Distributed Cognition in Classical Antiquity 9781474429764

Citation preview

D i st r i but e d C o g n i t i o n i n C la s s ica l Antiq u ity

Distributed Cognition in Classical Antiquity Edited by Miranda Anderson, Douglas Cairns and Mark Sprevak

Edinburgh University Press is one of the leading university presses in the UK. We publish academic books and journals in our selected subject areas across the humanities and social sciences, combining cutting-edge scholarship with high editorial and production values to produce academic works of lasting importance. For more information visit our website: edinburghuniversitypress.com

© editorial matter and organisation Miranda Anderson, Douglas Cairns and Mark Sprevak, 2019 © the chapters their several authors, 2019 Edinburgh University Press Ltd The Tun – Holyrood Road, 12(2f) Jackson’s Entry, Edinburgh EH8 8PJ Typeset in 10/12 Monotype Baskerville by Servis Filmsetting Ltd, Stockport, Cheshire and printed and bound in Great Britain. A CIP record for this book is available from the British Library ISBN 978 1 4744 2974 0 (hardback) ISBN 978 1 4744 2976 4 (webready PDF) ISBN 978 1 4744 2977 1 (epub) The right of Miranda Anderson, Douglas Cairns and Mark Sprevak to be identified as the editors of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988, and the Copyright and Related Rights Regulations 2003 (SI No. 2498).

C o n t e nts

List of Illustrations Series Preface

v vi

  1. Distributed Cognition and the Humanities Miranda Anderson, Michael Wheeler and Mark Sprevak

1

  2. Distributed Cognition and the Classics Douglas Cairns

18

  3. Physical Sciences: Ptolemy’s Extended Mind Courtney Roby

37

  4. Distributed Cognition and the Diffusion of Information Technologies in the Roman World Andrew M. Riggsby   5. Mask as Mind Tool: A Methodology of Material Engagement Peter Meineck

57 71

  6. Embodied, Extended and Distributed Cognition in Roman Technical Practice 92 William Michael Short   7. Roman-period Theatres as Distributed Cognitive Micro-ecologies Diana Y. Ng

117

  8. Cognition, Emotions and the Feeling Body in the Hippocratic Corpus George Kazantzidis

132

  9. Enactivism and Embodied Cognition in Stoicism and Plato’s Timaeus 150 Christopher Gill 10. Enargeia, Enactivism and the Ancient Readerly Imagination   169 Luuk Huitink

iv

conten t s

11. Group Minds in Classical Athens? Chorus and Dēmos as Case Studies of Collective Cognition 190 Felix Budelmann 12. One Soul in Two Bodies: Distributed Cognition and Ancient Greek Friendship 209 David Konstan 13. Distributed Cognition and its Discontents: A Dialogue across History and Artistic Genre Thomas Habinek and Hector Reyes

225

Notes on Contributors 240 Bibliography 243 Index 287

L i s t o f I l l us tra tions

Figure 6.1  Detail after an artist’s rendition of the Darius Vase (South Italian red-figure vase found at Canosa, Puglia, dating to about 340 to 320 bce, now housed at the Museo Archaeologico Nazionale in Naples, Italy). The seated calculator manipulates a mounted counting board while holding a writing tablet. 105 Figure 6.2  Schematic images metaphorically underlying Latin’s expressions of mental activity. 111 The plate section can be found between pages 70 and 71. Plate 1  Detail from the Pronomos Vase. Attic red-figured volute krater by the Pronomos Painter, c. 420–400 bce (Naples, NM 81673). Credit: akgimages/Album/Oronoz. Plate 2  Red-figure chous or oinochoe fragment, c. 430 bce. American School of Classical Studies at Athens: Agora Excavations (P. 32870). Plate 3  Raphael, School of Athens, Stanza della Segnatura, Vatican Palace. Photo credit: Erich Lessing/ART RESOURCE, New York. Plate 4  Nicolas Poussin, Landscape with Numa Pompilius and the Nymph Egeria, Musée Condé, Chantilly. Photo credit: Harry Bréjat/RMN-Grand Palais (Domaine de Chantilly).

S e r i e s P r ef a ce

This book, like the series of which it is part, explores the notion that the mind is spread out across brain, body and world, for which we have adopted ‘distributed cognition’ as the most comprehensive term. Distributed cognition primarily draws evidence from philosophy, cognitive science, psychology, linguistics and neuroscience. Distributed cognition covers an intertwined group of theories, which include enactivism and embodied, embedded and extended cognition, and which are also together known as ‘4E cognition’. An overview and explanation of the various strands of distributed cognition are provided in the general introduction. Distributed cognition can be used as a methodology through which to pursue study of the humanities and is also evident in past practices and thought. Our series considers a wide range of works from classical antiquity to modernism in order to explore ways in which the humanities benefits from thinking of cognition as distributed via objects, language and social, technological and natural resources and environments, and to examine earlier notions that cognition is distributed. Theories of distributed cognition are transformative in terms of how we understand human nature and the humanities: they enable a different way of perceiving our interactions in the world, highlighting the significant role of texts and other cultural artefacts as part of a biologically based and environmentally grounded account of the mind. Theories of distributed cognition offer an opportunity to integrate the humanities and the sciences through an account that combines biologically and culturally situated aspects of the mind. The series illuminates the ways in which past ideas and practices of distributed cognition are historically and culturally inflected and highlights the cognitive significance of material, linguistic and other sociocultural developments. This evidence has the potential to feed back into cognitive science and philosophy of mind, casting new light on current definitions and debates. Each volume provides a general and a period-specific introduction. The general introduction, which is replicated across all four volumes, aims to orientate readers unfamiliar with this area of research. It provides an overview of the different approaches within distributed cognition and discussion of the value of a distributed cognitive approach to the humanities. The period-specific introductions provide a more detailed analysis of work in the cognitive humanities in the period covered



ser ies pr eface vii

by the volume, before going on to reflect on how the essays in the volume advance understanding in the humanities via distributed cognition. The project from which this series emerged provided participants with an online seminar series by philosophers working on various aspects of distributed cognition. These seminars are publicly available on the project’s website (http://www. hdc.ed.ac.uk/seminars). The seminars aim to help researchers in the humanities think about how ideas in distributed cognition could inform, and be informed by, their work. Four workshops were held in the summer of 2015 at the University of Edinburgh and were attended by nearly all the volume contributors. The workshops brought participants together to collaborate in ways that contributed not just to the making of this series but to the development of this approach to the humanities. From the springboard of the seminar series, through the gathering together of scholars from across a range of disciplines and by ongoing interaction with the editorial team during the production of the final essays, the aim has been to provide a set of rigorous analyses of historical notions of distributed cognition. The series is deliberately exploratory: the areas covered by the essays are indicative of the benefits of the general approach of using distributed cognition to inform cultural interpretations. The first four volumes of the series concentrate primarily on Western Europe; however, we envisage further future volumes that will expand the scope of the series. If distributed cognition is understood as we contend, then this understanding has the potential to be valuable across the humanities as part of a new type of intellectual history. The four volumes are arranged chronologically and each of the volumes is edited by a period specialist (Cairns, Anderson, Rousseau, Garratt), a philosopher (Wheeler or Sprevak), and Anderson, whose central involvement in all four volumes ensured overall consistency in approach and style. At the very moment when modern-day technological innovations reveal the extent to which cognition is not just all in the head, this series demonstrates that, just as humans have always relied on bodily and external resources, we have always developed theories, models and metaphors to make sense of the ways in which thought is dependent on being in the world.

Acknowledgements This series emerged from the project ‘A History of Distributed Cognition’ (2014– 18), funded by the Arts and Humanities Research Council (AHRC), whom we should like to thank most warmly for their support. The idea for the project first came about in 2010. Miranda Anderson realised that the resonances she was exploring between recent ideas on distributed cognition and Renaissance notions of the mind were not just a matter of a correlation between two points in time, but reflected an important aspect of the mind in history that has been neglected, one that, fittingly, might be best explored through a collaborative project. Our interdisciplinary team has worked closely together developing the project, the monograph series and this general approach to the humanities: ours has been an intellectual endeavour akin to Hutchins’s description of a ship’s crew successfully navigating

viii

ser ies pr ef ace

by means of collective computation (1995). With Douglas Cairns as Principal Investigator and the core project team of Miranda Anderson, Mark Sprevak and Michael Wheeler, we have collaborated closely both between ourselves and with other scholars. For volumes 3 and 4 respectively, the editorial team were joined by period specialists George Rousseau and Peter Garratt, who helped shape these volumes. Boleslaw Czarnecki, research assistant on the project during 2016–17, helped liaise with contributors and with the organisation of public engagement activities during this time. We were fortunate to have had two excellent auditors, in the shape of Terence Cave and Tim Crane, who used their years of accumulated experience and wisdom to help monitor our progress. Our advisory board offered expertise from across a wide spectrum of academic disciplines: Andy Clark, Giovanna Colombetti, Christopher Gill, David Konstan, Karin Kukkonen, Duncan Pritchard, Andrew Michael Roberts and Patricia Waugh. We are very grateful to the philosophers who came on board to provide us with the online seminars and joined us in online discussion: Andy Clark, Giovanna Colombetti, Shaun Gallagher, John Sutton, Deb Tollefsen, Dave Ward, Dan Zahavi, as well as our own Michael Wheeler. The editors would also like to thank those scholars (John Bintliff, Orazio Capello, Nick Lowe, Cristóbal Pagán Cánovas and Lisa Sherbakova) who made valuable contributions to the project workshops but whose papers could not for various reasons be included in the final volume. The editorial team is especially grateful to Duncan Pritchard and Eidyn, the University of Edinburgh’s Philosophy Research Centre, for their support of a pilot of this project in 2012–13. The Balzan Project, ‘Literature as an Object of Knowledge’, led by Terence Cave, also kindly supported the project, providing funds for the images on our website. The project has benefited from the involvement of many of the participants from the Balzan Project in our workshops and volumes including Guillemette Bolens, Terence Cave, Mary Crane, Karin Kukkonen, Raphael Lyne, Emily Troscianko, and our own Miranda Anderson. Meanwhile, the AHRC-funded Cognitive Futures in the Humanities Network (2012–14), co-led by Peter Garratt, with Michael Wheeler as a founding s­ teering-committee member, has also fostered further productive interactions and cross-fertilisation. We warmly thank the National Gallery of Scotland for helping us to source and secure our website images. The National Museum of Scotland (NMS) was our supportive project partner, and NMS staff met with the team to discuss and assist with the development of public engagement activities. These activities included a series of recorded public lectures, during which museum curators and academics provided their perspective on the cognitive implications of museum artefacts. Malcolm Knight, the multitalented man behind the Scottish Mask and Puppet Centre, illustrated the cognitive dimensions of masks and puppets and provided much entertainment during one of the NMS lectures. NMS also provided us with their classrooms for our school workshops. Lisa Hannah Thompson was an invaluable contributor to the development of our ideas on how best to shape the material and the programme for children in order to connect in fun and effective ways.



ser ies pr eface ix

Editorial Notes Contributors have been allowed to follow their own preferences with regard to the use of subscript versus adscript iotas, as well as in the use of direct transliteration (versus the tradition anglicised or latinised forms) in rendering ancient Greek proper names. In following this policy the editors have sought to impose consistency within rather than between chapters.

The Edinburgh History of Distributed Cognition Series Editors: Miranda Anderson and Douglas Cairns Scholars from across the disciplinary spectrum track the notions of distributed cognition in a wide range of historical, cultural and literary contexts from antiquity through to the twentieth century. Distributed Cognition in Classical Antiquity Edited by Miranda Anderson, Douglas Cairns and Mark Sprevak Distributed Cognition in Medieval and Renaissance Culture Edited by Miranda Anderson and Michael Wheeler Distributed Cognition in Enlightenment and Romantic Culture Edited by Miranda Anderson, George Rousseau and Michael Wheeler Distributed Cognition in Victorian Culture and Modernism Edited by Miranda Anderson, Peter Garratt and Mark Sprevak Visit the series website at edinburghuniversitypress.com/series/ehdc

1 D i st r i but e d C o g n i t i o n a nd the Hu ma nities Miranda Anderson, Michael Wheeler and Mark Sprevak1

Consider counting on your fingers; or solving a challenging mathematical problem using pen and paper (or Napier’s bones, or a slide-rule); or the way in which we routinely offload the psychological task of remembering phone numbers on to our ubiquitous mobile phones; or a brainstorming scenario in which new creative ideas emerge from a process of collective group interaction; or the manner in which the intelligent feat of ship navigation is realised through a pattern of embodied, information-communicating social exchanges between crew members who, individually, perform purely local information-processing tasks (such as bearing taking) using specialised technology. All of these examples of brain-body-world collaboration are, potentially at least, instances of the phenomenon that, illuminated from a historical perspective, is the topic of this volume. That phenomenon is distributed cognition. So what, precisely, is distributed cognition? The term itself is standardly traced to the pioneering work of the cognitive anthropologist Edwin Hutchins (see, canonically, Hutchins 1995, from where the example of ship navigation is taken). However, in using this introduction to sketch the conceptual background for the chapters that follow, we shall adopt an understanding of distributed cognition that arguably diverges somewhat from Hutchins’s own (for one thing, we make no demand that the target elements, whether located inside or outside the brain, should be understood as representational media; see e.g. Hutchins 1995: 373). Here we are aiming for a general and inclusive notion of cognition alongside a general and inclusive notion of what it means for cognition to be distributed. Thus the term ‘cognition’ should be understood liberally, as it routinely is in the day-today business of cognitive science, as picking out the domain of the psychological, where that domain encompasses phenomena that we often identify using terms  1

We warmly thank Douglas Cairns for his assistance with this chapter. The authors would also like to thank the participants in a workshop at the University of Edinburgh in June 2017 who provided feedback on an earlier draft of this chapter: Felix Budelmann, Peter Garratt, Christopher Gill, Elspeth Jajdelska, Karin Kukkonen, Adam Lively, Andrew Michael Roberts, George Rousseau, William Short, Jan Söffner, Eleanore Widger and Clare Wright.

