euripidean Tragedy and Genre: The Terminology and its Problems’, in D. Sansone, M. J. Cropp, and K. Lee (eds.), Euripides and Tragic Theatre in the Late Fifth Century, special issue of Illinois Classical Studies 24/25: 23–39. Mastronarde, D. J. ed., (2002), Euripides: Medea, Cambridge. Mastronarde, D.  J. (2010), The Art of Euripides: Dramatic Technique and Social Context, Cambridge. Matzner, S. (2016), Rethinking Metonymy: Literary Theory and Poetic Practice from Pindar to Jakobson, Oxford. Mauduit, C. (2015), ‘L’aparté existe-t-il dans le théâtre grec du Ve siècle?’, in P. Paré-Rey (ed.), L’Aparté dans le théâtre antique:un procédé dramatique à redécouvrir, Paris, 63–99. Maynard, R (1994), ‘Seeing Double’, The Journal of Aesthetics and Art Criticism 52: 155–67. McCall, M. (1990), ‘The Chorus of Aeschylus’ Choephori’, in M.  G.  Griffith and D.  J.  Mastronarde (eds.), Cabinet of the Muses: Essays on Classical and Comparative Literature in Honor of Thomas G. Rosenmeyer, Atlanta, GA, 17–30. McCauley, R. N. (1988), ‘Walking in Our Own Footsteps: Autobiographical Memory and Reconstruction’, in U.  Neisser and E.  Winograd (eds.), Remembering Reconsidered: Ecological and Traditional Approaches to the Study of Memory, Cambridge, 126–44. McConachie, B. (2008), Engaging Audiences: A Cognitive Approach to Spectating in the Theatre, New York. McConachie, B. (2011), ‘An Evolutionary Perspective on Play, Performance, and Ritual’, The Drama Review 55.4: 33–50. McConachie, B. (2013), Theatre and Mind, Basingstoke. McConachie, B. and Hart, F. E. (eds.) (2006), Performance and Cognition: Theatre Studies and the Cognitive Turn, London. McGavin, J.  J. and Walker, G. (2016), Imagining Spectatorship: From the Mysteries to the Shakespearean Stage, Oxford. McHale, B. (2012), ‘Transparent Minds Revisited’, Narrative 20: 115–24.

Bibliography  245 McLure, L. (2015), ‘Tokens of Identity: Gender and Recognition in Greek Tragedy’, Illinois Classical Studies 40: 219–36. Meineck, P. (2011), ‘The Neuroscience of the Tragic Mask’, Arion 19.1: 113–58. Meineck, P. (2018a), ‘The Affective Ancient Theatre, a Bio-Cultural Cognitive Approach’, in J. Lauwers, H. Schwall, and J. Opsomer (eds.), Psychology and the Classics: A Dialogue of Disciplines, Leiden, 77–93. Meineck, P. (2018b), Theatrocracy. Greek Drama, Cognition, and the Imperative for Theatre, Abingdon. Meineck, P. (2019), ‘Mask as Mind Tool.:A Methodology for Material Engagement’, in Anderson et al. (2019), 71–91. Meineck, P., Short, W., and Devereaux, J. (2019a), ‘Introduction’, in Meineck et al. (2019b), 1–18. Meineck, P., Short, W., and Devereaux, J. (eds.) (2019b), The Routledge Handbook of Classics and Cognitive Theory, London. Menary, R. (2010), ‘Introduction to the Special Issue on 4E Cognition’, Phenomenology and the Cognitive Sciences 9: 459–63. Meyer, L. (1956), Emotion and Meaning in Music, Chicago, IL. Milgram, S. (1963), ‘Behavioral Study of Obedience’, Journal of Abnormal and Social Psychology 67: 371–8. Milgram, S. (1974), Obedience to Authority: An Experimental View, London. Miller, A.  G. (2009), ‘Reflections on “Replicating Milgram” (Burger 2009)’, American Psychologist 64: 20–7. Miller, J. G. (1984), ‘Culture and the development of everyday social explanation’, Journal of Personality and Social Psychology 4: 961–78. Minchin, E. (2011), ‘ “Themes” and “Mental Moulds”: Roger Schank, Malcolm Willcock and the Creation of Character in Homer’, Classical Quarterly 61: 323–43. Minchin, E. (2019), ‘The Cognition of Deception: Falsehoods in Homer’s Odyssey and their Audiences’, in P. Meineck et al. (2019b), 109–21. Miyamoto, Y. and Kitayama, S. (2002), ‘Cultural Variation in Correspondence Bias. The Critical Role of Attitude Diagnosticity of Socially Constrained Behavior’, Journal of Personality and Social Psychology 83: 1239–48. Moi, T. (2017), Revolution of the Ordinary. Literary Studies after Wittgenstein, Austin, and Cavell, Chicago, IL. Moorhouse, A. C. (1982), The Syntax of Sophocles, Leiden. Morgan, B. (2017), ‘Situated Cognition and the Study of Culture: an Introduction’, Poetics Today 38: 213–33. Morgan, B., Park, S., and Spolsky, E. (eds.) (2017), Situated Cognition and the Study of Culture, special issue of Poetics Today 38.2. Mossman, J. (2011), Euripides: Medea, Oxford. Most, G.  W. (1985), The Measures of Praise: Structure and Function in Pindar’s Second Pythian and Seventh Nemean Odes, Göttingen. Mueller, M. (2016), Objects as Actors: Props and the Poetics of Performance in Greek Tragedy, Chicago, IL. Müller, G. (1967), Sophokles: Antigone, Heidelberg. Munteanu, D.  L. (2012), Tragic Pathos: Pity and Fear in Greek Philosophy and Tragedy, Cambridge. Murnaghan, S. (1989), ‘Trials of the hero in Sophocles’ Ajax’, in M.  Mackenzie and C.  Roueché (eds.), Images of Authority. . . Proceedings of the Cambridge Philological Society Suppl. 16, Cambridge, 171–93.

246 Bibliography Neisser, U. (1989), ‘Domains of Memory’, in P. R. Solomon, G. R. Goethals, C. M. Kelley, and B. R. Stephens (eds.), Memory: Interdisciplinary Approaches, New York, 67–83. Neisser, U. (2009), ‘Memory With a Grain of Salt’, in H. H. Wood and A. S. Byatt (eds.), Memory: An Anthology, London, 80–8. Nestle, W. (1940), Vom Mythos zum Logos, Stuttgart. Nightingale, A. (1995), Genres in Dialogue: Plato and the Construct of Philosophy, Cambridge. Noel, A.-S. (2013), ‘Le vêtement-piège et les Atrides: métamorphoses d’un objet protéen’, in B. Le Guen and S. Milanezi (eds.), L’appareil scénique dans les spectacles de l’Antiquité, Saint-Denis, 161–82. Noel, A.-S. (2016), ‘Love Is In The Hands: Affective Relationships With Objects in Votive Dedications’, Center for Hellenic Studies Research Bulletin 4.2, November. Noel, A.-S. (2018), ‘Prosthetic Imagination in Greek Literature’, in J.  Draycott (ed.), Prostheses in Antiquity, London, 159–79. Noel, A.-S. (2019a), ‘Hybris and Hybridity in Aeschylus’ Persians: A Posthumanist Perspective on Xerxes’ Expedition’, in G. M. Chesi and F. Spiegel (eds.) 2019, 259–65. Noel, A.-S. (2019b), ‘What Do We Actually See on Stage? A Cognitive Approach to the Interaction between Visual and Aural Effects in the Performance of Greek Tragedy’, in Meineck et al. (2019b), 297–309. Nooter, S. (2012), When Heroes Sing: Sophocles and the Shifting Soundscape of Tragedy, Cambridge. Nordgren, L. (2015), Greek Interjections: Syntax, Semantics and Pragmatics, Berlin. Norman, D. (1991), ‘Cognitive Artifacts’, in J.  M.  Carroll (ed.), Designing Interaction: Psychology at the Human-Computer Interface, Cambridge, 17–37. Nünning, V. (2014), Reading Fictions, Changing Minds: The Cognitive Value of Fiction, Schriften des Marsiliuskollegs, vol. 11, Heidelberg. O’Connell, P. (2017), The Rhetoric of Seeing in Attic Forensic Oratory. Austin. O’Neill, Eugene, Jr. (trans.), (1938), Aristophanes. Ecclesiazusae. The Complete Greek Drama, vol. 2, New York. Ormand, K. (1999), Exchange and the Maiden: Marriage in Sophoclean Tragedy, Austin, TX. Orne, M.  T. (1962), ‘On the Social Psychology of the Psychological Experiment: With Particular reference to Demand Characteristics and their implications’, American Psychologist 17: 776–83. Ostenfeld, E.  N. (2018), Ancient Greek Psychology and the Modern Mind-Body Debate, 2nd edn, Baden-Baden. Ostrom, T.  M. (1989), ‘Three Catechisms for Social Memory’, in P.  R.  Solomon, G.  R.  Goethals, C.  M.  Kelley, and B.  R.  Stephens (eds.), Memory: Interdisciplinary Approaches. New York, 201–220. Otto, N. (2009), Enargeia: Untersuchung zur Charakteristik alexandrinischer Dichtung, Stuttgart. Owens, D. (1999), ‘The Authority of Memory’, European Journal of Philosophy 7: 312–29. Padilla, M. (1992), ‘The Gorgonic Archer’, Classical World 86: 1–12. Pagán Cánovas, C. (2010), ‘Conceptual Blending Theory and the History of Emotions’, in R.  Sun (ed.), Proceedings of the Workshop on Cognitive Social Sciences: Grounding the Social Sciences in Cognition?, New York. Pagán Cánovas, C. (2011), ‘The Genesis of the Arrows of Love: Diachronic Conceptual Integration in Greek Mythology’, American Journal of Philology 132: 553–79. Page, D. L. (1972), Aeschyli. septem quae supersunt tragoediae, Oxford. Palmer, A. (2004), Fictional Minds, Lincoln, NB.