2

m i randa a nder s on , michael w h eel er and m ark s p revak

such as mind, thought, reasoning, perception, imagination, intelligence, emotion and experience (this list is not exhaustive), and includes various conscious, unconscious-but-potentially-conscious, and strictly non-conscious states and processes. Given this broad conception of what cognition is, cognition may be said to be distributed when it is, in some way, spread out over the brain, the non-neural body and (in many paradigm cases) an environment consisting of objects, tools, other artefacts, texts, individuals, groups and/or social/institutional structures. Advocates of distributed cognition argue that a great many examples of the kinds of cognitive phenomena identified above (reasoning, perception, emotion, etc.) are spread out in this way. To see why the notion of distributed cognition has attracted so much attention, here’s a way of thinking about how the contemporary discourses stationed in and around cognitive science arrived at what might justifiably, in the present context, be called the received (non-distributed) view of mind. Although the very brief history lesson that follows involves the odd caricature, it is surely broadly accurate. According to the much-maligned substance dualists (the most famous of whom is arguably Descartes), mind is a non-physical entity that is metaphysically distinct from the material world. Here the material world includes not only the external tools and artefacts that human beings design, build and use, but also the thinker’s own organic body. On this model, the minds of other people become peculiarly inaccessible, and indeed one’s indirect knowledge of those minds, such as it is, seems to result from a precarious analogy with the correlations between thought and action in one’s own case. For this reason, plus a whole battery of others – some scientific, some philosophical – substance dualism is now officially unpopular in most of the relevant academic circles. Indeed, in the twentieth and twenty-first centuries, mind has been placed firmly back in the material and social world. Or rather, it has been placed firmly in a particular segment of that world, namely the brain. As apparently demonstrated by all those ‘pictures of the brain thinking’ that we regularly receive from functional magnetic resonance imaging (fMRI) scans and the like, the received view is now that the brain is where the cognitive action is. This neuro-centric orthodoxy is not an irrational position. Indeed, there is no doubt that many a good thing has come out of research programmes in psychology, neuroscience and elsewhere which embrace it. Nevertheless, the contemporary distributed cognition perspective is usefully depicted as a reaction against neuro-centrism’s (allegedly) distorting influence. To be clear, no advocate of distributed cognition believes that the brain is somehow unimportant. Rather, (part of) the proposal is that to understand properly what the brain does, we need to take proper account of the subtle, complex and often surprising ways in which that venerable organ is enmeshed with, and often depends on, non-neural bodily and environmental factors, in what is the co-generation of thought and experience. One consequence of adopting a general and inclusive notion of distributed cognition is that there turns out to be more than one version of the idea from which to choose when developing the view. How, then, may we articulate the notion



dis t rib uted c ognit ion and t h e h um ani t i es 3

further? One taxonomic move that is increasingly popular in the literature is to unpack distributed cognition in terms of 4E cognition, where the four Es in question are embodied, embedded, extended and enactive. In other words, it is possible to provide a more detailed picture of distributed cognition by thinking in terms of the four Es and the pattern of symbiotic and sometimes not-so-symbiotic relationships between them. That’s what we shall now do, starting with the notion of embodied cognition. According to the hypothesis of embodied cognition, psychological states and processes are routinely shaped, in fundamental ways, by non-neural bodily factors. In a full treatment of this idea, much more would need to be said about what the terms ‘shaped’ and ‘fundamental’ mean, but for present purposes the motivating thought will do: in order to understand cognition, the structures and forms of the non-neural body need to be foregrounded in ways that are absent from the neuro-centric orthodoxy. From this shared point of departure, the embodied cognition community has become home to a diverse kaleidoscope of projects. Thus embodiment is said to determine or condition the nature of concepts (e.g. Lakoff and Johnson 1980), the character of perceptual experience (e.g. Noë 2004), various factors such as orientation and posture that do not themselves enter into the content of experience, but which preconceptually structure that experience (e.g. Gallagher 2005), and the architectures, assemblages and processing mechanisms that enable intelligent action (in the philosophical literature, see e.g. Clark 1997, 2008b; Haugeland 1998; Wheeler 2005). As just one example of embodied cognition research, consider groundbreaking work in cognitive semantics on the role of embodiment in human sense-making (how we experience the world to be meaningful). Johnson (1987) argues that we experience our bodies fundamentally as three-dimensional containers into which we put things (e.g. food) and out of which things come (e.g. blood). The result is that the metaphor of containment becomes a preconceptual cognitive schema that heavily constrains other contexts of meaning. Thus, building on Johnson’s idea, Lakoff (1987) argues that the containment schema, as determined by our human experience of embodiment, even underlies abstract logical structures such as ‘P or not P’ (inside the container or outside of it). One apparent implication of this approach is that creatures with different experiences of embodiment will possess different preconceptual schemata and thus will inhabit different semantic landscapes. To the extent that embodiment is grounded in bodily acts, such as, say, the physical manipulations of instruments or tools, embodiment naturally encompasses a rich mode of environmental interaction, which is just to say that there is a natural route from embodied cognition to the second of the four Es, namely embedded cognition. According to the embedded view, the distinctive adaptive richness and/ or flexibility of intelligent thought and action is regularly, and perhaps sometimes necessarily, causally dependent on the bodily exploitation of certain environmental props or scaffolds. As an illustration, consider the phenomenon that Andy Clark has dubbed cognitive niche construction (e.g. Clark 2008b; see also Wheeler and Clark 2008). This occurs when human beings build external structures that, often in

4

m i randa a nder s on , michael w h eel er and m ark s p revak

combination with culturally transmitted practices, transform problem spaces in ways that promote, or sometimes obstruct, thinking and reasoning. A compelling example, which Clark sources from Beach (1988), is the way in which a skilled bartender may achieve the successful delivery of a large and complex order of cocktails (a relatively daunting memory task) by exploiting the fact that different kinds of cocktail often come in differently shaped glasses. Bartenders learn to retrieve the correct glass for each drink as it is requested, and to arrange the differently shaped glasses in a spatial sequence that tracks the temporal sequence of the drinks order, thus transforming a highly challenging memory task into a simpler (roughly) perception and association task. This reduces the burden on inner processing by exploiting a self-created environmental structure according to a culturally inherited social practice. Of course, as indicated in the definition given of cognitive niche construction, not all cases of the environmental scaffolding of cognition will result in enhanced performance. The background picture here is of ‘our distinctive universal human nature, insofar as it exists at all, [as] a nature of biologically determined openness to deep, learning- and development-mediated, change’ (Wheeler and Clark 2008: 3572) and thus, given a technologically saturated environment, of human organisms as what Clark (2003) calls natural born cyborgs, creatures who are naturally evolved to seek out intimate unions with non-biological resources. Overall, the ongoing operation of this evolved tendency has yielded myriad adaptive benefits, but sometimes the couplings that result will be adaptively neutral, inappropriate or dysfunctional. This observation points to an important vein of research on how ideas that are central to distributed cognition can contribute to areas such as psychopathology (e.g. Gallagher 2004; Fuchs 2005; Drayson 2009; Sprevak 2011). Despite the fact that the embedded theorist seeks to register the routinely ­performance-boosting, often transformative, sometimes necessary, but occasionally obstructive, causal contributions made by environmental elements (paradigmatically, external technology) to many cognitive outcomes, she continues to hold that the actual thinking going on in such cases remains a resolutely skin-side phenomenon, being either brain-bound or (on a less common, more radical iteration of the view) distributed through the brain and the non-neural body. By contrast, according to the advocate of extended cognition, it is literally true that the physical machinery of mind itself sometimes extends beyond the skull and skin (see, canonically, Clark and Chalmers 1998; for a collection that places the original Clark and Chalmers paper alongside a series of criticisms, defences and developments, see Menary 2010). More precisely, according to the hypothesis of extended cognition, there are actual (in this world) cases of intelligent thought and action, in which the material vehicles that realise the thinking and thoughts concerned are spatially distributed over brain, body and world, in such a way that certain external factors are rightly accorded fundamentally the same cognitive status as would ordinarily be accorded to a subset of your neurons. Thus, under the right circumstances, your mobile phone literally counts as part of your mnemonic machinery, alongside some of your neurons.



dis t rib uted c ognit ion and t h e h um ani t i es 5

To bring home the distinction between embedded and extended cognition, as we have just introduced it, consider the example of a mathematical calculation achieved, in part, through the bodily manipulation of pen and paper. For both the embedded and the extended view, what we have here is a brain-body-penand-paper system involving a beyond-the-skin element that, perhaps among other things, helps to transform a difficult cognitive problem into a set of simpler ones (e.g. by acting as storage for intermediate calculations). For the embedded theorist, however, even if it is true that the overall mathematical problem could not have been solved, at least by some particular mathematician, without the use of pen and paper, nevertheless the external resource in play retains the status of a non-­cognitive aid to some internally located thinking system. By contrast, for the advocate of the extended view, the coupled system of pen-and-paper resource, appropriate bodily manipulations, and in-the-head processing may itself count as a cognitive architecture, even though it is a dynamically assembled (rather than hard-wired) and essentially temporary (rather than persisting) coalition of elements. In other words, each of the differently located components of this distributed (over brain, body and world) multi-factor system enjoys cognitive status, where the term ‘cognitive status’ should be understood as indicating whatever status it is that we ordinarily grant the brain. Here it is worth pausing to note that, in the distributed cognition literature, one can certainly find the term ‘extended cognition’ being given a less specific reading than we have just suggested, a reading which is tantamount to the interpretation we have adopted here of the term ‘distributed cognition’, and which thus encompasses embedded cognition and (at least many forms of) embodied cognition. This liberal usage is not negligent. For one thing, the boundary between internal and external is, in some contexts, fixed by the skin – in which case gross bodily forms count as internal – while in others it is fixed by the limits of the brain or central nervous system – in which case gross bodily forms count as external. On the latter view, at least some forms of embodied cognition would count as cases of extended cognition. For another thing, given certain projects and purposes, the distinction between being a non-cognitive but performance-boosting scaffold and being a genuine part of one’s mental machinery may be a distraction, even if it is metaphysically legitimate. Nevertheless, it does seem clear that if one uses the term ‘extended cognition’ in the more inclusive way, one will need to find a different term for the case of what we might identify as metaphysical or constitutive extension (‘strictly extended’ maybe). Otherwise one will risk succumbing to what extended cognition theorists call cognitive bloat, an undesirable outcome in which one is forced to concede all sorts of mundane and unexciting cases of causal coupling between inner and external elements to be cases of extended cognition, thus generating a wildly counter-intuitive position. It looks, then, as if there is a genuine argument to be had over whether it is possible to make the transition from embodied-embedded cognition to extended cognition. And, indeed, this is a complicated and contested area (to sample just a small subset of views and the sometimes ill-tempered debate, see e.g. Rowlands 1999, 2010; Menary 2007, 2010; Adams and Aizawa 2008; Clark