Bibliography  247 Palmer, A. (2007), ‘Attribution Theory: Action and Emotion in Dickens and Pynchon’, in M. Lambrou and P. Stockwell (eds.), Contemporary Stylistics, London, 81–92. Palmer, A. (2010), Social Minds in the Novel, Columbus, OH. Papadimetropoulos, L. (2008), ‘Heracles as tragic hero’, Classical World 101: 131–8. Papadopoulou, T. (2011), Aeschylus: Suppliants, Bristol. Paré-Rey, P. (2015), L’Aparté dans le théâtre antique : un procédé dramatique à redécouvrir, Saint-Denis. Parlavantza-Friedrich, U. (1969), Täuschungsszenen in den Tragödien des Sophokles, Berlin. Pelling, C. B. R. (ed.) (1990), Characterization and Individuality in Greek Literature, Oxford. Peponi, A. E. (2015), ‘Dance and Aesthetic Perception’, in P. Destrée and P. Murray (eds.), A Companion to Ancient Aesthetics, Chichester, 204–17. Perry, G. (2013), Behind the Shock Machine: the Untold Story of the Notorious Milgram Psychology Experiments, Brunswick, CA. Pfeijffer, I.  L. (1999), Three Aeginetan Odes of Pindar: A Commentary on Nemean V, Nemean III and Pythian VIII, Leiden. Phelan, J. (1989), Reading People, Reading Plots: Character, Progression, and the Interpretation of Narrative, Chicago, IL. Phelan, J. (2007), Experiencing Fiction: Judgments, Progressions, and the Rhetorical Theory of Narrative, Columbus, OH. Poe, J. P. (1987), Genre and Meaning in Sophocles’ Ajax, Frankfurt am Main. Pohlenz, M. (1965), ‘Die Anfänge der griechischen Poetik’, in Kleine Schriften, Hildesheim, 436–72. Originally published in NGG, Phil.-Hist. (1920): 142–78. Pollard, T. (2015), ‘Tragicomedy’, in P. Cheney and P. Hardie (eds.), The Oxford History of Classical Reception in English Literature, vol. 2. 1558–1660, Oxford, 419–32. Pollard, T. (2017), Greek Tragic Women on Shakespearean Stages, Oxford. Pollard-Gott, L. (1993), ‘Attribution Theory and the Novel’, Poetics 21: 499–524. Polvinen, M. (2013), ‘Affect and Artifice in Cognitive Literary Theory’, Journal of Literary Semantics 42.2: 165–80. Polvinen, M. and Sklar, H. (2019), ‘Mimetic and Synthetic Views of Characters: How Readers Process “People” in Fiction’, Cogent Arts and Humanities 6.1: 1687257. Prauscello, L. (2010), ‘The Language of Pity: Eleos and Oiktos in Sophocles’ Philoctetes’, The Cambridge Classical Journal 56: 199–212. Prentice, D.  A. and Gerrig, R.  J. (1999), ‘Exploring the Boundary between Fiction and Reality’, in S. Chaiken and Y. Trope (eds.), Dual-Process Theories in Social Psychology, New York, 529–46. Radt, S. L. (1971), ‘Sophokles, Antigone 45f ’, Mnemosyne 24: 293–5. Read, S. (2013), ‘Shakespeare and the Arts of Cognition’, in J.  Post (ed.), The Oxford Handbook of Shakespeare’s Poetry, Oxford, 62–76. Rehm, R. (1994), Marriage to Death: The Conflation of Wedding and Funeral Rituals in Greek Tragedy, Princeton, NJ. Rehm, R. (2002), The Play of Space: Spatial Transformation in Greek Tragedy, Princeton, NJ. Reinhardt, K. (1947), Sophokles, 3rd edn, Frankfurt am Main. Reinhardt, K. (1960), ‘Die Sinneskrise bei Orest’, in C. Becker (ed.), Tradition und Geist: Gesammelte Essays zur Dichtung, Göttingen, 227–56. Originally in Die Neue Rundschau 68 (1957), 615–46. Revermann, M. (2006), ‘The Competence of Theatre Audiences in Fifth- and FourthCentury Athens’, The Journal of Hellenic Studies 126: 99–124. Richardson, A. (2004), ‘Studies in Literature and Cognition: A Field Map’, in E.  Spolsky and A.  Richardson, (eds.), The Work of Fiction: Cognition, Culture, and Complexity, Aldershot, 1–29.

248 Bibliography Richardson, A. and Steen, F.  F. (2002), ‘Literature and the Cognitive Revolution: An Introduction’, Poetics Today 23: 1–8. Richardson, B., Skov Nielsen, H., and Alber, J. (eds.) (2013), A Poetics of Unnatural Narrative, Columbus, OH. Ringer, M. (1998), Electra and the Empty Urn: Metatheater and Role Playing in Sophocles, Chapel Hill, NB. Ritchie, L. S. (2013), Metaphor, Cambridge. Roisman, H. M. (1997), ‘The Appropriation of a Son: Sophocles’ Philoctetes’, Greek, Roman, and Byzantine Studies 38: 127–71. Ronnet, G. (1975), ‘L’ironie d’Euripide dans Électre, vers 513 à 546’, Revue des Études Grecques 88: 63–70. Rosch, E. (1975), ‘Cognitive Representation of Semantic Categories’, Journal of Experimental Psychology 104: 192–233. Rosch, E. (1978), ‘Principles of Categorization’, in E. Rosch and B. Lloyd (eds.), Principles of Categorization, Hillsdale, NJ, 27–48. Reprinted in Aarts, B., D. Denison, E. Keizer, and G. Popova (eds.), Fuzzy Grammar. A Reader. Oxford, 2004, 91–108. Rosch, E. (1983), ‘Prototype Classification and Logical Classification: The Two Systems’, in E. Scholnick (ed.), New Trends in Cognitive Representation: Challenges to Piaget’s Theory, Hillsdale, NJ, 73–86. Rose, P.  W. (1976), ‘Sophocles’ Philoctetes and the Teachings of the Sophists’, Harvard Studies in Classical Philology 80: 49–105. Rose, P.  W. (1992), Sons of the Gods, Children of Earth: Ideology and Literary Form in Ancient Greece, Ithaca, NY, and London. Roselli, A. (1982), ‘Livelli del conoscere,’ Materiali e Discussioni per l’Analisi Dei Testi Classici 7: 9–38. Rosenmeyer, T.  G. (1955), ‘Gorgias, Aeschylus, and apate’, The American Journal of Philology 76: 225–60. Rosenmeyer, T. G. (1982), The Art of Aeschylus, Berkeley, CA. Rösler, W. (1980), ‘Die Entdeckung der Fiktionalität in der Antike’, Poetica 12: 283–319. Rotstein, A. (2010), The Idea of Iambos, Oxford. Roux, G. (1974), ‘Électre avait-elle de grands pieds? (Eschyle, Choéphores, v.164 sq. et Euripide, Électre, v.532 sq.)’, Revue des Études Grecques 87: 42–56. Roux, J. (1970), Les Bacchantes II: Commentaire, Paris. Ruffell, I.  A. (2008), ‘Audience and Emotion in the Reception of Greek Drama’, in M.  Revermann and P.  Wilson (eds.), Performance, Iconography, Reception: Studies in Honour of Oliver Taplin, Oxford, 37–58. Rutherford, I. (2013), State Pilgrims and Sacred Observers in Ancient Greece: A Study of Theōriā and Theōroi, Cambridge. Rutherford, R. B. (2012), Greek Tragic Style: Form, Language, and Interpretation, Cambridge. Ryan, M.-L. (1999), ‘Immersion vs. Interactivity: Virtual Reality and Literary Theory’, SubStance 28.2.89: 110–37. Ryan, M.-L. (2001), Narrative as Virtual Reality: Immersion and Interactivity in Literature and Electronic Media, Baltimore, MD. Ryan, M.-L. (2015), Narrative as Virtual Reality 2: Revisiting Immersion and Interactivity in Literature and Electronic Media, Baltimore, MD. Saltz, D. Z. (2006), ‘Infiction and Outfiction: The Role of Fiction in Theatrical Performance’, in D.  Krasner and D.  Z.  Saltz (eds.), Staging Philosophy: Intersections of Theater, Performance, and Philosophy, Ann Arbor, MI, 203–20.

Bibliography  249 Samson, D. (2013), ‘Theory of Mind’, in D.  Reisberg (ed.), The Oxford Handbook of Cognitive Pyschology, Oxford, 943–56. Sansone, D. (1975), Aeschylean Metaphors for Intellectual Activity, Wiesbaden. Schacter, D. L. (1995), ‘Memory Distortion: History and Current Status’, in D. L. Schacter (ed.), Memory Distortion: How Minds, Brains, and Societies Reconstruct the Past, Cambridge, MA, 1–43. Schadewaldt, W. (1926), Monolog und Selbstgespräch: Untersuchungen zur Formgeschichte der griechischen Tragödie, Berlin. Schadewaldt, W. (1955), ‘Furcht und Mitleid? Zur Deutung des aristotelischen Tragödiensatzes’, Hermes 83: 129–71. Schaeffer, J.-M. (1999), Pourquoi la fiction?, Paris. Schaper, E. (1968), ‘Aristotle’s Catharsis and Aesthetic Pleasure’, The Philosophical Quarterly 18.71: 131–43. Schein, S. L. (1975), ‘Mythical Illusion and Historical Reality in Euripides’ Orestes’, Wiener Studien NF 9: 49–66. Schein, S.  L. (1982), ‘Electra: A Sophoclean problem play’, Antike und Abendland 28: 69–80. Schein, S. L. (2012), ‘Sophocles and Homer’, in K. Ormand (ed.), A Companion to Sophocles, Oxford, 424–39. Schein, S. L. (2013), Sophocles, Philoctetes, Cambridge. Schein, S. L. (2016), Homeric Epic and its Reception: Interpretive Essays, Oxford. Schneider, R. (2001), ‘Toward a Cognitive Theory of Literary Character: The Dynamics of Mental-Model Construction’, Style 25: 607–40. Schollmeyer, J. (2021), Gorgias’ ‘Lobrede auf Helena’: Literaturgeschichtliche Untersuchungen und Kommentar, Berlin. Schudson, M. (1995), ‘Dynamics of Distortion in Collective Memory’, in D.  L.  Schacter (ed.), Memory Distortion: How Minds, Brains, and Societies Reconstruct the Past, Cambridge, MA, 346–64. Schwenger, P. (2006), The Tears of Things: Melancholy and Physical Objects, Minneapolis, MN. Scodel, R. (1984), Sophocles, Boston, MA. Scodel, R. (2009), ‘Ignorant Narrators in Greek Tragedy’, in J. Grethlein and A. Rengakos (eds.), Narratology and Interpretation: The Content of Narrative Form in Ancient Literature, Berlin, 421–47. Scodel, R. (2012), ‘Ἦ and Theory of Mind in the Iliad’, in M. Meier-Brügger (ed.) Homer, gedeutet durch ein großes Lexikon, Berlin, 319–34. Scodel, R. (2015), ‘Sunt lacrimae rerum’, The Classical Journal 111: 219–30. Seaford, R. (1987), ‘Pentheus’ Vision: Bacchae 918–22’, Classical Quarterly 37: 76–8. Seaford, R. (1994a), Reciprocity and Ritual: Homer and Tragedy in the Developing CityState, Oxford. Seaford, R. (1994b), ‘Sophokles and the Mysteries’, Hermes 122: 275–88. Seale, D. (1982), Vision and Stagecraft in Sophocles, London. Searle, J. R. (1969), Speech Acts, An Essay in the Philosophy of Language, Cambridge. Searle, J. R. (1976), ‘A Classification of Illocutionary Acts’, Language in Society 5.1: 1–23. Searleman, A. and Herrmann, D. (1994), Memory from a Broader Perspective, New York. Sedgwick, E. K. (2003), Touching Feeling: Affect, Pedagogy, Performativity, Durham, NC. Seeman, A. (2011), Joint Attention: New Developments in Psychology, Philosophy of Mind, and Social Neuroscience, Cambridge, MA.

250 Bibliography Seeman, A. (2016), ‘Reminiscing Together: Joint Experiences, Epistemic Groups, and Sense of Self ’, Synthese 196: 1–16. Segal, C. P. (1962), ‘Gorgias and the Psychology of the Logos’, Harvard Studies in Classical Philology 66: 99–155. Segal, C. P. (1966), ‘The Electra of Sophocles’, Transactions and Proceedings of the American Philological Association 97: 473–545. Segal, C. P. (1981), Tragedy and Civilization: An Interpretation of Sophocles, Cambridge, MA. Segal, C. P. (1997), Dionysiac Poetics and Euripides’ Bacchae, Princeton, NJ. Seghier, M. L. and Price, C. J. (2018), ‘Interpreting and Utilising Intersubject Variability in Brain Function’, Trends in Cognitive Science 22: 517–30. Segre, C. (1980), ‘A Contribution to the Semiotics of Theater’, Poetics Today 1.3: 39–48. Seidensticker, B. (2008), ‘Character and Characterization in Greek Tragedy’, in M.  Revermann and P.  Wilson (eds.), Performance, Iconography, Reception. Studies in Honour of Oliver Taplin, Oxford, 333–46. Seidensticker, B. (2009), ‘Charakter und Charakterisierung bei Aischylos’, in J. Jouanna and F. Montanari (eds.), Eschyle à l’aube du théâtre occidental, Geneva, 205–56. Semino, E. and Steen, G. (2008), ‘Metaphor in Literature’, in Gibbs 2008, 232–46. Sewell-Rutter, N.  J. (2007), Guilt by Descent: Moral Inheritance and Decision Making in Greek Tragedy, Oxford. Shapiro, L. and Spaulding, S. (2021), ‘Embodied Cognition’, in The Stanford Encyclopedia of Philosophy, available at https://plato.stanford.edu/archives/win2021/entries/embodiedcognition/ last accessed 13 July, 2022. Sharrock, A. (2018), ‘How Do We Read a (W)hole? Dubious First Thoughts About the Cognitive Turn’, in S.  Harrison, S.  A.  Frangoulidis, and T.  D.  Papanghelis (eds.), Intratextuality and Latin Literature, Berlin, 15–31. Shaw, J. and Porter, S. (2015), ‘Constructing Rich False Memories of Committing a Crime’, Psychological Science 26: 291–301. Shrout, P. E. and Rodgers, J. L. (2018), ‘Psychology, Science, and Knowledge Construction: Broadening Perspectives from the Replication Crisis’, Annual Review of Psychology 69: 487–510. Shteynberg, G. (2015), ‘Shared Attention’, Perspectives on Psychological Science 10: 579–90. Shteynberg, G., Hirsh, J.  B., Roese, N.  J., Larsen, J.  T., Galinsky, A.  D., and Apfelbaum, E. P. (2014), ‘Feeling More Together: Group Attention Intensifies Emotion’, Emotion 14: 1102–14. Sier, K. (2000), ‘Gorgias über die Fiktionalität der Tragödie’, in E. Stärk and G. Vogt-Spira (eds.), Dramatische Wäldchen: Festschrift für E. Lefèvre, Hildesheim, 575–618. Silk, M. S. (1974), Interaction in Poetic Imagery, Cambridge. Sinding, M. (2002), ‘After Definitions: Genre, Categories, and Cognitive Science’, Genre 35.2: 181–220. Singer, J.  A. and Salovey, P. (1993), The Remembered Self: Emotion and Memory in Personality, New York. Slater, N.  W. (2002), Spectator Politics: Metatheatre and Performance in Aristophanes, Philadelphia, PA. Slings, S. R. (2003), Platonis Respublica, Oxford. Sluiter, I. (2008), ‘General Introduction’, in R. M. Rosen and I. Sluiter (eds.), Kakos: Badness and Anti-Value in Classical Antiquity, Leiden, 1–27. Sluiter, I. (2017), ‘Anchoring Innovation: a Classics Research Agenda’, European Review 25: 20–38. Sluiter, I., Corthals, B., van Duijn, M. J., and Verheij, M. (2013), ‘In het hoofd van Medea: Gedachtenlezen bij een moordende moeder’, Lampas 46: 3–20.

Bibliography  251 Smith, J. (2015), ‘Filmmakers as Folk Psychologists: How Filmmakers Exploit Cognitive Biases as an Aspect of Cinematic Narration, Characterization, and Spectatorship’, in L.  Zunshine (ed.), The Oxford Handbook of Cognitive Literary Studies, New York, 483–504. Sommerstein, A. H. (1994), Aristophanes: Thesmophoriazusae, Warminster. Sommerstein, A.  H. (2008), Aeschylus: Persians, Seven against Thebes, Suppliants, Prometheus Bound, Cambridge, MA. Sommerstein, A. H. (2010), Aeschylean tragedy, 2nd edn, London. Sommerstein, A. H. (2019), Aeschylus: Suppliants, Cambridge. Sourvinou-Inwood, C. (1989), ‘Assumptions and the Creation of Meaning: Reading Sophocles’ Antigone’, Journal of Hellenic Studies 109: 134–48. Sperber, D. and Wilson, D. (1995), Relevance: Communication and Cognition, 2nd edn, Oxford (first edition 1986). Sperber, D. and Wilson, D. (2008), ‘A Deflationary Account of Metaphors’, in Gibbs 2008, 84–108. Spolsky, E. (1993), Gaps in Nature: Literary Interpretations and the Modular Mind, Albany, NY. Spolsky, E. (2001), Satisfying Skepticism: Embodied Knowledge in the Early Modern World, Aldershot. Spolsky, E. (2002), ‘Darwin and Derrida: Cognitive Literary Theory as a Species of PostStructuralism’, Poetics Today 23: 43–62. Spolsky, E. (2015), The Contracts of Fiction: Cognition, Culture, Community, Oxford. Squire, M. (2016), ‘Introductory Reflections: Making Sense of Ancient Sight’, in M. Squire (ed.), Sight and the Ancient Senses, London, 1–35. Stanford, W. B. (1963), Sophocles: Ajax, London. Stanford, W. B. (2014), Greek Tragedy and the Emotions: An Introductory Study, Abingdon. Originally published 1983. States, B. O. (1985), Great Reckonings in Little Rooms: On the Phenomenology of Theater, Berkeley, CA. Steiner, D. (2001), Images in Mind: Statues in Archaic and Classical Greek Literature and Thought, Princeton, NJ. Stendhal (1928), Racine et Shakespeare, Paris. Stenger, J. (2004), Poetische Argumentation: Die Funktion der Gnomik in den Epinikien des Bakchylides, Berlin. Stockwell, P. J. (2006), ‘On Cognitive Poetics and Stylistics’, in H. Veivo, B. Petterson, and M. Polvinen (eds.), Cognition and Literary Interpretation in Practice, Helsinki, 267–82. Stopel, B. (ed.) (2017), From Mind to Text: Continuities and Breaks between Cognitive, Aesthetic and Textualist Approaches to Literature, New York. Sweetser, E. (1990), From Etymology to Pragmatics: Metaphorical and Cultural Aspects of Semantic Structure, Cambridge. Tal-Or, N. and Cohen, J. (2010), ‘Understanding Audience Involvement: Conceptualizing and Manipulating Identification and Transportation’, Poetics 38: 402–18. Tamir, D. I., Bricker, A. B., Dodell-Feder, D., and Mitchell, J. P. (2016), ‘Reading Fiction and Reading Minds: The Role of Simulation in the Default Network’, Social Cognitive and Affective Neuroscience 11.2: 215–24. Taplin, O. (1972), ‘Aeschylean Silences and Silences in Aeschylus’, Harvard Studies in Classical Philology 76: 57–97. Taplin, O. (1977), The Stagecraft of Aeschylus: The Dramatic use of Exits and Entrances in Greek Tragedy, Oxford. Taplin, O. (1978), Greek Tragedy in Action, London.