6

m i randa a nder s on , michael w h eel er and m ark s p revak

2008b; Rupert 2009; Sprevak 2009; Sutton 2010; Wheeler 2010). And it is possible that there won’t be a universal resolution. That is, it may be that while some cognitive phenomena reward an extended treatment (leading candidates might include memory, reasoning and problem-solving), others will not. There is, for example, an ongoing debate over the credentials of extended consciousness (Hurley and Noë 2003; Noë 2004; Clark 2009; Hurley 2010; Ward 2012; Wheeler 2015). Our final ‘E’ is enactive. In the most general terms, a position is enactivist if it pursues some version of the claim that cognition unfolds (is enacted) in looping sensorimotor interactions between an active embodied organism and its environment. For the enactivist, then, cognition depends on a tight and dynamic relationship between perception and action. Enactivism also tends to foreground the disciplined examination of lived experience as a methodological tool in cognitive theory. This leads many enactivists to draw on the phenomenological philosophical tradition, as represented centrally by thinkers such as Husserl, Heidegger and Merleau-Ponty, a tradition which concentrates on the structures of, and the conditions for, lived experience. This productive engagement with phenomenology is especially prominent in relation to the enactivist understanding of the body not simply as a physical mechanism, but as a lived structure though which the world is experienced. (Although enactivism foregrounds phenomenology more so than the other branches of distributed cognition, that is not to say that it has a monopoly on phenomenology’s insights and conceptual machinery; see e.g. Gallagher 2005; Wheeler 2005; Zahavi 2014 for essentially non-enactivist yet systematic appeals to phenomenology, in and around the distributed cognition literature.) The two most common forms of enactivism are sensorimotor enactivism (e.g. O’Regan and Noë 2001a; Noë 2004) and autopoietic enactivism (e.g. Varela et al. 1991; Di Paolo 2005; Thompson 2007b), although another recent, and increasingly important, variant that we shall not discuss here is the so-called radical enactivism of Hutto and Myin (2012). Sensorimotor enactivism is rooted in the thought that perceptual experience is constituted by implicit knowledge of so-called sensorimotor contingencies – the law-like effects that either my movement or the movements of objects in my sensory field have on the sensory input that I receive – where the implicit knowledge in question is to be understood in terms of the possession and exercise of certain bodily skills. Thus consider my visual perception of a tomato. Although my visual access to that entity is aspectual (there is an obvious sense in which, given my embodied spatial perspective, I have visual access only to certain portions of it), my ordinary experience is nevertheless of the tomato as an intact, solid, three-dimensional object. As one might put it, the tomato’s hidden-from-­ perspectival-view aspects are nevertheless experientially present to me. According to the sensorimotor enactivist, this is explained by my implicit mastery of the relevant sensorimotor contingencies – very roughly, the visual inputs my eyes would receive if I moved around the tomato, or if I turned it, or if it span round. This implicit sensorimotor knowledge is constitutive of my perceptual experience. Autopoietic enactivism is based on the idea that cognition is a process of sense-making by adaptively autonomous systems, where an autonomous system



dis t rib uted c ognit ion and t h e h um ani t i es 7

is a network of interdependent processes whose recurrent activity (a) produces and maintains the very boundary that determines the identity of that network as a unitary system, and concurrently (b) defines the ways in which that system may encounter perturbations from what is outside it while maintaining its organisation. A system is adaptively autonomous when it is able to alter its behaviour in response to changes in its environment in order to improve its situation, for example by sensorimotor activity. To illustrate this with an example that the autopoietic enactivists themselves often use, bacteria sense and swim towards the environmental area containing the greatest concentration of glucose molecules. Thus, as a consequence of the specific metabolically realised autonomy of the bacteria, glucose emerges as – is brought forth as – significant for those organisms as food. As this example nicely illustrates, autopoietic enactivism is distinctive in forging a close connection between life and mind (cognition). As Thompson (2007b: 128) puts it: ‘life and mind share a set of basic organizational principles, and the organizational properties distinctive of mind are an enriched version of those fundamental to life. Mind is life-like and life is mind-like.’ One effect of enactive life-mind continuity is to place affective phenomena such as emotions and moods at the very centre of the cognitive stage. As the glucose example highlights, enacted significance is fundamentally a matter of valence, that is, of being appropriately attracted and repelled by environmental factors that might improve or diminish organisational integrity. In this way, these factors are things that the organism cares about. This is the enactive root of affectivity. (For a developed enactive account of emotional episodes as self-organising patterns of the whole embodied organism, see Colombetti 2014.) More generally, a 4E-friendly treatment of affective phenomena will tend to reject a commonly held view in the psychology of emotions according to which there is a neat distinction between the cognitive components of emotions (e.g. the appraisal of a situation in relation to one’s well-being) and their bodily components (e.g. arousal and facial expressions). For the advocate of embodied emotions, appraisal is itself a phenomenon that is spread out over both neural and non-neural bodily factors (see, again, Colombetti 2014). In the background here is Damasio’s (e.g. 1999) influential notion of somatic markers, i.e. specific feelings in the body that accompany specific emotions (e.g. nausea with disgust) and which strongly shape subsequent decision-making. A more controversial application of 4E thinking in the vicinity of affective phenomena is the claim that such phenomena may be extended beyond the skin of an individual, either over artefacts such as musical instruments (see e.g. Colombetti and Roberts 2015) or over other people (see e.g. Slaby 2014; Krueger and Szanto 2016). This final point brings us neatly to the issue of the social dimension of the 4E mind. Consider three possible ways in which cognition might be socially distributed: 1. I think some of the thoughts I think because, or perhaps only because, I am part of a particular social group. 2. My cognitive states or processes are socially as well as technologically extended,

8

m i randa a nder s on , michael w h eel er and m ark s p revak

such that some of my cognitive machinery is located partly in the brains of other people. 3. Groups may have minds in much the same way that individuals have minds. Option 1 is perhaps most naturally understood as an embedded view according to which some psychological capacities realised wholly by neural states and processes are nevertheless manifested only in certain kinds of social circumstances, because they are socially scaffolded by those circumstances. Option 2 is the social version of the hypothesis of extended cognition. Consider Tom and Mary, a couple with a long and interdependent relationship. Perhaps Tom might come to trust, rely on and routinely access information stored in Mary’s brain, in such a way that, in certain contexts, her brain comes to play essentially the same role as his own neural resources, and thus constitute a repository of his memories (cf. Clark and Chalmers 1998). Both option 1 and option 2 may count as versions of what Wilson (2005) calls the social manifestation hypothesis, which maintains that cognition remains a ‘property of individuals, but only insofar as those individuals are situated or embedded in certain physical environments and social milieus’. By contrast, option 3 shifts the ownership of the relevant psychological states and processes from the individual to the group. According to 3, the group mind hypothesis, we should take at face value statements such as ‘the team desires a victory’ and ‘the crowd thinks the game is over’. Whole groups, and not merely the individuals out of which those groups are constructed, may non-metaphorically be attributed with beliefs, desires, other psychological states and processes of reasoning (for versions of this view, see e.g. Huebner 2014; Tollefsen 2006). As one might imagine, much of the philosophical debate in this area concerns the conditions and circumstances under which it would be correct to adopt either option 2 or option 3, both of which might, to our modern thinking at least, seem counter-intuitive or, from a theoretical perspective, metaphysically profligate (for an argument for the latter conclusion, see e.g. Rupert 2014). But even the seemingly less radical option 1 constitutes a prompt for a careful examination of the distributed causal mechanisms and social contexts that drive the processes concerned. Two further modulations of social cognition, when developed in a 4E register, bring out a final way in which the distributed perspective changes the shape of the received theoretical terrain. It has become a commonplace view in the psychology of social cognition that we navigate our social spaces either by predicting and explaining one another’s observable behaviour using what is tantamount to a commonsense theory of the hidden inner causes of that behaviour – the beliefs, desires and other mental states inside other people’s heads – or by predicting and explaining that behaviour by internally simulating what we ourselves would think and experience in the same circumstances, in order to produce the same behaviour. On either of these models, and partially echoing the Cartesian dualism of days gone by (see above), our access to the mental states of the other person is fundamentally indirect, involving inference or simulation. However, some distributed cognition theorists (e.g. Gallagher 2008), drawing centrally on phenomenology, argue that



dis t rib uted c ognit ion and t h e h um ani t i es 9

our grasp of the mental states of another person with whom we are in perceptual contact is ordinarily direct, in that, rather than, for example, inferring another person’s joy from some of her facial movements, we see that joy directly, in her laughter. Relatedly, but moving away even further from the conventional assumption that a theory of social cognition should begin with a conception of people as isolated spectators, a distinctively enactive account of social cognition has emerged in the form of the participatory sense-making approach (De Jaegher and Di Paolo 2007). This approach begins with a conception of people not as spectators, but as engaged participants, and focuses on the ways in which individuals interactively coordinate their movements and utterances in social situations. For participatory sense-making theorists, the interaction process itself can come to exhibit an enactive form of autonomy (see above), one that is generated by, but also constrains and scaffolds, the activity of the interacting individuals. The alleged directness of social cognition is grounded in the fact that we are so proficient at social interaction that the interactive process becomes transparent to us (De Jaegher 2009). This section has surveyed views that fall under the general heading of ‘distributed cognition’. What these views have in common is that they accord some kind of special role to the environment (perhaps including the bodily environment or social environment) that is missing in the traditional Cartesian or in more recent neuro-centric cognitive theories. More to the fore in this introduction, however, is the diversity of views within the distributed cognition camp. We have seen that distributed cognition can roughly be split into 4 Es (embodied, embedded, extended and enactive) with each ‘E’ admitting of further, sometimes competing, articulations in the hands of different philosophers and cognitive scientists (we have already met two kinds of enactivism). Philosophers and cognitive scientists who work in this area often adopt what might appear to be a mix-and-match approach: they accept some ‘distributed’ claims but reject others. Moreover, they may accept a different combination of distributed (or non-distributed) claims about different parts of human mental life: some aspects of our mental life (perhaps our feelings of joy and pain) may be treated as purely internal while others (perhaps some of our memories and decision-making) are distributed. A further complication is that, even if attention is restricted to a single aspect of human mental life, a distributed theorist may give a different distributed/non-distributed answer for different human subjects in different environmental contexts at different times (some humans are more inclined than others to distribute their memory on to external devices). How should researchers in the humanities make sense of all this disagreement and diversity within the distributed cognition camp? We suggest that they cut through philosophical disagreements and explore the specific combination of distributed views that suits their interests. An application of distributed cognition in the humanities should be assessed on its own merits. A specific combination of ‘distributed’ views may prove more or less fruitful to understanding a particular historical phenomenon. And different combinations of ‘distributed’ views may prove suited to different historical phenomena. There is no reason why a single, one-size-fits-all approach should be adopted.

10

m i randa a nder s on , michael w h eel er and m ark s p revak

The merits of a particular combination of distributed views, in a particular historical setting, should be based on its pay-off for our understanding in the humanities. We hope that the essays in the volume demonstrate both the value and the diversity of conceptual tools offered by a distributed cognition approach.

Distributed Cognition and the Cognitive Turn The question of how new insights into the nature of the mind illuminate our understanding of being human and our engagement with the world is one of the greatest issues facing the current generation of humanities scholars. Knowledge that is emerging from cognitive science and neuroscience, along with related research in disciplines such as philosophy, psychology and linguistics, casts a new light on issues that are central to the humanities, and enables us to better explain the nature of forms of human culture and how and why they emerge and evolve. Knowledge from the sciences can help to make a case for culture’s significance to being human. In turn, the humanities provide an archive of examples concerning how humans develop in a range of environments and evidence diverse ways in which we use and create resources as means to extend our capacities. This section provides a general overview of the emergence of the cognitive humanities and considers how distributed cognition interrogates and supplements existing humanities methodologies. In spite of the humanities’ increasing interdisciplinarity since the late 1950s, the long-standing disciplinary antagonism between the arts and sciences continues (Snow [1959] 1993; Leavis 1962; Ortolano 2005). This tension is now further exacerbated by wider movements towards education’s commodification, quantification and rebranding as merely employment-based training, with accompanying questions about the relevance and value of the arts and humanities (Collini 2009; Nussbaum 2010). The defensive response by many in the humanities has been to fall back on claims about the qualitative and irreducible nature of aesthetic values in the humanities and to view humanities scholars who draw on scientific knowledge as reductionists.2 The presumption is that scientific knowledge entails a constraint on, and a diminution of the value of, their own complex material and methodologies, rather than adding a further perspective. Yet, even for the hard problem of consciousness – that is, the way it feels to be me or you – philosophers draw on insights from neuroscience. There is no reason why questions about the qualitative nature of our experience of cultural artefacts cannot be addressed, even if not solved outright, by neuroscientific insights. Moreover, a broad-based approach that draws not just on the mind’s neural basis, but on the whole range of the human mind as evidenced in cultural texts and practices, seems likely to provide us with the fullest insights into these enduring questions. Phenomenology has influenced both philosophers focusing on culture or lan 2

As Helen Small describes, ‘the humanities value qualitative above quantitative reasoning’ (2013: 57). Raymond Tallis (2011) provides an example of the anti-scientific bent with his dismissal of a focus on the physical aspects of being human as ‘neuromania’ or ‘Darwinitis’.