252 Bibliography Taplin, O. (1979), ‘Yielding to Forethought: Sophocles’ Ajax’, in G.  W.  Bowersock, W.  Burkert, and M.  C.  J.  Putnam (eds.), Arktouros: Hellenic Studies Presented to Bernard M. W. Knox on the Occasion of his 65th Birthday, Berlin, 122–9. Taplin, O. (1986), ‘Fifth-Century Tragedy and Comedy: A Synkrisis’, The Journal of Hellenic Studies 106: 163–74. Taplin, O. (2016), ‘Aeschylus, “Father of Stage-Objects” ’, in A.  Coppola, C.  Barone, M. Salvadori (eds.), Gli oggetti sulla scena teatrale ateniese: funzione, rappresentazione, comunicazione, Padova, 145–54. Tarkow, T.  A. (1970), ‘The dilemma of Pelasgus and the nautical imagery of Aeschylus’ Suppliants’, Classica and Mediaevalia 31: 1–13. Taylor, J. R. (2011), ‘Prototype theory’, in C. Maienborn, K. von Heusinger, and P. Portner (eds.), Semantics: An International Handbook of Natural Language Meaning, Berlin, 643–64. Telò, M. and Mueller, M. (2018), The Materialities of Greek Tragedy: Object and Affect in Aeschylus, Sophocles and Euripides, London. Tendahl, M. and Gibbs, R.  W. (2008), ‘Complementary Perspectives on Metaphor: Cognitive Linguistics and Relevance Theory’, Journal of Pragmatics 40: 1823–64. Thaler, R.  H. (2000), ‘From Homo Economicus to Homo Sapiens’, Journal of Economic Perspectives 14: 133–41. Thumiger, C. (2007a), Hidden Paths: Self and Characterization in Greek Tragedy. Euripides’ Bacchae, London. Thumiger, C. (2007b), ‘Visione e identità nelle Baccanti di Euripide’, ACME II: 3–30. Thumiger, C. (2013), ‘Vision and Knowledge in Greek Tragedy’, Helios 40: 223–45. Tobin, V. (2008), ‘Literary Joint Attention: Social Cognition and the Puzzles of Modernism’, College Park, MD. Tobin, V. (2018), Elements of Surprise: Our Mental Limits and the Satisfactions of Plot, Cambridge, MA. Trendall, A. D. (1959), Phlyax Vases, BICS Supplement 8, London. Trenkner, S. (1957), The Greek Novella in the Classical Period, Cambridge. Tribble, E. (2011), Cognition in the Globe: Attention and Memory in Shakespeare’s Time, Basingstoke. Troscianko, E. T. (2014), Kafka’s Cognitive Realism, New York. Turner, F. and Pöppel, E. (1983), ‘The Neural Lyre: Poetic Meter, the Brain, and Time’, Poetry Magazine 142: 277–309. Turner, M. (1996), The Literary Mind, New York. Turner, M. (2016), ‘A Response: The Performing Mind’, in R.  Blair and A.  Cook (eds.),  Theatre, Performance and Cognition. Languages, Bodies and Ecologies, London, 68–74. Turner, M. and Fauconnier, G. (1995), ‘Conceptual Integration and Formal Expression’, Metaphor and Symbolic Activity 10.3: 183–204. Uithol, S., van Rooij, I., Bekkering, H., and Haselager, P. (2011), ‘Understanding Motor Resonance’, Social Neuroscience 6: 388–97. Ussher, R. G. (ed.) (1990), Sophocles Philoctetes (intr., trans., comm.), Warminster. Vaccari, A. P. (2016), ‘Against Cognitive Artifacts: Extended Cognition and the Problem of Defining “Artifact” ’, Phenomenology and the Cognitive Sciences 16: 879–92. Vandaele, J. and Brône, G. (2009), ‘Cognitive Poetics: A Critical Introduction’, in G.  Brône and J.  Vandaele (eds.), Cognitive Poetics: Goals, Gains and Gaps, Berlin, 1–29.

Bibliography  253 van Duijn, M. J. (2016), The Lazy Mindreader: A Humanities Perspective on Mindreading and Multiple-order Intentionality, Doctoral thesis, Leiden University. van Duijn, M. J. (2018), ‘Readers’ Mindreading Challenges, and How They Can Inform Cognitive Science’, Review of General Psychology 22.2: 188–98. van Duijn, M. J., Sluiter, I., and Verhagen, A. (2015), ‘When Narrative Takes Over: The Representation of Embedded Mindstates in Shakespeare’s Othello’, Language and Literature 24.2: 148–66. van Emde Boas, E. (2017), ‘Analyzing Agamemnon: conversation analysis and particles in Greek tragic dialogue’, Classical Philology 112, 4: 411–34. van Emde Boas, E. (2018), ‘Euripides’, in K. De Temmerman and E. van Emde Boas (eds.), Characterization in Ancient Greek Literature, Leiden, 355–74. van Emde Boas, E. (forthcoming), ‘Imagining Tragic Messenger Speeches: A Cognitive and Experimental Approach’, in F.  Budelmann and K.  Earnshaw (eds.), Cognitive Visions. Poetic Image-Making and the Mind. van Kuijk, I., Verkoeijen, P., Dijkstra, K., and Zwaan, R. A. (2018), ‘The Effect of Reading a Short Passage of Literary Fiction on Theory of Mind: A Replication of Kidd and Castano (2013)’, Collabra. Psychology 4.1: art. 7. van Nes, D. (1963), Die maritime Bildersprache des Aischylos, Groningen. Vasseur-Legangneux, P. (2004), Les tragédies grecques sur la scène moderne: une utopie théâtrale, Villeneuve d’Ascq. Vermeule, B. (2010), Why Do We Care About Literary Characters?, Baltimore, MD. Vermeule, E. (1966), ‘The Boston Oresteia Krater’, American Journal of Archaeology 70, 1–22. Vernant, J.-P. (1965), ‘Figuration de l’invisible et catégorie psychologique du double: le colossos’, in J.-P. Vernant and P. Vidal-Naquet, Mythe et pensée chez les Grecs, 2nd edn, Paris, 65–78. Vervain, C. and Wiles, D. (2001), ‘The Masks of Greek Tragedy as Point of Departure for Modern Performance’, New Theatre Quarterly 17.3: 254–72. Vidal-Naquet, P. (1972), ‘Chasse et sacrifice dans l’Orestie d’Eschyle’, in J.-P. Vernant and P. Vidal-Naquet, Mythe et tragédie en Grèce ancienne, vol. 1, Paris, 135–58. Visch, V. T. and Tan, E. S. (2008), ‘Narrative versus Style: Effect on Genre-Typical Events Versus Genre-Typical Filmic Realizations on Film Viewers’ Genre Recognition’, Poetics 36: 301–15. Visch, V. T., Tan, E. S., and Molenaar, D. (2010), ‘The Emotional and Cognitive Effect of Immersion in Film Viewing’, Cognition and Emotion 2: 1439–45. von Wilamowitz-Moellendorff, T. J. W. (1917), Die dramatische Technik des Sophokles, Berlin. Waldock, A. J. (1951), Sophocles the Dramatist, Cambridge. Walker, J. (2006), ‘The Text/Performance Split across the Analytic/Continental Divide’, in D.  Krasner and D.  Z.  Saltz (eds.), Staging Philosophy: Intersections of Theater, Performance, and Philosophy, Ann Arbor, MI, 19–40. Walton, K. L. (1978), ‘Fearing Fictions’, The Journal of Philosophy 75: 5–27. Walton, K. L. (1990), Mimesis as Make-Believe: On the Foundations of the Representational Arts, Cambridge, MA. Wayne, V., (ed.) (2017), Cymbeline, The Arden Shakespeare, Third Series, London. Waytz, A., Epley, N., and Cacioppo, J.  T. (2010), ‘Social Cognition Unbound’, Current Directions in Psychological Science 19.1: 58–62. Waytz, A., Morewedge, C. K., Epley, N., Monteleone, G., Gao, J., and Cacioppo, J. T. (2010), ‘Making Sense by Making Sentient: Effectance Motivation Increases Anthropomorphism’, Journal of Personality and Social Psychology 99.3: 410–35.

254 Bibliography Waytz, A., Heafner, J., and Epley, N. (2014), ‘The Mind in the Machine: Anthropomorphism Increases Trust in an Autonomous Vehicle’, Journal of Experimental Social Psychology 52: 113–17. Webb, R. (2009), Ekphrasis, Imagination and Persuasion in Ancient Rhetorical Theory and Practice, Farnham. Weiss, N. (2020), ‘Opening Spaces: Prologic Phenomenologies of Greek Tragedy and Comedy’, Classical Antiquity 39.2: 330–67. Welcker, F. G. (1829), ‘Über den Aias des Sophokles’, Rheinisches Museum 3: 43–92, 229–64. Welzer, H. (2008), ‘Communicative memory’, in A. Erll and A. Nünning (eds.), Cultural Memory Studies: An International and Interdisciplinary Handbook, Berlin, 285–98. Wheeler, M. (2010), ‘Minds, Things and Materiality’, in Malafouris and Renfrew (2010), 29–37. Whitman, C. H. (1951), Sophocles: A Study of Heroic Humanism, Cambridge, MA. Whitman, C. H. (1974), Euripides and the Full Circle of Myth, Cambridge, MA. Willems, R.  M. and Jacobs, A.  M. (2016), ‘Caring About Dostoyevsky: The Untapped Potential of Studying Literature’, Trends in Cognitive Science 20.4: 243–5. Willemsen, S., Kraglund, R. A., and Troscianko, E. T. (2018), ‘Interpretation: Its Status as Object or Method of Study in Cognitive and Unnatural Narratology’, Poetics Today 39: 597–622. Williams, T. (1945), The Glass Menagerie, London. Reprint, London, 2009. Wilson, E. (2012), ‘Sophocles and Philosophy’, in A. Markantonatos (ed.), Brill’s Companion to Sophocles, Leiden, 537–62. Winnington-Ingram, R.  P. (1948), Euripides and Dionysus: An Interpretation of the Bacchae, Cambridge. Winnington-Ingram, R. P. (1980), Sophocles: An Interpretation, Cambridge. Wolf, W. (1993), Ästhetische Illusion und Illusionsdurchbrechung in der Erzählkunst: Theorie und Geschichte mit Schwerpunkt auf englischem illusionsstörenden Erzählen, Tübingen. Wolf, W. (2008), ‘Is Aesthetic Illusion “Illusion référentielle”? “Immersion” in (Narrative) Representations and its Relationship to Fictionality and Factuality’, Journal of Literary Theory 2: 99–173. Wolf, W. (2013), ‘Aesthetic illusion’, in Immersion and Distance. Aesthetic Illusion in Literature and other Media, eds. W. Wolf, W. Bernhart and A. Mahler. Amsterdam: 1–63. Woolf, V. (1968), To the Lighthouse, Middlesex. Wong, Patrick  C.  M. and Diehl, R.  L. (2002), ‘How Can the Lyrics of a Song in a Tone Language Be Understood’, Psychology of Music 30: 202–9. Yates, F. A. (1966), The Art of Memory, Chicago, IL. Young, P. H. (1991), ‘Fighting in the Shade: What the Ancient Greeks Knew About Humor’, Soundings: An Interdisciplinary Journal, 74.1/2: 289–307. Zandberg, E. (2010), ‘The Right to Tell the (Right) Story: Journalism, Authority, and Memory’, Media, Culture and Society 32: 5–24. Zangwill, N. (2012), ‘Listening to Music Together’, British Journal of Aesthetics 52: 379–89. Zanker, A. T. (2019), Metaphor in Homer: Time, Speech, and Thought, Cambridge. Zanker, G. (1981), ‘Enargeia in the Ancient Criticism of Poetry’, Rheinisches Museum für Philologie 124: 297–311. Zeitlin, F. I. (1990), ‘Thebes: Theater of Self and Society in Athenian Drama’, in J. J. Winkler and F.  I.  Zeitlin (eds.), Nothing to Do with Dionysus? Athenian Drama in its Social Context, Princeton, NJ, 130–67. Zeitlin, F.  I. (1996), Playing the Other: Gender and Society in Classical Greek Literature, Chicago, IL.