dis t rib uted c ognit ion and t h e h um ani t i es 11

guage (so-called ‘continental’ philosophers, such as Foucault, Derrida and Lacan) and models (in particular enactivism) that attempt to conceptualise scientific evidence of distributed cognition. However, in cultural studies phenomenological ideas became entangled with a tendency towards relativism, as part of a backlash against earlier humanist or structuralist notions of culture as revealing universal human attributes and values. Throughout history, oscillating drives towards universalism and relativism often culminate in polarised models, or one extreme sets the pendulum swinging back to the other extreme. In recent decades literary, historical and cultural criticism have focused on various kinds of postmodern relativism and social constructivism which resist anything interpretable as ‘facts’, ‘truth’ or ‘reality’ and in which human bodies are presented as merely cultural constructs (the tendency is notable in new historicism, cultural materialism and feminist, queer and globalisation studies). From classical antiquity onwards, there have been those who have questioned the extent of our access to a mind-independent world: from Plato’s shadow-watchers in the cave, to Descartes’s sceptical cogito ergo sum, to more recent thought experiments suggesting that our experience might be the same were we brains in a vat, to Bayesian predictive coding models, which some internalist-minded philosophers argue are a basis for assuming that cognition is skull-bound, so that ‘conscious experience is like a fantasy or virtual reality’ (Hohwy 2013: 137). Early artificial intelligence and cybernetics influenced both cognitive scientists and the continental philosophers who in turn have informed cultural theory. Lacan, for instance, discussed the influence that cybernetics had on him, though in his dark version, humans become the processed rather than the processor: ‘It is the world of words which creates the world of things . . . Man speaks therefore, but it is because the symbol has made him man’ (Lacan 1981: 39). Higher-level shifts in norms of understanding and focus cascade down, and so shape and are reshaped by disciplines into idiosyncratic manifestations of the higher-level propensities into which (along with many other factors) they feed back. The linguistic turn in the humanities in the 1970s and 1980s argued that language consisted of a system of codes, with words caught up primarily in their relation to other words, with a consequent endless deferral of meaning and a disconnectedness to referents in the world. At the same time, classical cognitive science described cognition as occurring through computational manipulation of internal symbols. Such theories emphasised the role of arbitrary, abstract syntactic structures at the expense of attention to the emergence of meaning through our engagement in the world. More recently, Daniel Dennett suggested that ‘A scholar is just a library’s way of making another library’ (1991: 202). This may seem to relate straightforwardly to Lacanian-type claims. Dennett himself has commented on the correspondences between these ideas and postmodern deconstructionism (410). However, Dennett reaches his suppositions through biologically grounded or organically inspired notions, such as memes, an idea developed by Richard Dawkins (who coined the term) to describe units of cultural transmission akin to genes as units of biological transmission (202–3). In contrast to such biologically grounded accounts,

12

m i randa a nder s on , michael w h eel er and m ark s p revak

­ revailing ­postmodern methodologies have argued that sociocultural forces are p entirely responsible for human concepts and behaviour. For example, the elision of the physical body is evident in Judith Butler’s claim that ‘gender is a performance that produces the illusion of an inner sex’ (1998: 728). Whilst such accounts make visible sociocultural influences on biological categories, their rebuttal of the hegemony of the natural world and of the fixed nature of biological categories is problematic to the extent that it simply inverts the relationship and asserts the dominance of sociocultural forces over the realm of the natural and physical. The significant difference between a postmodern stance and Dennett’s viewpoint is highlighted by his concluding comment that ‘I wouldn’t say there is nothing outside the text. There are for instance all the bookcases, buildings, bodies, bacteria . . .’ (1991: 411). Embodied cognition – that is, the notion that our physical bodies enable and constitute cognitive processes – presents a greater challenge to postmodern accounts than do claims about its extended nature. Furthermore, distributed cognitive models make evident that the mind’s embedded or extended nature is not simply a matter of unconstrained cultural determinism. The elision of the physical body and world in postmodern accounts helps to perpetuate an apparent conflict between the arts and sciences that risks miring notions of being human (and the humanities) in isolated idealism. The recent rise of digital humanities, despite its use of distant reading and quantitative analyses,3 might be explained in part by technological advancements. These advancements massively increase the ways in which we can creatively use programs and technologies to inform and study our interaction with texts and artefacts. However, its rise may also be due to a theoretical tendency towards the virtual and virtualisation, with an over-emphasis on the distinctiveness of human nature as arising from meta- or trans-physical capacities, a tendency which has been evident throughout history in idealist accounts of human nature (Hayles 1999; Nusselder 2009). As part of a countermovement in the humanities today, there is a shift towards the study of material culture, with a renewed focus on specific physical objects, physical environments and ecological contexts. We also find elision, if not of the physical body as a whole, then at least of the significance of the specifics of it, in some more functionalist accounts of the extended mind. Andy Clark accuses thinkers such as Damasio of biochauvinism because of the essential role they ascribe to the body in cognition. Instead, Clark argues, ‘the perceptual experience of differently embodied animals could in principle be identical, not merely similar to our own’ (2008b: 193). Clark’s functionalist approach allows for the possibility that a non-biological resource can play the same role as a biological one. Yet, elsewhere, Clark emphasises that external resources need not be functionally identical to internal ones to qualify as extended (1997:  3

Franco Moretti coined the term ‘distant reading’ to describe the need for a world-scale study of literature, which would draw on other researchers’ scholarship, in contrast to the conventional close reading of a small number of canonical texts (Moretti 2000). In his later work, and more generally, the term is now often used to describe the study of literature through data analysis or text-mining of large-scale corpora (Moretti 2013).



dis t rib uted c ognit ion and t h e h um ani t i es 13

220): a laptop does not store or compute information in the same way as a brain, and it can, for that reason, be useful in supplementing our neural capacities. On Clark’s view, there is a question about how much the material nature of a resource matters to how it fulfils its function, and this is important for how we evaluate the significance of different resources – neural, bodily and non-biological ones – in different historical contexts. Our view is that in certain contexts, or while performing certain functions, a difference between human bodies or physical resources may matter; in other cases, it may not be significant. The richer idea that emerges from this perspective is that through differences, as well as similarities, various forms of representational, computational and mnemonic resources can supplement our biological limitations. We have seen that distributed cognition presents an array of competing and sometimes conflicting theories. This may appear a sign of fragility, but this diversity can be seen as reflecting the ways in which different cognitive models come to the fore in relation to different mental capacities or contexts. Shifting trends and debates about cognitive hierarchies, such as an emphasis on the role of embodiment or on particular methods or resources as extending representational or phenomenological possibilities, can be seen to emerge in relation to the development of, or reaction against, new genres, cultural modes and technological, scientific and sociocultural changes. Therefore, the multifaceted nature of distributed cognition as a theory is, in our view, a strength that reflects the operation of different cognitive norms and modes in the world. From the late 1980s and 1990s onwards a few ‘first-wave’ thinkers in the cognitive humanities, primarily based in the US, adopted notions from evolutionary psychology or from cognitive linguistics that emphasised the universal aspects of humans’ cognitive and physical characteristics.4 A particularly influential idea was that humans tend to conceptualise non-physical domains in terms of physical ones, and in contrast to postmodern deconstructionism, this suggested a specific way in which language is embodied. Engagement with these ideas in the humanities challenged existing disciplinary divisions and opened up new ways to think about what human beings have in common and how humans from diverse cultures can have some understanding of one another. Yet both evolutionary-psychological and cognitive-linguistic humanities approaches tend to operate without due attention to the many historical (and geographical) variables involved in cultural, linguistic and literary constructions. Such early cognitive humanities scholars therefore tended to set postmodern and cognitive approaches in irreconcilable opposition to one another. This risked simply repeating within the humanities the persisting methodological tensions between the arts and sciences through the siding of critics with oppositional explanatory paradigms. However, these first-wave thinkers have remained largely on the peripheries of mainstream literary and cultural methodologies, against which some of them tend to situate themselves. Yet even from the early stages, cognitive humanities scholars drew attention to  4

See, for example, Lakoff and Johnson 1980; Carroll 1994; Gottschall and Wilson 2005.

14

m i randa a nder s on , michael w h eel er and m ark s p revak

the benefits of engaging with scientific work on the mind, and not all their efforts were so oppositional. One of the first scholars to adopt a cognitive approach, which he argued was compatible with the more conventional approaches he was already using, was the psychoanalytical literary critic Norman Holland (1988). Most other early adopters of scientific work, models and metaphors as a means to inform our reading of texts in the emerging field of ‘cognitive poetics’ (such as Tsur 1992; Gibbs 1994; Turner 1991, 1996) were influenced by and developed cognitive linguistics’ notion of conceptual schema and argued for the everyday nature of figurative thought and language. There were, however, already outliers such as Ellen Spolsky, who was instead focused on reapplying the then fashionable theory of the modular mind to her analysis of literary texts as a means of engaging with poststructuralism (1993). Distributed cognition suggests another perspective to universalising models, one that takes account of the way in which embodiment is a crucial aspect of our extendedness, since it is our biological nature that enables us to incorporate sociocultural and technological resources into our cognitive systems. The significance of the human body arises also from its capacity for engagement and interaction. The dynamic cognitive roles of linguistic, sociocultural and technological resources are made possible by neurological plasticity.5 This raises the possibility that social constructivist models may have a neurological basis, as our ability to be (at times transparently) constructed by sociocultural forces relates to the adaptability of the human brain. At the same time, human adaptability and extendedness temper any notion of universal communal features shared across all humans that could be based on embodiment. While humans exhibit certain enduring biological characteristics, these characteristics dynamically interact with our sociocultural and environmental contexts. Together these lead to the manifestation of different kinds of minds and to the expression or suppression of particular forms of cognitive paradigms. This perception enables a reassessment of polar representations of the mind as either fixed and universal or as socially constructed and culturally relative – two models which have constrained understandings of historical, as well as modern, concepts of the mind. Humans are particularly talented not just at evolutionarily adapting to niches, but also at adapting our niches to supplement our cognitive and other needs (Wheeler and Clark 2008). While humans’ capacity to exist within cognitive niches, with ongoing reciprocal interactions between niches and organisms, is shared across generations, these niches also reflect technological and sociocultural developments; ultimately, rather than either universalism or relativism, this implies that we shall find a rich combination of shared features and particular divergences across history and cultures. The paradigm of distributed cognition provides a middle way between relativism and universalism by highlighting the vital roles played by both physical and  5

For just a few works that discuss neurological plasticity and adaptivity, see Ramachandran and Ramachandran 1996; Berti and Frassinetti 2000; Maravati and Iriki 2004; Gallagher 2005; Rizzolatti and Sinigaglia 2008; Clark 2008b.



dis t rib uted c ognit ion and t h e h um ani t i es 15

cultural resources in cognition. As a methodology, this provides a potential strategy for making headway in the science wars. In the conflict between radical postmodern relativism and science-based realism, the question is whether facts about the world are merely culturally determined or whether those facts are grounded in some mind-independent reality. Analysis of cognition needs to take into account not only the findings of current cognitive science, but also the imagery and narratives used in scientific, cultural and literary discourses. Cultural factors play an essential role in shaping the world, and acknowledging this should not require rejecting or devaluing the role of science. The influence of culture and its role in cognitive niche construction shapes even our disciplinary taxonomies – constantly evolving scaffolds that accrete knowledge about specific domains. A combination of scientific and cultural knowledge is necessary to understand the nature of cognition and of being human in the world. Distributed cognition is a theoretical framework that enables one to grasp the mutual entanglement of science and culture. The importance of distributed cognition as a methodological approach grows out of its rootedness in bodies of knowledge from across the disciplinary spectrum, which together are reflective of the full scope of human nature. During the period in which our project has taken place, second-wave thinkers in the cognitive humanities have begun to consider a more diverse range of approaches to the mind; the expansion of recent work in the field is discussed in each of the period-specific introductions to the volumes. Our project sought to strengthen and stimulate increased interest in distributed cognition. The current book series provides a rigorous engagement with these ideas by humanities scholars and demonstrates that distributed cognition can helpfully illuminate cultural studies. Our aim is to inspire a broader re-evaluation in the humanities of what is understood to count as cognition and to suggest a new way of doing intellectual history. The four current volumes on distributed cognition examine the practices and the explicit and implicit conceptual models that were in use from antiquity to the early twentieth century. Together the essays trace across Western European history developments and divergences among various concepts of distributed cognition that were circulating. The essays engage with recent debates about the various models of distributed cognition and bring these into discourse with research in the humanities through examination of the parallels to (and divergences from) these models and these debates in earlier cultural, philosophical and scientific works. The essays make evident some of the explicit and implicit grounds on which our present suppositions about cognition stand and also some of the knowledge and insights, as well as superstitions and ungrounded beliefs, that have been lost and obscured along the way. The reiterations and the diversity in expressions and practices of distributed cognition enrich and enlighten our understanding of wider forces and rhythms in the history of cognition. One important point that emerges is that notions of cognition can be shown to be fundamental to how we conceptualise debates in every discipline – the study of cognitive phenomena cannot be considered a specialist niche, but is rather a necessary underpinning of any study of humans in the world. This series bears out the premise that ‘current philosophical

16

m i randa a nder s on , michael w h eel er and m ark s p revak

notions [of distributed cognition] are simply the most recent manifestation of an enduring paradigm that reflects the non-trivial participation of the body and world in cognition’ (Anderson 2015: xi). In return for providing a scientific basis to our understanding of the mind in the humanities, historical studies have the potential to feed back and interrogate our current philosophical and scientific understanding of how cognition may be distributed across the body and the world. The historical lineage of non-brain-bound concepts of cognition demonstrates that such ideas are not merely a product of our own age. Cultural beliefs and philosophical interpretations map on to underlying physical features and processes in ways that function practically within a society. The volume introductions provide detailed overviews of the development of the cognitive humanities in relation to the periods they cover, but to briefly outline how distributed cognition can illuminate cultural interpretation by challenging models that view the body or the environment as peripheral to understanding the nature of cognition, we shall present a few instances that arise from our collaboration with the National Museum of Scotland. Distributed cognition raises questions about the nature and role of galleries and museums, demonstrating ways in which many of the artefacts extend either human cognitive or other physical capacities. There are some (perhaps obvious) examples of distributed cognition, such as the tally stick, which was a piece of bone or wood scored across with notches that was used, from around 30,000 years ago, to record numbers or messages. This was an ancient memory aid. Like the oft-cited modern example of an iPhone, it remembers, so you need not. Basic tools controlled by bodily action gave way to mechanised tools that outsourced mechanical processes, through devices such as steam engines, and to automated and programmable agents, such as the Jacquard Loom. The Jacquard Loom not only moved faster and more reliably than a human weaver, it took over the weaver’s cognitive load and allowed greater design complexity than would ordinarily be available from the average human weaver’s brain alone. In a trajectory familiar to those who work on the history of the book, a museum also makes evident the shift from oral traditions to literacy through the preservation of early memorial stones carved first with only images, then manuscripts which enabled more detailed storage, manipulation and communication of information, then printed books and presses that enabled the sharing of information on a larger scale, and finally we emerge into the modern world of computers and the internet. Encounters with museum artefacts involve both a conceptual encounter with an object’s caption and an experiential encounter with the object itself – a metal helmet may automatically trigger a sense of weightiness (Bolens 2017). Yet, in some instances the contingent nature of one’s tacit knowledge may affect the extent to which artefacts prompt a kinesic or kinaesthetic response, with the cognitive capacity to simulate holding, wearing or interacting with an object relating partly to prior embodied and cultural experience, with more conceptual scaffolding (for example, via illustrations of past uses) needed for more obscure artefacts. Similarly, devices that may seem intuitive in one period often require significant amounts of cultur-



dis t rib uted c ognit ion and t h e h um ani t i es 17

ally embedded knowledge that belie their apparent simplicity (Phillipson 2017). Distributed cognition invites a broad spectrum of multidisciplinary approaches, enabling a richly diverse appreciation of the reciprocal ways in which artefacts and humans have shaped each other. Distributed cognition creates a scientific basis for understanding the fundamental significance of culture to humans and the humanities. Describing perception, Alva Noë says, ‘We continuously move about and squint and adjust ourselves to . . . bring and maintain the world in focus’ (2015: 9). When we read literature, view art or engage with historical artefacts in a museum, we remain linked to our own particular ever-shifting perspective and yet these cultural resources allow us to experience the world beyond our usual cognitive range. Each genre, each author or artist, and each work provides distinct forms of cognitive mediation. It does this in a way that reflects back on ourselves and the world around us, at the same time as it recalibrates and adds to the numerous virtual coordinates through which we more generally orient ourselves and enact our worlds. Texts, artworks and other cultural artefacts are imbued with mind, the mind of their creator and their context, and that of the spectator, reader or interactor. Objects, images and language, particularly those in consciously crafted literary and art works, provide catalytic scaffolding for perceptual flights into and beyond the usual constraints of our own imaginations, and can trigger a rich array of responses that are grounded in and recalibrate our emotional, physical and cultural natures, extending and revitalising our mental panoramas. Our series questions assumptions that have been made about historical notions of the mind, begins to trace the lineages of ideas about the mind across periods and cultures, and highlights the ways in which certain aspects of the mind come to the fore in certain contexts and traditions. The realisation of the distributed nature of cognition, which can induce both mind-forged manacles and mind-extending marvels, upholds the role of the humanities in wider society and more broadly challenges humans’ ways of being in the world. The extent of our capacity to extend our minds, across our current sociocultural panorama and physical world, and via the cognitive scaffolding provided by earlier generations, places in question the relatively short-term individualistic ends that are currently being prioritised and endorsed in most modern societies. The value of the humanities in concert with the sciences is their capacity to extend our cognitive range beyond everyday constraints, by scaffolding critical thinking and enabling our minds to soar to the heights needed to tackle world-sized and epic-scale issues, as well as to supplement our ability to grasp more fully the diversity of other minds. Distributed cognition invites a more inclusive approach, which acknowledges that experience and understanding of the world and of the humanities is multifaceted and involves biological and sociocultural dimensions. Distributed cognition offers a reconsideration of the nature of the human mind, and so of being human, and, in turn, of the humanities.

18

dou glas c ai rns

2 D i s t r i but e d C o g n i t i on a nd the C la s s ics Douglas Cairns

The purpose of this chapter is to provide a background to current research in Classics on topics related to distributed cognition and to consider how the various chapters in this volume represent, reflect and advance work in this area. This volume brings together eleven chapters by international specialists in the historical period from archaic Greece to late antiquity, with the majority of essays focusing on the period from the fifth century bce to the third century ce. It includes essays on the ways in which cognition is explicitly or implicitly conceived of as distributed across brain, body and world in Greek and Roman technology, culture, science, medicine, philosophy, art, literature and drama. Together the chapters make evident the ways in which the particular range of sociocultural and technological contexts that existed in antiquity fostered and reflected particular notions of distributed cognition. In what follows I attempt to put these contributions in their wider research context, in terms of (a) mainstream classical scholarship, (b) earlier work in Classics which draws on the theories and findings of the cognitive sciences in general, and (c) existing applications of specific aspects of distributed cognition theory to classical material. In each of these sections I freely intersperse references to the chapters in this volume, where they can be seen as taking forward the issues in question or where they provide further information on topics canvassed here, before turning, in the concluding section, to the chapters themselves.

Distributed Cognition and Traditional Classical Scholarship To a considerable extent, traditional ways of doing Classics, ancient history and classical archaeology either lend themselves to recasting in terms of distributed cognition or in fact already employ, usually without realising it, approaches and assumptions that can be expressed in those terms.1 For example, as contributors to  1

For assistance with this chapter I am very much indebted to Miranda Anderson, Felix Budelmann, Jennifer Devereaux, Alexander Forte, Christopher Gill, William Short, Mark Sprevak and Michael Wheeler, with special thanks to Felix Budelmann and Katharine Earnshaw for pre-publi-



dist rib uted c ognit ion and t h e cl as s i cs 19

this volume such as Christopher Gill and George Kazantzidis make clear, ancient philosophical theories of the nature of the psychē, its capacities, affections and afflictions, as well as its relation to the body, regularly rest on models that assume the interaction of body and soul in most if not all aspects of cognition and affectivity. The primacy of the body, as Kazantzidis demonstrates, is a given of Hippocratic medical theory. Aristotle’s hylomorphic theory of the relation between body and psychē implicates the body in all or virtually all aspects of mental functioning. And the Stoic pneumatic theory, as Gill notes, links mind, body and world in ways that present intriguing points of contact and discontinuity with the theories that have inspired the project that lies behind this volume. No one can deal with the ‘world soul’ in Plato’s Timaeus or with the Stoic conception of an interconnected, intelligent universe without discussing forms of cognition that extend beyond the body and into the world. Naturally, therefore, the phenomena that most interest us about these theories have long been central to traditional scholarship.2 For both Aristotle and the Stoics, too, the role of the body in mental functioning goes far beyond its salience in that prototypical subcategory of the pathē of the soul that we call emotions; but still, in recognising that salience, scholarship on ancient theories of emotion has likewise focused on concerns that have become fundamental to theorists of embodied and enactive cognition.3 Nor has the social embeddedness of affective cognition (or cognitive affectivity) been neglected: the sociality of emotion, both in ancient theory and in ancient literature, has been extensively emphasised.4 In a general sense, indeed, embeddedness has long been a term of art in classical studies, where an emphasis on shared, social and collective forms of experience has dominated a number of subfields, but especially that of ancient religion, still shaped as that field is by scholarly approaches that go back ultimately to Durkheim.5 In much of this scholarship, especially in view of its emphasis on social (especially festive) contexts and on shared experience of ritual, forms of social cognition (including forms of shared emotional cation access to the gradually expanding bibliography on ‘Cognitive Classics’ which now appears at (last accessed 19 February 2018).  2 For older bibliography on the Hippocratics, see Kazantzidis, this volume. For the centrality of the body in Aristotle’s accounts of mental functions, see most recently Mingucci 2015; among earlier studies, see especially Van der Eijk 1997 (with yet earlier literature in n. 34, 239–40); cf. Van der Eijk 2000; also Van der Eijk 2005 on the relation between medicine and philosophy. On the Timaeus, see Johansen 2004. On Stoic psychology, see Annas 1992; Long 1999. For medical and philosophical theories on the psychē-sōma relation in Hellenistic and Imperial philosophy and medicine, see von Staden 2000.  3 On Aristotle, see especially Fortenbaugh 2002, which goes beyond the first (1975) edition in emphasising Aristotle’s insistence that, as prototypical pathē of the psychē, what we call ‘emotions’ are paradigmatically ‘with body’. On Stoic emotion theory, see Graver 2007.  4 See especially Konstan 2006. For recent surveys of approaches to emotion in classical studies, see Chaniotis 2012; Cairns and Nelis 2017.  5 For the ‘embeddedness’ of ancient religions, see e.g. Bremmer 1994: 1–4; Parker 1996: 1–2; Beard et al. 1998: i. 43; Price 1999: 89; Rüpke 2001: 13.

20

dou glas c ai rns

experience) are central, albeit typically studied without reference to the issues raised in 4E cognitive approaches discussed in our series introduction.6 In ancient (especially Greek) history, on the other hand, a long-standing focus on collective decision-making, collective representations, cultural memory and shared forms of life is currently being reinvigorated through the application of New Institutionalist theory, with its emphasis on the agency of collective bodies.7 Related to this widespread notion of the embeddedness of individuals, their thoughts and their emotions in structures of social relations is the interpretation (especially associated with Christopher Gill) of classical Greek, Hellenistic and Roman models of mind and personality as themselves dialogic and socially conditioned.8 Theories of socially embedded cognition share with extended mind theories what Habinek and Reyes (in this volume) identify as ‘one of the basic principles of distributed cognition, namely that a system may well have properties not held by any individual within it’.9 Studies of ancient technologies in mainstream archaeological and historical scholarship do of course recognise that from the earliest beginnings human beings have created instruments, devices and prostheses of various sorts as tools which express, enact and extend their cognitive powers.10 Studies of the interrelation between thought and language, of the properties of poetic and other literary forms, genres and traditions, and of cultural artefacts such as laws and legal processes all implicitly identify ways in which human cognition makes use of external and intersubjective media all of which extend, enhance, scaffold and condition the abilities of individuals.11 But the explicit confrontation of traditional studies in these areas with the methods, theories and debates of the cognitive sciences is a phenomenon of no more than the past twenty years. In the next section, I discuss the general impact of the cognitive sciences on classical studies, before moving on to consider, in the subsequent section, the growing influence of 4E approaches.