Bibliography  255 Zunshine, L. (2006), Why We Read Fiction: Theory of Mind and the Novel. Columbus, OH. Zunshine, L. (ed.) (2010), Introduction to Cognitive Cultural Studies, Baltimore, MD. Zunshine, L. (ed.) (2015), The Oxford Handbook of Cognitive Literary Studies, New York. Zwaan, R. A. (1993), Aspects of Literary Comprehension: A Cognitive Approach, Amsterdam. Zwaan, R.  A. (2014), ‘Embodiment and Language Comprehension: Reframing the Discussion’, Trends in Cognitive Sciences 18.5: 229–34.

Index Locorum For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages. Aelianus Varia Historia 14.40 142n.28 Aeschylus Agamemnon 3 120–1 8 123n.39 18 121n.22 26–7 121n.23 37  120–1, 128–9 38 120–1 496–8 124n.41 1090 121 1091–7 121n.27 1117–18 121n.27 1186–90 121n.27 1192 121n.27 1197 120 1217–22 121n.27 1242–4 121n.27 1291 120n.19 1309 120n.21 1532–3 120n.21 Edonians TrGF fr. 58  122n.31 Libation Bearers 50 120n.21 168 123n.37 170 122–3 171 122n.35 172 123n.36 174 123n.38 175 122n.35 176  123nn.37–8 178 123n.37 187 123n.37 187–94 122–3 195 128–9 195–204 123 197 124n.43 201 124n.43 226 124n.44

231–2 124 238 122n.30 262–3 120n.21 744–5 120 808 121–2 810–11 121–2 843 120n.21 934 121n.29 963 121–2 971 121n.27 973–4 181–3 978 186–7 980  182–3, 189–90 980–9 181–3 985–6 183 1034 183–4 1054 183–4 1057 183–4 1061 183–4 Suppliants 335–47 63 368–9 69–70 379–80 69–70 399 69–70 407–9 70 437 72 438–42 71–2 452–4 73 468–71 73 Aristophanes Acharnians 440–1 138 Ecclesiazusae 296–9c 138n.17 1154–62 147 Lysistrata 759 205–6 Ranae 303 148n.49 908–35 194n.51 1043–4 142n.27

258  Index Locorum Thesmophoriazusae 1015 180n.24 1060 180n.24 Aristotle Nicomachean Ethics 1161a34–62b9 129n.65 1161b11 127n.57 Physics 243b11 125n.49 Poetics 1449b27 157n.20 1454a28–9 208n.40 1454b37–55a4 125n.51 1455a4–6 122n.34 1455a21–55b23 194n.51 1460b34 31n.23 Rhetoric 1385b13–15 149n.53 1386a17–24 158n.25 1413a 86 Athenaeus 11.13 Kaibel  150n.58 14.30 Kaibel  145n.38 Augustinus Confessiones 10.8.12 121n.26 Aulus Gellius Noctes Atticae 6.5 149n.54 Demosthenes Against Aristocrates 76 119–20 Dio Chrysostom 52.8 131n.75 Dissoi Logoi 3.10 156n.15 Euripides Alcestis 68–9 207 328–33 207 348–56 207 948 127n.58 Bacchae 918–21 191 924 191 927 193 1075 191 1122–3 191

1200–3 192 1232 192–3 1238–9 192–3 1244 192–3 1246–7 192–3 1257–8 192–3 1264 192–3 1279 192–3 Heracles 79 127n.58 986 127n.58 1376–85 126 1379  127, 128–9 1381–2 127 Medea 36–7 37 37 38n.44 38–43 37n.39 117–21 37–8 123–4 47n.41 214–58 32 292–315 40n.45 314–20 38 317 38n.44 344–7 40n.45 368–70 39 401–4 40 555 37n.38 568 37n.38 569–71 37n.38 588–90 37n.38 621 37n.38 689 37–8 772–7 39–40 791–6 40 792 37n.40 869–905 40n.45 873–81  40, 40n.45 929 38n.43 1008–14 37–8 1019–80 40 1042 40–1 1044 40–1 1048–9 40–1 1052 40–1 1057 40–1 1069 40–1 1076–7 40–1 1079–80 40–1 1129–31 38n.43 1236–50 40 1265–7 38n.43 Orestes 38 205 321 205

Index Locorum  259 836 205 866–954 205 1650 205 Phoenician Women 300 127n.59 306 127n.59 Suppliant Women 1174 121n.25 Trojan Women 762 127n.59 1142 127n.59 Gorgias fr. B 23 DK  20, 153–172 Helen fr. B 11.9 DK  156–7 fr. B 11.14 DK  159n.29 Heraclitus fr. B 5.22 DK  119n.15 Herodianus 1.417.27 186n.38 Herodotus 6.21.2  141, 157–8 8.41.2 205–6 Hesiod Theogonia 27 155 Hippocrates De victu 24 156 Homer Iliad 1.240–1 110n.50 3.229 98 6 107n.39 6.5 98 6.440–6 52n.13 7 105–6 7.211 98 11.609–10 110n.50 23 165–6 Odyssey 3.190 201–2 8 157–8 Pausanias 1.24.7 205–6 8.15.1–3 144–5

Pindar Nemean 7 20–3 155n.12 Plato Ion 535c5–8 157n.20 Republic 382d1–9 171n.71 439e–40a 194n.52 440a1–4 194 605c6–d5 171 606a3–d7 171 606b 158n.23 608a7 171 Pliny the Elder Naturalis Historia 35.65–6 155–6 Plutarch De Alexandri magni fortuna aut virtute 334a 142n.28 De gloria Atheniensium 5.348c  154n.6, 155–156 Quomodo adulescens poetas audire debeat 2.15d 154n.6 Vita Solonis 29.4–5 142n.26 Pollux Onomasticon 4.126–7 139n.18 Rhetorica ad Herennium 3.16–22 121n.26 Sannyrion Danae fr. 8  148n.49 Sophocles Ajax 5–8 53–4 18–20 53–4 31–3 53–4 36 53–4 40 53–4 51–2 53–4 77–8 103 118–20 53–4 119–22 103 289–90 106 292 106

260  Index Locorum Sophocles (cont.) 319–20 106 418–26 104–5 434–40 105 442–4 104–5 479–80 88 500–2 106–7 506–9 107–8 514–21 107 520–4  108, 116 522–4 51–2 523–4 107 594–5 51–2 652–3 109 646–53 46 646–92 45 651  47–48, 49–50 662–3 105–6 664–5 56 667 46–7 669–76 56 693–718 48 706–18 48–9 711 53 714 57 735–6 50 743–4 50 745  50–1, 52–3 807–8 50–1 817–18 105–6 822 129n.64 835–44 49 860 186n.38 866–1048 190n.45 962–3 109–10 964–5 109–10 1003  192, 195 1003ff. 190n.45 1006–22 112–13 1050 110 1053–4 110 1067–70  110, 112–13 1087–8 112–13 1099 111 1102 111 1133 112–13 1234 110–11 1236–8  101, 110–11 1245 110–11 1246–9 112–13 1266–79 102 1268–70 111–12 1273  112, 116 1273–87 112

1290 112 1334–5 114–15 1336–41 114 1336–42 115–16 1340–1 113 1343–4  113, 114 1354 113–14 1355 114 Antigone 2–6 88–89 18–19 87 31 87–8 38–9 88 45–6 88 49–68 89 58–60 89 59 93 61–2 90 65–8 90 72–3 91 74–5 92 76–9 91 80–1 92 82–5 93 89 92 90 90–2 92 90–1 99 86–7 175–7 83 450–7 96 460–8 94 542 121n.25 543 93 549 93 553 93–4 555 93–4 557–8 94–5 559–60 95 561–2 86–7 603 86–7 806–16 96 856 86n.13 857–66  86n.13, 87 Electra 36–7 153–4 37 160n.33 47 167–8 48–50 153 124–5 160n.33 124–6 153–4 170 160n.33 171–2 169 197 160n.33

Index Locorum  261 279 160n.33 481 121n.24 483 121n.24 491 121n.24 504–15 166n.48 649 160n.33 681–4 163–4 685–95 163–4 701–8 163–4 709–11 163–4 714–22 163–4 715–17 166–7 718 167n.52 720–1  163, 168n.55 723–7 163–4 729 164–5 731–46 163–4 744–5 163 745  164–5, 168n.55 745–8 164–5 747–56 163–4 749–56 165 761–3 169 774 169 807 167n.53 831 169 877–8 169–70 885–6 168 892–906 168 898 168n.55 900–2 168n.55 1027 93 1109 169 1261 169 1274 169 1285 169 Oedipus at Colonus 394 55 607–13 55–6 Philoctetes 15 223–4 24–5 221–2 26 223–4 26–36 221–2 49 221–4 50 222n.41 50–1 225n.60 53 223–4 54  221–2, 222n.41 57 222n.41 66–9 222n.42 86–111  214–15, 222 90 212–13

93 223n.49 93–4  222, 223n.50 94 223–4 96–9 205n.33 103 224n.51 113 213n.8 116 222 120  222, 222n.44 122  222, 222n.44 134 205–6 191–200 213n.8 242 215 268–9 131n.74 280–4 131n.74 385–8 205n.33 417 215n.17 470–1 131n.74 477 222–3 486–7 131n.74 507 212–13 528–9 223n.48 542–627 160n.33 874–5 215 889 222–3 890 223n.47 902–3 215 904 215 909 215–16 927–8 215–16 929 215–16 940 215 941 223n.48 950 215 965 212–13 993–4 222n.41 1018 130–1 1130–1 129 1226 223–4 1243–51 223n.50 1257–8 223n.50 1284 215n.17 1293 121n.25 1310–11  215, 215n.17 1327–8 205–6 1329–47 213n.8 1367 223n.48 1398 223n.48 1402 223n.48 1404 226n.64 1422 206–7 1440–4 206–7 1441–4 226n.63 1447 223–4 1470–1 201–2

262  Index Locorum Sophocles (cont.) Women of Trachis 21–3 185n.35 99–102 183n.32 102 185n.36 217 186–7 217–26 186–7 222 186n.39 223–4 187 226 187n.40 248–90 160n.33 904 127n.58 992  188–9, 195 1064 189n.44 1066 189n.44 1070 189n.44 1072–3  188–9, 188n.43 1076–8 189 1078 189 1079–80 189–90

Stobaeus 1.97.18 57n.19 Suetonius Nero 21.3 138n.15 Theognis 549 124n.41 Timocles Dionysiazusae fr. 6 K–A  148 fr. 6.5–7 Koch  157–8 Vita Aeschyli 9 141 Vitruvius De architectura 5.6.8 139n.18

General Index For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages. 4E approach, see cognition Achelaos 185 Achilles  110n.50, 113, 116, 139–40, 204–5 armour of, see Judgement of the Arms model for Neoptolemus  211–13, 215, 223n.46, 225, see also phusis un-Achillean behaviour  212–13, 215, 225–6 Actor 131n.75 Admetus 207–8 Aegeus  37–8, 41, 55 Aegisthus  121n.24, 130–1, 181–2, 188, 205 Aegyptids  63, 72, 77 Aeschylus Agamemnon House of Atreus  120–2, see also objects Eumenides audience response to chorus’ presentation 141, see also emotions Libation Bearers  83, 121–2 capacious sight invitation  181–4, see also sight invitations recognition scene  122–5 tokens 122–5 see also emotions, extended cognition, objects Persae 201–2 Philoctetes  131n.75, 204–5 Proteus 202 Suppliants dilemma and political authority  69–70, 76–7 (extended) metaphor and deliberation  69–78 supplication scene  61–2, 64–5, 69n.44, 75, 77–8 see also metaphor Aesop 136–7 aesthetic illusion  20, 153–61, 167 deceit and  170–2, see also immersion Agamemnon  69–70, 121n.24, 122–3, 153–4, 160n.33, 168–70, 183 memory of Ajax  99–104, 109–16 see also Atreidae

agathos 87–8 Agave 190–5 Ajax  11–12, 18–19, 43–59, 88, 98–116, 190n.45, 192n.50 memory of his past  102–15 mindreading attempts by characters  48–53 remembering Telamon  100n.8, 101–2, 104–5, 107–8, 112–13, 115 suicide  45–59, 103–6, 109, 129n.64 see also change of mind, deception, deception speech, Judgement of the Arms, memory, mindreading Alcestis 207–8 alêtheia 213–14 Alexander (tyrant of Pherai)  142–3 Amphitryon 130–1 anagnôrisis, see objects, recognition scenes anchoring anthropomorphism in animistic tendency  130 innovation in the condition humaine 228 material in conceptual blending  149–50, 183 ‘obedience to authority’ in sociocognitive structure of primate life  224 ‘self ’ in sociocultural context  225 ancient criticism  153–4, 171–2 Andromache  52n.13, 107n.39 anoia 86–7 anthropomorphism  5–6, 19–20, 119, 121–2, 124–30, 132 cognitive reflex  129–32 see also cognitive life of things, make-believe, objects Antigone  81–97, 138n.17 exchange with Ismene  11–12, 15–16, 19, 87–93, see also attribution theory low valuing of own life  82, 93–6 perceived dispositional irrationality  82, 86–7 rational self-presentation  82, 84–5, 87, 93–6 trait inheritance  87 apatê 153–72, see also deception Apollo  63, 153–4, 186, 197–8, 206–7, 208n.40 aporia  122–3, 127 Arendt, Hannah  219–20, see also obedience to authority

264  General Index Argos  63, 69–70, 76 Aristophanes of Byzantium hypothesis to Alcestis 207–8 hypothesis to Hippolytus 142n.27 hypothesis to Orestes  206n.35, 208n.40 Aristophanes  180, 194n.51 Acharnians doubleness of theatre  138 Ecclesiazusae choral address to audience  147–8 paradoxical blends  138 negative portrayal of Euripides  196 see also blending, gaze (comic) Aristotle  31–2, 86 on emotions, see emotions on recognition  122–3, 125n.51, 127n.57, 129n.65 on tragedy  142–3, 196, 198–9, 202 precursor of cognitive criticism  2–3 Artemis 186 Athena  53–5, 98, 102–4 Victory Athena the City Goddess  205–6 Athenaeus 145n.38 attribution theory  5–6, 14, 19, 34–5, 81–97, see also hyperbole, Malle Atreidae  45–7, 49, 52–3, 103–4, 109–16, 120n.21, see also Agamemnon, Menelaus Atreus House of, see Aeschylus’ Agamemnon sons of, see Atreidae audience cognition, see spectating audience  135–52, 173–95, see also emotions, joint attention, spectating authority figure  210–12, 216, 218n.26, 223–4, see also obedience to authority, Odysseus authority collective vision and  188 of memory  99–100, 102 political  69–70, 76 see also obedience to authority bacchantes 191–2 Bacchic  122n.31, 186 Beliefs, see attribution theory bia 213–14, see also violence Bigger Mysteries (Great)  144–5 blending  1, 5–6, 14, 19–20, 57–8, 136–40, 143–6, 177–8, 190–1, 193–4 actor-character  180–1, 183–4, see also Electra metaphor and  68–9, 77–8 see also conceptual blending theory, emotions, immersion, mask Boston Oresteia Krater  198n.6 bouleumata 39–40

Cadmus 191–3 Calchas 50–1 Caphereus (Cape)  201–2 Cassandra  121, 187–8 category concepts  198, see also genre Causal History of Reason, see attribution theory Cave, Terence  12–13, 17–18, 18n.40, 64, 210n.1, 226n.65 change of mind  37, 48–53, 204–5, see also mindreading, Theory of Mind characterization  5–6, 18, 25–42 mimetic character  33–6 realism  18, 26–7, 31–6, 41–2 synthetic character  32–4, 36 thematic character  32–4 see also mindreading chorus as mindreaders  48–53, 58 as vehicle of tradition and wisdom  47 emotional response to presentation of  141 focusing joint attention  186–7, see also Sophocles’ Women of Trachis Chrysothemis 168–70 Cithaeron (Mt.)  191 Clark, Herbert  21, 187–8, 211–12, 218–27 see also joint commitments Clytaemnestra  83, 121–2, 121n.24, 130–1, 153–4, 161, 167–70, 181, 188 cognition 4E approach  1, 6–7, 11, 118n.6, 210 see also embedded cognition, embodied cognition, enactive cognition, extended cognition, situated cognition, situated metacognition cognitive aesthetics of reception  1 cognitive archaeology  118, see also cognitive life of things cognitive categorization, see genre cognitive dissonance  52 cognitive historicism  1, 11 cognitive humanities  1–3, 6–8, 11–13, 16 cognitive life of things  19–20, 118–19, 121, 127–9, 132, see also extended cognition, objects cognitive linguistics category concepts  198 on metaphor  60–78 see also genre, image schemas, metaphor, relevance theory cognitive literary studies  18, 30n.19, 209 ability to generate readings  60–2, 64–5 development and history  1–3 epistemological status of cognitive criticism  7–10