Classics and the Cognitive Sciences Perhaps inevitably, given the nature and history of the discipline, a great many of the applications of ‘first-wave’ cognitive sciences to classical studies concentrate  6

See most recently Chaniotis 2011a. Budelmann’s chapter in this volume brings New Institutionalism into relation with theories of social cognition and tests their application to two areas of classical Athenian society that might be regarded as paradigmatic forms of collective agency. For bibliography on New Institutionalism, see his chapter; for its application to ancient history, see Canevaro 2018.  8 See Gill 1996, 2006 (Part iii).  9 Cf. Habinek and Reyes, this volume, p. 229, citing Hutchins 1995. 10 Cf. Roby, this volume, for some of the mainstream studies that have focused on the phenomena that she treats in terms of distributed cognition. 11 Cf. e.g. Habinek and Reyes (below, p. 234 n.19) on Cic. De or. 1.39, where ‘Scaevola . . . anticipates by two millennia Gallagher’s acute discussion [2013] of the legal “system” as a potential technology of cognition’.  7



dist rib uted c ognit ion and t h e cl as s i cs 21

on the composition, transmission and reception of texts (whether oral in origin or composed with the aid of writing). Prominent here have been a number of studies focusing on attention, memory and information-processing, pre-eminent among which are two monographs and a large number of shorter contributions by Elizabeth Minchin. Minchin’s recurrent focus has been memory, in all the varieties identified in contemporary cognitive psychology,12 and its importance for our understanding of the oral poetics and performance of the Homeric poems.13 Her 2001 book, Homer and the Resources of Memory (Minchin 2001a), discusses the indications that the texts of the Homeric poems give of the strategies of memory (episodic, visual, spatial and auditory) that allowed the oral poets from whose work the Iliad and Odyssey derive to compose creatively in performance. In the later book Homeric Voices (Minchin 2007), Minchin draws on earlier studies (Minchin 1992, 2002) to explore speech-acts and speech-genres in Homer from the perspective of cognitive linguistics. Minchin’s research is rooted in a tradition of oral poetics that begins, as far as Homeric studies are concerned, with Milman Parry in the 1920s, and can be seen more broadly against a background of Classics’ more general anthropological focus on the cognitive implications of orality and literacy and the contributions that can be made by oral tradition and written documents in the preservation of memory.14 As William Short observes in his chapter in this volume, ‘Walter Ong, Jack Goody, Ruth Finnegan, Jacques Le Goff and Jan Assmann (among others) have emphasised that the invention of writing has been perhaps the most powerful instrument for exteriorising and thus preserving memory.’15 These and other studies are widely used by classicists and ancient historians working on literacy and especially on its interfaces with orality.16 But so far few have investigated writing as a cognitive technology or reading as a cognitive skill. A notable exception is Jocelyn Penny Small’s pioneering study, Wax Tablets of the Mind (Small 1997), which ranges 12

For which see Short, this volume, pp. 93–5. Cf. Minchin 1996 on auditory, semantic and spatial memory in the organisation of Homeric lists and catalogues; 2001b on working memory and its limits (as an explanation for hysteron proteron in Homer); 2005 on autobiographical memory (as represented in Homer); 2008 on spatial memory; 2014 on the constraints that the limitations of short-term memory place on the oral-poetic technique of composition in performance; 2016a on the role of spatial and visual memory in creating the collective memories that preserve traditions relating to so-called ‘heroic tumuli’ in the Troad; 2016b on strategies, themselves explained in cognitive psychological terms, which facilitate the organisation of the oral poet’s repertoire of stories. 14 Cf. Short, this volume, p.  96 (citing Rubin 1995, Watson 2001 and Mackay 2008 alongside Minchin 2001a and 2007 on Homer). As a pioneer of such approaches in Classics Havelock 1963 (cf. Havelock 1986) deserves an honourable mention. 15 See Short, pp. 95–6 below (with references in n. 18). 16 Here the bibliography is vast, but since very little of it makes explicit use of the cognitive sciences we refer the reader only to recent standard works and surveys by Harris 1989; Humphrey 1991; Thomas 1992; Bowman and Woolf 1994; Pébarthe 2006; Johnson and Parker 2009. On orality and literacy, see especially the Brill series Orality and Literacy in the Ancient World, of which the two most recent volumes are Scodel 2014a and Slater 2016. 13

22

dou glas c ai rns

widely over technologies for the production, dissemination, storage and retrieval of ancient texts, the interface between literacy, orality and ancient mnemotechnics,17 and the importance of memory both in the composition of ancient texts and in the technologies used to produce them.18 One area of ancient literary studies that has seen a proliferation of approaches implied by the cognitive sciences is the study of ancient narratives: cognitive narratology is a flourishing and expanding field in Classics as elsewhere. A landmark study in this direction is N. J. Lowe’s 2000 volume, The Classical Plot and the Invention of Western Narrative, an ambitious exploration of what Lowe presents as a canonical ancient approach to plot-construction and its huge influence on Western literature. As part of this project, Lowe (2000: 17–35) develops a ‘cognitive model’ according to which the variously labelled elements that narratologists identify as constitutive of a work of narrative fiction – such as (in his version) text, story and narrative, or (in other versions) text, fabula and story – are aspects of the reader’s activity in reading the work. The features of these elements (the way that we understand them in terms of notions such as time, space, agency, causality, etc.) in turn rely upon underlying cognitive mechanisms (such as memory, imagination and associative learning) that we use from infancy onwards to make sense of our experience in the world. In particular, he argues, we use these capacities to construct model worlds which, though they can be described to some extent in terms of scripts (Schank and Abelson 1977), are more productively thought of using the metaphor of games (Lowe 2000: 31–4). Thus he concludes (2000: 34, original italics): Plot . . . is the affective predetermination of a reader’s dynamic modelling of a story, through its encoding in the structure of a gamelike narrative universe and the communication of that structure through the linear datastream of a text. In a classical plot, this narrative universe is strongly closed, privileging the values of economy, amplitude, and transparency. In other cognitively inflected approaches to narrative a major focus has been Theory of Mind, defined (in psychology and the cognitive sciences) as the ability to attribute mental states to others as well as to oneself and to understand that others’ representations and perspectives may differ from one’s own. Budelmann and Easterling (2010) show how the texts of Greek tragedy engage, as a basic element in their construction of dramatic character, the mindreading capacities of audiences and readers. Similarly, in a number of papers, Ruth Scodel has explored topics such as the limitations of tragic narrators (such as Messengers) in drawing influences about others’ intentions (Scodel 2009); the role of the particle ἦ (‘to be sure’) in passages in which Homeric characters claim to understand others’ mental states (Scodel 2012); the creation of indeterminacy about speakers’ and agents’ motives in Homeric ‘Theory of Mind’ scenes (Scodel 2014b); and the similari17 18

Cf. Habinek and Reyes, Short, this volume. Cf., more recently, Battezzato 2009, analysing the textual layout of ancient Greek script from the perspective of cognitive load and efficiency in terms of retrieval and retention.



dist rib uted c ognit ion and t h e cl as s i cs 23

ties and differences between the deployment of Theory of Mind in Homer and Vergil and its importance for Vergilian narratology (Scodel 2015). Comparable but ­somewhat different in perspective is Minchin 2011, which uses the theories of Schank and Abelson 1977 (already deployed in Minchin 1992 and 2007) to argue that understanding of the plans and goals that underlie the scripts of everyday events is central to Homer’s representation of character and motivation. Though Minchin’s article is not directly concerned with Theory of Mind, its approach bears comparison in some respects to the narrative account of Theory of Mind put forward by Shaun Gallagher and Daniel Hutto.19 The representation of collective experience (rather than collective minds) through narrative (in the historian Thucydides and the late-antique novelist Heliodorus) is the focus of a recent article by Jonas Grethlein (Grethlein 2015b), though Grethlein is less concerned with social cognition as such than with the presentation of collective experience as a way of immersing the reader in a narrative. Grethlein has also recently engaged with those (such as Alan Palmer (2004) and Lisa Zunshine (2006)) who have seen the exercise of complex Theory of Mind skills as central to the activity and the attraction of reading fiction, arguing that an ancient novel such as Heliodorus’ Ethiopica offers much less purchase for making inferences about characters’ mental states than do the modern novels discussed by Palmer and Zunshine.20 More broadly, approaches from cognitive linguistics have also begun to have a substantial impact on the study of Greek and Latin texts. As well as editing a volume of Oral Poetics and Cognitive Science (Antović and Pagán Cánovas 2016), Cristóbal Pagán Cánovas and Mihailo Antović (2016) have discussed the oral formulaic theory of composition in Homer in the light of ‘usage-based’ approaches to grammar and language acquisition as deployed in cognitive linguistics, linking the formulas, themes and story patterns of Homeric poetry (qua ‘formal-­­semanticmetrical templates’) to approaches which link the acquisition of linguistic and grammatical skills to human beings’ general cognitive capacities. Pagán Cánovas is also well known for his work on image schemas, pre-linguistic patterns fundamentally rooted in embodiment and experience that, in the work of George Lakoff and Mark Johnson, underpin conceptual metaphor theory.21 In a recent article, however, Pagán Cánovas argues against what he describes as the static source-totarget projections of cognitive metaphor theory and in favour of a more dynamic and flexible model in which infants’ innate or at least early command of spatial primitives gives rise to ‘the skeletal stories, or image schemas, into which these primitives are integrated’ and ‘which constitute the first conceptual structures’. These ‘basic spatial stories . . . do not make sense in isolation but as part of a narrative’ and ‘serve as building blocks for forming schemas with event structure’ (Pagán 19

See Gallagher 2006, 2012; Gallagher and Hutto 2008; Hutto 2007, 2008a, 2008b. Grethlein 2015a. For a more extensive and fundamental challenge to the approach of Zunshine 2006 in particular, see Van Duijn 2016. 21 See Lakoff and Johnson 1980, 1999; Johnson 1987; Lakoff 1987. 20

24

dou glas c ai rns

Cánovas 2016: 121–2). It is image schemas of this sort, he argues, that are then available for the creation from disparate aspects of experience of new (often ad hoc and creative) conceptual blends, as argued by the blending or conceptual integration theory put forward by Gilles Fauconnier and Mark Turner (2002). This model is then applied to ancient and modern Greek poetry, in a treatment similar to that found in Pagán Cánovas’s earlier articles on emotion metaphor, in particular the image of the ‘arrows of love’.22 By contrast, the Lakoff–Johnson approach to cognitive metaphor is alive and well, and being applied to classical texts, by scholars such as William Short, Fabian Horn and Douglas Cairns. In an extensive and expanding series of studies,23 Short has explored a wide variety of ways in which mappings from more concrete, empirically accessible domains scaffold and structure abstract concepts in Roman life, literature and thought – in the Romans’ spatial models of mental activity in general (Short 2008, 2012; cf. below, Chapter 6); in the way that the Romans’ spatial conceptualisation of truth and knowledge gives rise to metaphors that present ‘mistakenness’ in terms of wandering (errare, Short 2013a); in Latin’s presentation of communication in terms of cooking, serving, eating and digesting (Short 2013c); and in the construction of the concept of time in terms of spatial relations that structures both Latin semantics and other, non-linguistic forms of symbolic representation (Short 2016b). On the Greek side, Horn has pursued a similar approach, investigating a number of individual poetic metaphors (Horn 2015a, 2015b, 2016a, 2016b) and co-editing a volume on spatial metaphors in ancient texts (Horn and Breytenbach 2016).

Explicit Engagement with Distributed Cognition In so far as the studies mentioned in the previous paragraph explore metaphors arising from the embodied nature of human beings, they already imply forms of embodied cognition, and thus directly apply one of the approaches to distributed cognition that are the focus of this volume. More explicit in their emphasis on embodiment, however, are the essays contained in Short’s recent volume Embodiment in Latin Semantics, where a team of nine mostly younger scholars, many of whom have published on relevant topics elsewhere, explore the contention that (as Short puts it in his introduction) ‘people’s ability to make sense of and communicate about their experience is underwritten by conceptual structures and cognitive processes that emerge from human bodily interaction with the sorts of physical and social environments we inhabit (or have historically inhabited)’ (Short 2016a: 1).24 Cairns’s studies of emotion metaphor share a similar commitment, but are also characterised (first) 22

See Pagán Cánovas 2010, 2011, 2014. For a further application of blending theory to classical Greek poetry, see Budelmann and LeVen 2014. A book-length study applying the theory to Aeschylean imagery is forthcoming from Michael Carroll. 23 Cf. also the overview in Short 2014. 24 Cf. Short 2013b on the preposition de.



dist rib uted c ognit ion and t h e cl as s i cs 25

by an emphasis on the fundamental role of the body in emotional experience and (second) by close attention to the role played by the phenomenology and expression of emotion in the construction of embodied emotion metaphors. Thus the ancient Greek concept of phrikē (piloerection, goose pimples, shuddering, but also – by metonymy – a fear- or horror-like emotion in its own right) takes a symptom that has its roots in basic somatic mechanisms of temperature regulation, that is manifested in a range of non-emotional contexts, and that is shared with other animals, and constructs a concept of fear or horror in which physical symptoms are intimately related to, perhaps even constitutive of, cognitive appraisals and evaluations (Cairns 2013). At the same time, the specifically Greek concept of phrikē also takes on specific social and cultural connotations by virtue of its typical association with religious experiences (such as initiation, epiphany) and aesthetics (cf. also Cairns 2017c). The role of the symptomatology of emotion in the metaphors that structure emotion concepts also allows Cairns (2016b) to investigate the extent to which ancient Greek elpis (hope, expectation) manifests the qualities of more prototypical members of the category of emotion, while Cairns 2016a looks at the influence of the pragmatics of dress and gesture on the formation of ancient Greek garment metaphors for emotions such as shame and grief. In a related study (Cairns 2014; cf. Cairns 2017a), he looks at the metaphorical construction of the concepts of psychē (soul) and thymos (‘spirit’) in Homer and Plato, with a view to demonstrating how the phenomenology of self-division and mental conflict is reflected in forms of personification that in fact emphasise the unity of the embodied person as subject of thought and emotion. More recently, the pursuit of this project has led to an account of the importance of the body and the environment in the ways in which both Euripides’ and Seneca’s versions of the Phaedra myth implicitly recognise the ways in which our experience, perception and conception of the world depend on our existence as holistic, physically embodied organisms in sensory contact with the physical environment through which we move (Cairns 2017b). A somewhat different approach to embodied cognition is taken by Peter Meineck, who, both in this volume and in previous work, has stressed the embodied aspects of cognition and emotion in classical Athenian tragedy, emphasising that translation and interpretation must take account of the spatial and physical elements of performance, with all the traces that these leave in the texts as we have them and all their implications for the embodied cognition of both performers and spectators (Meineck 2012, 2016). In a more specific study, reprised in part in his contribution to this volume, Meineck (2011) uses contemporary work in cognitive neuroscience and psychology (with particular reference to capacities for facial recognition and the production and decoding of emotional expression) to demonstrate convincingly the powers of the fifth-century tragic mask as a physical means of both suggesting and arousing emotion: as a physical object, the mask not only exploits spectators’ abilities to read and infer mental states, complementing, embodying and interacting with the indications offered in the performance script, but also throws into added relief the importance of the body in manifesting mental processes and eliciting embodied emotional responses in the audience.