General Index  265 Greek tragedy and  3–7 methodology 7–18 modes of variation  10–13, see also cultural specificity, universality second-generation  1–4, 6–7, see also 4E approach, cognition cognitive materialism  1, 11 cognitive narratology  1, 2n.4, 3–4, 14n.33, 26–7 emphasis emotional effects  159 inattentiveness to literary convention  33 cognitive poetics  1, 62–5 cognitive psychology  1–2, 5–6, 11n.30, 19, 43, 81, 118 cognitive turn  21, 219 conceptual blending theory  1, 20, 68–9, 136–40, 144n.37, 150–2, see also blending, conceptual metaphor, Fauconnier, spectating, Turner conceptual integration theory  7 conceptual metaphor, see metaphor conversation analysis  221n.38 Cook, Amy  20, 139 Correspondence Bias  81 co-variation model of attribution, see attribution theory Creon (king of Corinth)  37n.40, 38–40 Creon (ruler of Thebes)  83–9, 91–6 cultural specificity (of cognition)  10–13, 30–1, 219 attribution 81–2 joint attention  175 see also situated cognition, universality Cypria 199–200 Danaids  19, 60–78 Death (personified)  207, 208n.40 deceit, see deception deception  21, 37, 39–40, 43–59, 153–72, 210–16, 222, 226–7, see also aesthetic illusion deception speech, see Sophocles’ Ajax defamiliarization contradictory mythology versions  199–200 disappointment of expectation  199 see also genre Deianeira 185–9 delegated responsibility  211, 216, 218–19 Delos 63 Delphi  170n.69, 183–4 Demeter Eleusinian 144–5 Kidaria 144–5 Demodocus 157–8 Demosthenes 119–20 Desdemona 138n.15 Desires, see attribution theory

deus ex machina  206–7, 212–13, 224n.54 Dicaeopolis 138 dikaios 155–6 Dio Chrysostom Oration 52  204 Oration 59  204 Diomedes 204–6 Dionysia  150n.58, 186–7 Dionysus  186, 190–4 Theatre of  11–12, 197–8, 204–5 disobedience  212, 224–6 divine agency  53–5 cyclic change  55–7 Dokimasia Painter, see Boston Oresteia Krater dolos  39, 160n.33, 214–16, see also deception dramatic irony  47 duslogiston 53–4 Egypt 202 Eichmann, Otto Adolf  218–19 trial  216, 218–19 see also obedience to authority eidotas 156 Electra  19–20, 93, 153, 160–1, 167–9, 205–6 actor-character blending  149–50, see also blending interpreting recognition tokens  122–5, 128–32, see also emotions, extended cognition, objects eleos 213n.10, see also pity embodied cognition  1–7, 13–14, 29–31, 37–8, 43–4, 60–1, 64–5, 68–9, 71n.50, 72, 75–8, 118n.6, 126–7, 147, 159–60, 162, 172, 180–1, 199, 210, 226n.65 emotions ancient audiences  135, 141–6, 148–50, 157–9 Aristotle on  142–3, 157–9 blending and  150–1 expectations 198–9 joint attention  174–5, 177–8 memory and  19, 99–108, 112–15 mindreading  27, 40–1, 43–4 objects  124, 128–9, 131–2 oversimplified attribution  83 pity and fear  142–3, 148–9, 149n.53, 151–2, 158, 212–13 Plato on  142–3, 157, 158n.23 see also blending, embodied cognition, extended cognition, immersion, memory Enabling Factors, see attribution theory enactive cognition  1, 20, 29, 34n.30, 125, 154, 160–72 enargeia 169–70 Erechtheum 205–6

266  General Index Erinyes  121, 183–4, 205 Eteocles 69–70 Euboea 201–2 eugeneia  88, 91 Eumenides 205, see also Erinyes Euripides 203–8 Alcestis happy ending  207 pro-satyric 207 Bacchae double vision  191 limits joint attention  190–4 metatheatricality 190–1 Electra  197–8, 204 Euripidean comedy  208n.42 Helen 207–8 Heracles  118, 201–2 anthropomorphic thinking  126–7 bow and arrows  19–20, 118–19, 126, 128–31 embodied memory  126–7 see also objects Hippolytus Garlanded 142 Hippolytus Veiled rejection of  142, 196 Ion 207–8 Iphigeneia in Tauris 207–8 Medea  25–42, 44 Great Monologue  40–1 intertextuality with Neophron’s Medea  37n.40 see also mindreading Orestes 197–8 Athenian flavour  205 disappointment expectations  205–6 Philoctetes  131n.75, 204–5 see also anthropomorphism, characterization, extended cognition, genre, joint attention, metatheatricality, objects, Pseudo-Euripides Eurysaces 107–8 Event-Related Brain Potentials  199–200 evolutionary approach to religion  54–5, 130 evolutionary literary theory  1, 11n.29, 28n.12, 151n.63 Exceptionality Thesis  27, 33n.26, see also realism experiment (scientific) archive of human (meta)cognition  210–11 special form of social interaction  221n.37 extended cognition  1, 5–6, 117–19, 128–9, see also cognitive life of things, emotions, make-believe, objects false-merchant scene (Sophocles’ Philoctetes)  160n.33 false-messenger speech, see Sophocles’ Electra

Fauconnier, Gilles  7, 20, 68–9, 136–7, 139n.19, 140, 143–4, 151–2, 178n.19, 179n.21, 181n.26, 190–1 fear, see emotions feminist criticism  1–2, 15–16, 87n.14 fMRI neuro-imaging  1–2, 130n.71, 158–9 focus, see joint attention, spotlight folk psychology  13–14, 28, 30–1, 35–6, 41–4, 52, 54–5, 82 folk tale  202n.24, 203–4, 207 fourth wall  178, 180–1 Fundamental Attribution Error  81, see also attribution theory gaze, tragic and comic  135, 150–2, see also spectating genre  13, 20–1 cognitive approaches to  196–201, see also image schemas generic expectations  196, 201–8 predictive processing of narrative  200–3, 207–9 probability designs  201–7 prototype theory  197–8 see also emotions, gaze, spectating gestalt 74–5 Gorgias fr. B 23 DK  153–72 Helen  156–60, 171 see also aesthetic illusion, cognitive narratology, deception, immersion Great Monologue, see Euripides’ Medea group attention  175n.9 group psychology  174 Hades, see underworld happy ending  201–2, 202n.24, 206–8, 223–4, see also genre Hector  52n.13, 57, 105–6, 107n.39, 109, 112 hêdonê 143 Hegelochus 148n.49 Helenus 214n.8 Heliodorus Aethiopica 81n.1 Helios, see Sun Heracles  19–20, 126–32, 137–8, 185, 188–90, 192n.50, 195, 206–8, 212–13, 223–4, 225–6 Heraclitus  56n.17, 119–20, 122 hermeneutics  15–16, 64, 123, 171, see also cognitive literary studies Hermione 207–8 Herodotus  141, 157–8 Hesiod 155 history play  202–3 Holocaust  211, 218–19

General Index  267 Homer  166n.47, 197–8 consensus-building in  221n.37 Homeric hexameter  199–200 metaphor in  61n.5 model of heroism  52n.13, 98, 104–6 Hyllus 188–9 hyperbole  66, 130–1, 158–9 extreme attribution  86, 93 iambos 198 identity blended 138–40, see also blending dual 190–1, see also blending group  99–100, 107–8 individual 100 layered 180n.24 memory and  99–103, 107–8, 115, see also memory illocutionary force, see also speech acts image schemas  5–7, 14, 19, 62, 71–5, 77, 197 as abstract conceptual structures  67–9 simulations of embodied experience  68, see also embodied cognition immersion  20, 135, 155–6, 160–70 affective response of  135, 148–50 contextualization and  137–8, 150–1 immersive features of narrative  162–3 interaction and  147–8 reflexive distance and  148–50, 155–6, 161–2, 167 see also aesthetic illusion, blending, deception, spectating interaction engines  219 interaction theory  29–30 intersubjectivity  29–31, 174–5 mindreading 25n.2 networks of  176–8, 181–5, 188, 190–4 see also joint attention, sight invitations Io 63 Iole 160n.33 irrationality  53–4, 82, 87 Ismene  55–6, 81–97 exchange with Antigone  11–12, 15–16, 19, 87–93, see also attribution theory Jason  37, 38n.43, 39–40 joint actions  220–3 joint attention  5–6, 14, 20, 173–95 affective power of  176–8, 185 epistemic limits of  190–4 joint knowledge and  184–90 mutually manifest  174–5, 180–1, 185, 193–4 see also memory, sight invitations, speech acts joint commitments  212 exploitation 220–4

negotiation 219–21 overcommitment 220–4 stacked joint actions  219–20, see joint actions see also Clark, obedience to authority Judgement of the Arms  53–5, 101–2, 104–5 kalon 91 katharsis  143, 159n.29, see also emotions kharis  51–2, 111 Laertes  214, 215n.17 Lakoff, George  20–1, 60–1, 130n.67, 136, 197–8, see also genre, image schemas Lemnos uninhabited  130–1, 204–5, 212–13 Leontius 194–5 Lichas  160n.33, 187, 192 literature as archive of cognition  53, 57–8, 210–12, 226–7 as archive of metacognition  210–12, 226–7 see also storytelling logos 158n.23 contrasted with phusis  211, 214 Lycurgus 122n.31 Macbeth 177–8 madness  5, 43, 54–5, 98, 102–3, 106, 109, see also irrationality maenads  186, 191 make-believe  5–6, 8–9, 19–20, 128–32, 143n.32, 155–6, see also aesthetic illusion, anthropomorphism, cognitive life of things, deception, objects Malafouris, Lambros  19–20, 118–19, 128–9, 132, see also cognitive life of things, extended cognition, objects Malle, Bertram  19, 43, 82–6, see also attribution theory mask as mind tool  181 comic  138, 147 (failed) blending and  137–8, 181 in mystic performance  144–5 tragic  5, 44n.1, 137–8 see also blending, sight invitations McConachie, Bruce  20, 137–40, 145–6 Medea  25–42, 44 mindreading attempts by characters  36–42 realism and  31–6 see also characterization, mindreading Megara 127 memory  1, 4–5, 19, 34–5, 98–116 collective  100, 107–8 image schemas and  68

268  General Index memory (cont.) joint attention and  175 world view and  99–103, 113–15 see also embodied cognition, emotions, enactive cognition, identity, rhetoric memory play  98–9, 115 Menelaus  206–7, 208n.40 memory of Ajax  99–100, 103–4, 109–16 see also Atreidae metaphor  2n.5, 4–5, 14–15, 19, 136n.5, 197 as leitmotiv  124 deliberation and  69–75 embodied experience and  75–8 ‘mainstream’ view of  62, 64–6, 69, 74–6 memory and  101n.9, 121 models of comprehension  65–9 personification and  130n.67 see also blending, image schemas metatheatricality  135, 148–50, 193–4, 215n.14 metonymy  120–1, 166–7, 169, 197 Miletus, see Phrynichus Milgram, Stanley  21, 211–28, see also obedience to authority mind-body dualism (Cartesian)  1–3, 13–14 mindreading  18, 25–46, 52–4, 57–9 ‘higher-level’ processing  26–31 ‘lower-level’ processing  26–31 real and fiction minds  26–31 see also attribution theory, change of mind, characterization, Theory of Mind Mycenae  153, 169, 183 narrative experiential quality of  154, 160–71, see also aesthetic illusion Narrative Practice Hypothesis  29, 31, see also mindreading natural disposition, see phusis Neophron Medea 37n.40 Neoptolemus  11, 129, 131–2, 204–7, 210–27, see also joint commitments, obedience to authority, phusis Nero 137–8 Nero-Hercules 137–8 Nestor 201–2 neuroaesthetics 7n.20 neuroscience  1–4, 6n.19, 7–10, 16–18, 28n.12, 43, 121n.26, 158–9, 175n.13 New Comedy  202 New Criticism  14 New Materialism  1–2, 118 Niobe 148 nomos 89

Nostoi 201–2 nostos-plot 201–2, see also genre, New Comedy novella, Greek  203–4 Nurse  36n.36, 37–8, 40 obedience to authority (experiments in)  210–12, 216–27, see also Clark, joint commitments, Milgram objects (material)  5–6, 19–20, 117–32 affective power of  117, 124 as cognitive artifacts  117–18 embodied memory and  126–7 extended cognition and make-believe  128–32 immaterial 177 resisting and collaborating  118, 122–5 see also anthropomorphism, cognitive life of things, emotions, extended cognition Odysseus  157–8, 206–7 authority figure  212–16, 221–6 memory of Ajax  103–4, 109, 113–15 mindreading 53–5 use of deception  129, 153–4, 204–7, 210–11, 215 Oedipus  27–8, 31, 47n.7, 55–7, 149n.53, 202 Oenomaus 166 oiktos 212–13, see pity Old Comedy  178, 180, 187–8 oracles  5, 55, 139, 153–4, 213n.8 Orestes  83, 121–5, 149–50, 153, 160–70, 185–7, 189–90, 192, 195, 197–8, 204–7 capacious sight invitation by  181–4, see also sight invitations Orestes-Electra plays  197–8 Orsino 57–8 Paean 186, see Apollo Paedagogus (Sophocles’ Electra)  15, 20, 153, 160–70 paradox of fiction  135, 143–4 paratragic 202 Parrhasius 155–6 pastoral 202–3 Patroclus 57 peithô 215, see persuasion Pelasgus  15, 19, 60–78 Pelops 166 Pentheus 190–4 performance studies  20, 118, 137–8 Persians 141–2 personality circumstances and  212, 224–5 persuasion  39, 40n.45, 86, 156–7, 162n.37, 206–7, 211, 213–15, 224–7 Petroma 144–5

General Index  269 Phaeacians 157–8 Phelan, James  32–4 Pheneos 144–5 phenomenology  1–2, 8, 20, 29n.15, 155–6, 176, 194 Pheraean elders  207 Pherai 142 philia 205–6 Philochorus 150 Philoctetes  201–2, 204–7, 210–27 bow  129, 131–2, 206–7, 212–13, 215, 222, see also objects social isolation, see Lemnos philosophy of mind  18, 26, 43 Phineus 148 Phocis 161–2 phusis  211, 215–16, 221, 224n.54, 225–7 contrasted with logos  211, 214 Phrygians 110 Phrynichus fined for Sack of Miletus  141–3, 157–8, 196 Pindar 154n.8 pity, see emotions Plato on emotions, see emotions on aesthetic illusion  171 precursor of cognitive criticism  3n.8 sight invitation in  194–5 Plutarch  142–3, 154–6 Polus 149–50 Polynices  84–5, 88, 90–3 posthumanism  6, 118 postmodernism  27n.10, 202 poststructuralism 1–2 Praxagora 138 predictive processing of narrative, see genre probability designs, see genre problem play  206–8 pro-satyric 207 prototype theory, see genre Prytaneion 119–20 Pseudo-Euripides Rhesus 139 Pylades  205–6, 208n.40 Pythian Games  153 reader response, cognitive models of  1–2, 33–4, 153–4, 159–60, 170 realism, see characterization Reasons, see attribution theory recognition scenes  5, 19–20, 122, 125, 177–8, 190–4, 208n.42, see also Aeschylus’ Libation Bearers, Aristotle, Euripides’ Bacchae

reductionism  16–17, 221–2 relevance theory  19, 62, 65–9, 71, 74–8, see also metaphor rhetoric  1, 4–5, 20, 40n.45, 47–8, 60, 166–7 attribution and  81, 86, 93 objects and  128, 132 persuasion and  212, 214–15, 225–6 use of memory in  98–100, 102, 107n.38, 108 see also enargeia Richardson, Alan  1, 6–7, 8n.21, 11 romance (play)  203, 207–8 Rosch, Eleanor  20–1, 197–8, see genre Russian Formalism  199 Salamis 54–5 Sannyrion mockery of Hegelochus  148n.49 satire 202–3 satyr play  135n.3, 202, 204n.31, 207–8 Scamander 104–5 selfhood 225 as product of social interactions  225 individuation and socialization  225 see also personality sensorimotor system  68, 160, 163–5 Shakespeare All’s Well That Ends Well 207–8 cognitive approaches to  4–5, 64–5, 174n.5 Cymbeline mixed genre  203 Hamlet 199n.12 Macbeth sight invitation  177–8, 183–4 Measure for Measure 207–8 Othello failed blending in  138n.15 The Tempest 207–8 The Winter’s Tale 207–8 Troilus and Cressida 203 genre categorization  203 Twelfth Night 57–8 see also blending, genre, sight invitations shared attention  174–5, 175n.11, 181, see also joint attention sight invitations  20, 173, 176–95 see also blending, gaze, joint attention, objects, speech acts simulation theory  28 Sisyphus 215 situated cognition  3–4, 5–6, 11n.28, 210–27 situated metacognition  210–27, see also situated salience situated problem-solving  210n.3, 218n.28 situated salience (differential)  21, 212, 226–7

270  General Index skopein 184–5 Skyros 201–2 social cognition  14n.34, 25n.2, 28–30, 34–5, 43–4, 173, 190, 194 social pressure  223n.50, 224–5 social psychology  1–2, 5–6, 11n.30, 12, 19–20, 129–32, 210–11, 216, see also attribution theory Socrates  171, 194–5 Solon 142 sophia 154–6 sophists influence on Sophocles  211 social anthropological interests  213–14 Sophocles  5, 192n.50 Ajax  18–19, 43–59, 98–116 deception speech  45–57, 105–6 see also change of mind, characterization, deception, mindreading Antigone  19, 81–97, see also attribution theory Athamas 206n.36 Electra  20, 121n.24, 149–50, 153–72, 197–8 Chrysothemis’ rhesis 168–70 epic and epinician echoes  165–6 false-messenger speech  160–70 see also aesthetic illusion, deception, immersion, narrative (experiential quality of) Oedipus at Colonus 55–7 Oedipus the King 202 on Euripides’ characterization  31–2 Peleus 206n.36 Philoctetes  11, 20–1, 129–31, 201–8, 210–27, see also genre, situated cognition Women of Trachis  177–8, 184–90, 201–2, see also sight invitations Spartans 111 spectating  4–5, 135–52 dynamic model of  135, 146–50 see also blending, gaze, genre, joint attention, subjunctive reality, suspension of disbelief speech acts  161–2 illocutionary force  176–7, 192n.49 sight invitations as  178–81 see also sight invitations, theatrical communication spotlight  176–7, 190–1, 194 social cognition and  173 see also joint attention, sight invitations Stesichoros 197–8 Stobaeus 57n.19

Stoicism precursor of enactivism  3n.8 storytelling evolutionarily beneficial  28n.12 scaffolding folk psychology  31n.21 structuralist theory  199 subjunctive reality in ritual performance  144–5 in theatrical performance  139–40, 151–2 see also blending, spectating Suda 57n.19 suicide, see Ajax Sun (personified)  181–5 two suns (double vision)  191, 193 supplication, see Aeschylus’ Suppliants Susanna and the Elders, story of  203 suspension of disbelief  5, 135, 139–40, see also spectating Tecmessa misattributing Ajax’ mindstates  18–19, 45–55, 58 remembering Ajax  103–4, 106–10, 114–16 tekhnê  214–15, 221 tekmêria 169 Telamon, see Ajax Telemachus 201–2 Telephus 148 Terence Hecyra 150n.57 terpsis 157–8 tetralogy  200, 202, 204n.31, 206–7 Teucer  50, 190n.45 remembering Ajax  99–100, 102–4, 109–16 theatrical cognition focus on individual spectator  174 theatrical communication  178–81, see also sight invitations theatrical space  139, 144, 177, 180, 190, see also blending, joint attention theatron 175 Thebes  92, 177–8, 190–4 Theodorus 142 Theognis 124n.41 Theory of Mind  5–7, 14, 18–19, 25–7, 28n.12, 30n.19, 43–4, 81, 91–2, 174–5, see also characterization, mindreading ‘theory theory’  28 Thersites 203 Theseus  55, 130–2 Thespis 142 Thetis 206n.36 Thucydides 164n.42

General Index  271 Thyestes 121 Timocles  148, 157–8 Timotheus 68n.41 tragicomedy 203 Traits, see attribution theory Trojan War  100, 112, 156–7, 203, 225–6 tropoi 50 Trugrede, see deception speech Turner, Mark  7, 20, 60–1, 68–9, 136–7, 139n.19, 140, 143–4, 151–2, 178n.19, 179n.21, 181n.26, 190–1 Tutor (Euripides’ Medea)  36n.36, 37–8 Tychoism 32n.24 Tyndareus 205–6 universality (of cognition)  10–13, 19–20, 30–2, 130, 151, 210n.1, 225–7 attribution 81–2 ‘obedience to authority’  220 see also cultural specificity, situated cognition

underworld  120, 144–5, 183 unnatural narratology  2n.4, 3–4, 14n.33, 26–7, see also cognitive narratology Valuings, see attribution theory violence  53–4, 114–15, 213–17, 223, 226–7 vision 173–95 individual (limits of)  185, 188–9, 192–3 theatrical  177–8, 183–4, 188–9, 194 see also blending, joint attention, spectating, Sun visual attention  20, 175–7, see also joint attention, sight invitations Watchman (Aeschylus’ Agamemnon)  19–20, 120–1 Wooden Horse  157n.22 Woolf, Virginia To the Lighthouse 129–31 World War II  218–19 Zeus  48, 183, 206–7 Zeuxis 155–6