26

dou glas c ai rns

The theatre is also an important context for the investigation of forms of embedded and social cognition, as demonstrated by the contributions of Budelmann and Ng to this volume. So far, however, the most significant study of the extent to which the cognition of groups may be said to manifest properties that are not wholly identical to that of the individuals who comprise them is Garrett Fagan’s The Lure of the Arena: Social Psychology and the Crowd at the Roman Games (Fagan 2011), a pioneering and historically sophisticated account of the ways in which group membership and social context condition the appetite for shared consumption of violent spectacle, not only in the Roman arena – a performance space that exhibits intriguing points of similarity and difference with respect to that of the theatre – but in other historical contexts too. Beyond that, most research in Classics that is concerned with socially conditioned, socially embedded and shared forms of cognition has focused on religion, where recent work in other fields on the sociality of emotion, on social network theory and on distributed cognition has prompted new approaches to the familiar phenomenon of the ‘embeddedness’ of Greco-Roman religion.25 On some accounts of extended mind theory, many of the phenomena studied under the heading of embedded, embodied, enactive and social cognition would count also as examples of extended cognition, especially if we adopt Menary’s ‘integrationist’ approach to the hybridisation of organism and environment.26 Applications of extended mind theory may also be found in one or two contributions by classical scholars. Elizabeth Minchin’s 1996 study of the role of lists and catalogues in scaffolding the oral poet’s ability to improvise in performance by developing auditory, semantic and spatial memory provides an interesting precursor to Riggsby’s study in this volume.27 Similarly, a study that complements Gill’s chapter in this volume may be found in Thomas Habinek’s well-known and authoritative piece on the ‘tentacular mind’ of the Stoics, a model of the mind that not only emphasises, as distributed cognition theories do, the continuity between mind and world, but also locates cognition beyond the individual organism in a connected, sentient universe in which ‘boundaries between bodies are in effect useful fictions’.28 The Stoics, as Habinek shows, insisted on involvement of the whole embodied organism in cognition and on the active participation of the organism in the pro25

See especially Chaniotis 2011a, on the role of ritual (including its spatial and material aspects) in creating what Barbara Rosenzweig has called ‘emotional communities’, and a number of studies by Esther Eidinow on the models of mind, both divine and human, that emerge from the embeddedness of Greek religion in specific social networks (Eidinow 2011, 2015a, 2015b). Cf. e.g. Lulić 2014 on the role of material culture in embodying and anchoring shared religious concepts. 26 Menary 2006. For Slaby 2014: 37 (cf. and contrast Krueger and Szanto 2016: 865) such an approach means relaxing Clark and Chalmers’ ‘parity principle’; but this involves taking the parity principle to demand that external and internal processes be functionally identical, when all that is required by Clark and Chalmers 1998 (cf. Clark 2008b: 77–8, 91) is that external processes should work in such a way that, were they internal, they would be regarded as cognitive. 27 Cf. Riggsby, this volume, p. 58, n. 2. 28 Habinek 2011, esp. 65–6 (quotation p. 66). Cf. Gill, this volume.



dist rib uted c ognit ion and t h e cl as s i cs 27

cesses by which it perceives and interacts with its environment. In all these respects, the Stoic model of cognition and emotion (again, processes that the Stoics see as aspects of a whole) lends itself to discussion also in enactivist terms. Explicitly enactivist accounts of classical materials and sources, however, have so far been scarce.29 One very promising example is Grethlein and Huitink 2017’s exploration of the much-celebrated ‘vividness’ (enargeia) of the Homeric poems as an enactive process (cf. Huitink, this volume). Whereas previous approaches to phantasia (imagination) and enargeia have taken at face value ancient sources’ apparent ­­privileging of the single modality of vision (so that they typically talk about Homeric enargeia in terms of ‘mental images’),30 Grethlein and Huitink point to the sparsity of explicitly visual cues in Homer and instead emphasise the extent to which audiences and readers, encouraged by narrative cues that highlight affordances for sensorimotor interaction, imaginatively place themselves within the world in which the narrative unfolds by synthesising a range of sensory modalities in line with their knowledge and experience of the scenarios involved. The phenomenon of enargeia is so fundamental to ancient rhetorical and aesthetic theory and the capacity of phantasia so crucial to ancient psychology that scholars interested in these subjects from a traditional point of view would indubitably benefit from acquaintance with enactivist and other distributed cognition theories.

The Chapters in this Volume As participants in the research project from which this volume derives, our contributors were given the remit of considering either (a) ways in which classical sources might be regarded as representing approaches to human cognition that in some way prefigure modern theories of distributed cognition or (b) ways in which such modern theories might shed useful new light on ancient texts and the traditions by which they have been interpreted. Beyond that, contributors were free to decide on their own perspectives and material. In particular, we did not seek to impose a requirement that each of the four modes of distributed cognition (embodied, embedded, extended and enactive) should be covered to the same extent in each of the four volumes of this series. In fact, the preponderance of one or more of these approaches in the four workshops that preceded our volumes provided interesting testimony to the differing priorities and predilections of researchers across the periods and disciplines that the project has covered. Perhaps the most inclusive of the chapters in integrating the full range of 4E approaches is Courtney Roby’s wide-ranging exploration (Chapter 3) of distributed cognitive strategies in the work of the second-century ce scientist Claudius Ptolemaeus, or Ptolemy. Ptolemy’s approach in disciplines such as geography, astronomy and music entails a profound interest in how human beings acquire, assess and deploy knowledge in these disciplines. At every level of these processes, 29

For limited forays into this area, see Cairns 2017b, forthcoming. Even in the cognitively inflected approach of Minchin 2001c.

30

28

dou glas c ai rns

it emerges, Ptolemy consistently regards the cognitive processes involved as distributed. First, he works with a materialist notion of psychē in which soul and body interact and exist in a relationship of mutual interdependence. Though his views on the body-soul relation have affinities with those of the Stoics (as elucidated by Gill in this volume), Ptolemy differs from them not only in regarding the brain as the location of the soul’s most important cognitive functions, but also in being open to the possibility that there may be more than one such hēgemonikon, distributed throughout the body. Thus conceived, the body is an integrated system whose sensorimotor functions and cognitive capacities are interrelated in various complex ways in the process of scientific investigation. This embodied model underpins Ptolemy’s approach to the use of the artificial instruments which extend our capacity to perceive and interpret the world. Whether the focus is on the use of geographical tables and star catalogues as textual instruments, on the scientific instruments used by ancient astronomers or on harmonic instruments such as the kanōn, Ptolemy’s primary concern is the greatest and most efficient integration of mind, body and external resource as aspects of the process of scientific investigation. With practice, expert use of such external instruments in effect remodels not only the mind but the sensorimotor processes by which cognition is executed. Ptolemy’s scientific approach thus, as Roby argues, shows a close affinity with models of distributed cognition as developed by Edwin Hutchins. But in so far as the investigator and the instruments deployed in the investigation become a ‘coupled system’ to whose efficacy the instrument is crucial, Ptolemy’s approach also fulfils the requirements of extended mind theory as developed by Clark and Chalmers. And finally, since this is an approach in which ‘Mind, eye, hand, instrument and text cooperate within a space that blends the cognitive and the material, sensory experience and theoretical knowledge, internal faculties and external tools, integrating the scientist into the world he explores’ (p. 54), Ptolemy’s ‘emphasis on exploring the environment through skilled manipulations of body and instrument suggests that his cognitive model might be situated somewhere on the enactivist spectrum’ (p. 55). Andrew Riggsby’s account of the use of tables and nested lists in the Roman world (Chapter 4) nicely complements Roby’s chapter in demonstrating how our own adoption of a distributed approach to cognition can help elucidate ancient practices that might otherwise be difficult to explain. Riggsby defines a table as a conceptual device in which two axes, consisting of individual columns and rows, are used to represent identifiable variables. In a nested list, on the other hand, each of the entries gives rise to another list (e.g. chapter, section, subsection), and the list is ‘multiply-nested’ if its lower levels correspond in following the same system of identifying the sections. Both tables and nested lists facilitate retrieval of information about items and the relations between them. Some have assumed that the capacity to understand and use tables is universal, but Riggsby observes that the use of tables and nested lists (as opposed to ordinary lists, no matter how formatted) in the Latin-speaking world is very rare and where it does occur it is in a restricted range of specialised contexts. This is because, he argues, there is



dist rib uted c ognit ion and t h e cl as s i cs 29

a cost to acquiring the skill of using tables comfortably, useful though that skill may be once acquired. It becomes easier to acquire the skill where use of tables is common, and dissemination of the skill encourages use; where tables are not widely used, on the other hand, the skills required to make sense of them are rarer. But, though rare, forms of tabular representation do occur (in the practices of the Roman land-surveyors and in fragments of army duty rosters), while nested lists (in triple-index form) are found almost exclusively in large government archives, both in Rome and in Roman Egypt. In all these cases, the practice in question is sustained by the social and organisational context in which it is embedded. These are all specialised and limited contexts in which the ability to create and to make use of the cognitive artefact in question can flourish – the relevant devices are not designed for widespread public consumption. More than that, however, Riggsby shows how these particular expedients depend on a ‘close connection between the material implementation and the physical actions of the various agents involved in producing and using the records in question’ (p. 66). In the case of the practices of the land-surveyors, the tabular structure is actually imposed on the land itself by means of a grid in which squares are identified in terms of their distance from each of two perpendicular roads. ‘Here’, Riggsby observes, ‘the reality directly models the information-organising device, and that reality is created not by a single act of imagination, but by repetition of the specific procedures of the survey process’ (pp. 64–5). These are all complex distributed systems in which problems of linear processing are offloaded on to external forms of representation that rearrange the items to facilitate retrieval of the information concerned and to permit the storage of that information over time. Comparable to Roby’s chapter in the range of distributed cognitive strategies that it canvasses is Peter Meineck’s discussion (in Chapter 5) of the role of the mask in Attic tragedy. Meineck draws on Lambros Malafouris’s Material Engagement Theory (Malafouris 2013) to offer a model of mask, performer and performance environment as an integrated system, encompassing embodied, embedded, extended and enactive forms of cognition. As an object and vehicle of cognition, the mask is patterned on but differs from the human face. Its use as a physical extension of the actor’s body has substantial implications for performance style, especially in terms of the need for the actors and chorus, for reasons of audibility, to face the audience, but also in the way that the expressive powers of the mask itself are enhanced not only by verbal cues in the text but also by the expressive powers of the body. As a focus of interaction not only between characters, but also between characters and audience, the mask is not a single, isolated feature of ancient performance practice, but rather has entailments that implicate the actors’ bodies, their movement in the performance space, and their engagement of the audience’s emotional responses. Focus on the gestures and movements of the masked performers (no doubt supplemented and enhanced by cues in the text) may promote what Meineck calls ‘kinaesthetic empathy’ in the audience, as the masked body moving in space acts as anchor for responses based on sensorimotor resonance and emotional mirroring.

30

dou glas c ai rns

Complementary to Meineck’s chapter, which deals with classical Greek evidence, is William Short’s similarly wide-ranging discussion of embodied, extended and distributed cognition in Roman technical practice (Chapter 6). Short deals in particular with theories which ‘posit that reasoning in abstract as well as in concrete domains is underwritten by conceptual structures and cognitive processes arising from our sensory and motor interactions with the world’ (p.  92). He begins by focusing on techniques of memorisation and calculation, two areas which are paradigmatic in their use of external resources to offload cognitive functions on to the environment, using not only technologies such as writing, reckoning tables and the abacus, but also (in the case of calculation) one’s own fingers or (in memorisation) the memory capacities of other human beings (such as the mnēmones of Greek myth, the amanuenses known as nomenclatores in Roman society or the monitor or prompt in Roman oratorical practice).31 But others’ powers of recall can be as fallible as one’s own, and so Short also discusses ancient techniques for strengthening the memory. Ancient mnemotechnics (whose historical development Short outlines) relied on spatial visualisation, focusing first (in the earliest sources) on the text itself as a series of quasi-spatial locations, but developing in time into the imaginative creation of fully developed environments or memory palaces through whose spaces an orator would move in memorising and rehearsing the sequential structure of a speech. Similar processes of using human beings’ physical interaction with the environment as an internalised form of mental scaffolding can be observed also in Roman politics, where the senatorial practice of ‘voting with the feet’ made it possible to gauge at a glance the disposition of support for or opposition towards given proposals. In this way, the floor of the Senate became not only a board for tallying votes but a symbolic representation of political networks and alliances. Short’s main point involves the observation of continuity and complementarity between the ways in which external resources support cognition and the metaphors and image schemas that construct the concepts that human beings use to think and talk about these processes. Writing, for example, is not only a technology for enhancing human memory but a recurrent metaphor, from antiquity to the present, for the very function that it enables. In the same way, Short argues, Roman ways of enlisting the spatial environment to scaffold cognitive functions reflect and are reflected by Latin speakers’ conceptualisation of thought itself in spatial terms, so that thoughts are conceived of as locations and thinking as movement through space. Though such metaphors and image schemas are ultimately conditioned by fundamental features of the human body and basic modes of interaction between the individual and the environment – and so can be paralleled in other languages and cultures – nonetheless, Short argues, the ‘embodied spatial metaphorics’ of the Latin language constitutes the Romans’ ‘ “preferential” or “privileged” conceptual model for comprehending the mind and its various operations’ (p. 107). Meineck’s emphasis on distributed cognition in theatrical space and Short’s on spatial constructions of social and political meanings resurface in Diana Ng’s 31

Cf. Konstan, this volume, p. 210.



dist rib uted c ognit ion and t h e cl as s i cs 31

chapter on ‘Roman-period Theatres as Distributed Cognitive Micro-ecologies’ (Chapter 7). Inspired by Evelyn Tribble’s analysis of Shakespeare’s Globe Theatre as a ‘cognitive ecology’ (Tribble 2005, 2011), Ng investigates how in the stonebuilt Roman theatre, especially in imperial Asia Minor, ‘cognition was distributed across the individuals in attendance, the seating arrangements and the decoration of the building. To be in the theatre was to see, occupy and know one’s place relative to others and – in some cases – to the larger entities of region and empire’ (p. 117). Unlike Meineck, Ng is not specifically concerned with forms of distributed cognition at work in the creation and appreciation of a dramatic spectacle in an open-air theatre, but rather with theatrical space as a medium for the storage and retrieval of various forms of social information – a function that is as applicable to the theatre as a civic and political arena (e.g. as a venue for public meetings) as it is to the theatre as the location of dramatic performances. As had earlier Greek theatres in their own historically specific ways, imperial Roman theatres made social status visible, not only in the way that the auditoria themselves were segregated by rank, but also by the dress codes in which attendees of different ranks displayed themselves and in the use of the theatre as a locus for the display of specific marks of esteem, such as the wearing of golden crowns. In the same way, the visual and public prominence of the theatre as a space also made them places of memory, for the commemoration of civic benefactors, for the celebration of civic identity through sculptural representations of important local mythology, and even, as in third-century ce Aphrodisias, as the location of an epigraphic archive of the city’s correspondence with the emperor in Rome. The focus of all the chapters discussed so far on embodiment, and on the role and function of the embodied organism as a whole in making sense of its environment, is echoed in several of the other chapters. Christopher Gill (Chapter 9) discusses embodied and enactivist approaches to cognition in Stoic philosophy and in Plato’s Timaeus. Stoic theory, in particular, involves a strongly embodied account of mental functioning: though cognition is subject to the control of the hēgemonikon in the heart, all functions of the hēgemonikon itself involve physical changes, the body is an integrated psychophysical system and the pneuma that is concentrated in the hēgemonikon is itself not only material but also diffused throughout the organism as a whole. The Stoics’ famously ‘cognitive’ approach to emotion (in which all ‘emotions’ or pathē – in the Stoics’ own strict sense of the term – involve ‘assent’) is also strongly embodied: a pathos for the Stoics is an occurrent psychophysical event, involving specific physical changes in the heart and (in the first stirrings of emotion, or propatheiai, to which assent may or may not be granted) a wider set of changes in the body as a whole. In the Timaeus, the three functions of the tripartite soul (reason, spirit and appetite) are located in different bodily centres (in the head, the chest and the abdomen), but unified by a communications network involving other bodily organs (such as the brain, the heart, the lungs, the liver and the spleen), the blood vessels, myelos (bone marrow and cerebro-spinal fluid) and so on. One might, therefore, say that the account of psychic functioning in the Timaeus is in some ways less centralised than that of the Stoics. On the other hand, the Timaeus works with

32

dou glas c ai rns

a distinction between immortal, non-material psychē (located in the head) and the mortal psychē located in the chest and abdomen. Both the Timaeus and the Stoics rely on a natural teleology that makes organisms at least potentially fit for purpose and capable of increasing degrees of environmental adaptation. For Gill, the Stoic view resembles contemporary enactivist approaches to cognition in so far as Stoic creatures organise their physical and mental functions in a holistic, coordinated way that answers to the autopoiesis of enactivism, while the Stoic process (oikeiōsis) by which an organism progressively adapts to and appropriates environments of increasing complexity approximates quite closely to enactivist accounts of adaptivity and sense-making. Finally, both the Timaeus and Stoic theory raise questions pertinent to this volume in that both regard the universe as a living system, permeated by the pneuma (for the Stoics) or the psychē (in the Timaeus) that is responsible for human rationality. The resemblance to contemporary accounts of the extended mind, however, is largely superficial: whereas for extended mind theorists aspects of the world extend the capabilities of human thought, performing aspects of our cognition for us, in the Timaeus and for the Stoics, by contrast, human reason is a local instantiation of the rationality of the world as a whole. Embodied and enactivist approaches also form the focus of George Kazantzidis’s chapter (Chapter 8) on the presentation of emotion in the Hippocratic Corpus. The Hippocratics differ from the philosophical approaches to emotion found in (e.g.) Aristotle and the Stoics in focusing almost exclusively on physical symptomatology to the virtual exclusion of the cognitive-evaluative element (even if, one can reasonably surmise, some form of appraisal of external states of affairs is often taken as read). This is a tendency that Aristotle himself perhaps criticises, albeit obliquely, at De anima 403b13–14. The authors are alive to a wide range of organic causes of emotion and emphasise (as do many modern theorists) various ways in which bodily states, either in themselves or as responses to environmental phenomena, may be sufficient to cause a range of affective states. As a consequence of this emphasis, Kazantzidis writes (p. 134), the Hippocratic authors’ ‘pronounced emphasis on the body as essentially a sense-making mechanism which absorbs lived experience and filters an individual’s interaction with his or her environment shares something of enactivism’s tendency to “decentralise” consciousness and to tie it intimately with embodied activity, not only within the brain, but also in the full body in action’. In fact, however, the Hippocratic approach to emotion seems rather to elide emotion’s external causes, and so does not give any very prominent sense of that dynamic interaction between organism and environment that is important to enactivism. Thus it does not quite capture the totality of emotion as an interactive way, at once both cognitive and affective, of making sense of the world. The Hippocratic focus, too, is more a matter of assumed physiological underpinnings than of the phenomenality of the experiences involved; if the Hippocratics had more to say about how the relevant phenomena present themselves to their subjects, they might also have to say more about the intentionality of those experiences. Embodiment is also fundamental to the enactivist approach taken by Luuk



dist rib uted c ognit ion and t h e cl as s i cs 33

Huitink (Chapter 10) to the examination of enargeia, or ‘vividness’, a key term in ancient aesthetics, literary criticism and rhetorical theory. Contemporary enactivist approaches to readers’ experiences in processing literary narratives work with a view of the reader’s imagination that is based on embodied experience in the world, rather than on mental pictures or representations. Ancient theories, on the other hand, explicitly emphasise the visual and in particular regularly comment on the ability of ‘vivid’ narratives to elide the difference between readers and eye-­­witnesses. But in fact the appearance of a disjunction between ancient and modern theory is deceptive, and it is possible, as Huitink shows, to use ‘the non-­ representationalist enactivist theory of vision and imagination’ to develop ‘a more precise account of some crucial stylistic and experiential aspects of enargeia’ (p. 175) that reveal ancient theory to have been much more enactivist in orientation than is normally assumed. First, as Huitink reminds us, vision itself is neither a passive process nor a matter of creating a snapshot of an entire visual field; it responds above all to aspects of the world that are salient from the perceiver’s point of view as an embodied creature in a particular spatial and temporal environment, and prominent among such features are those that particularly offer scope for physical interaction. Imagination, similarly, involves the as-if rehearsal of embodied experience; it is a multimodal and not an exclusively visual experience, rooted in sensorimotor engagement with the physical environment. If we re-examine ancient accounts and examples of enargeia with these points in mind, we see that the ancient emphasis on mental images and on the visual in general involves a kind of synecdoche: the ‘placing before the eyes’ that enargeia is said to entail regularly involves the idea of feeling oneself physically present, located in the scene in question with a perspective of one’s own and thus able to respond with all one’s senses, not only visually. Presence also implies movement, and in fact scenes which are praised for their enargeia typically highlight both physical movement of agents and objects and the affordances that the scene described might offer for physical interaction. The experiential, physical, embodied aspects of ancient enargeia are considerable. Thus it is not merely the case that contemporary theory and criticism can shed light on ancient aesthetics and rhetoric, but rather that ancient theories of enargeia can usefully be embraced and deployed by contemporary cognitive criticism. Several of the chapters focus on social, group or intersubjective manifestations of cognition, i.e. forms of cognition that count as distributed in so far as the properties of the group experience differ from those of the experiences of the individuals involved. The notion of ‘group minds’ – whether a group can in fact be said to have a mind that is not resolvable into the minds of the individuals who comprise it – is problematised and explored especially by Felix Budelmann (Chapter 11), a chapter which not only employs contemporary approaches to social cognition as a means of exploring possible candidates for consideration as examples of the phenomenon in classical Athens but also uses the Athenian cases to probe and to test theories of social cognition. The Athenian dēmos, Budelmann shows, is treated as a singular and homogeneous collective agent; even in circumstances of faction, the dēmos is a whole, a player in its own right, rather than an entity that dissolves into

34

dou glas c ai rns

classes or parties. Athenian decrees record decisions as if they had been arrived at unanimously (much as ‘the will of the British people’ is alleged to have been established by a referendum in which the votes were divided 52 per cent to 48 per cent). Choruses (both dramatic and non-dramatic) are, Budelmann argues, in some ways comparable (both as characterised agents in the world of the play or performance and as groups of performers), though these perhaps have more of a tendency to divide into smaller units such as semi-choruses, even in tragedy; beyond tragedy, as in Alcman’s Louvre Partheneion, chorus leaders can be named and set apart from ordinary members, or choreuts given individual names, as in comedy and satyr play. Occasional exceptions (divided choruses, a divided dēmos in times of tension) confirm the general rule. Though the Athenians themselves refer repeatedly to the agency of the dēmos or chorus without apparent interest in the questions that might engage modern theory, Aristotle (especially at Politics 3.11, 1281a40–b10) can raise issues such as the wisdom of crowds and the possibility that a group, like a single organism, may have a character of its own that is not simply the product of the contributions of its individual members. Even if, Budelmann argues, neither the behaviour of dēmos and chorus nor the ways in which the Athenians thought about that behaviour would convince a modern sceptic of the existence of group minds, nonetheless they ‘provide an example of a culture that, in certain specific instances, appears to be more extensively committed than we are today to treating the group rather than the individual as the default level at which mental activity is situated’ (p. 207). He thus ends with a challenge to contemporary scholarship to adopt the emic perspective of the Athenians themselves and begin from the group rather than the individual level. David Konstan (Chapter 12) discusses ancient notions of friendship, with a particular focus on Aristotle’s well-known description of a friend as a second self. Friendship, Konstan shows, differs from other ways in which one’s own cognitive capacities might be extended by means of interaction with other human beings (such as the use of a slave nomenclator in Seneca’s 27th Letter, as discussed by Short in this volume): whereas a slave might provide cognitive resources for the use of his master, friendship is a mutual relation of equals. ‘Friendship’, Konstan argues (p.  214), ‘is an ideal instance of the social distribution of cognition, for it is by nature a relationship that depends on a special kind of reciprocal consciousness and is not intelligible solely as a representation in the mind of a single individual.’ Aristotle, on this account, prefigures some modern approaches to intersubjective cognition in seeing friendship as requiring ‘two minds, not just one’. For Aristotle, ‘in friendship proper, cognition is distributed’, for, in order for it ‘actually to exist, it is necessary that both parties in fact feel the same way and that both know this to be the case’ (p. 216). Being friends is an activity (like virtue or happiness, in Aristotle’s thinking), not a state of affairs, and, unlike liking or loving, this is an activity that must be carried out by more than one person; its mutual, shared character means that this activity of being friends has qualities that my liking for you and your liking for me do not possess in themselves. The final chapter of the volume, by Thomas Habinek and Hector Reyes, shares



dist rib uted c ognit ion and t h e cl as s i cs 35

Konstan’s focus on dialogue and shared activity as modes of cognition, especially in contexts of philosophical investigation and pedagogy, but it does so in ways that both contrast with and complement the approaches adopted in the previous chapters. Habinek and Reyes’ discussion takes us into the continuous and culturally embedded process of receiving, appropriating and interpreting classical sources, in so far as they take as their cue two early modern paintings with classical themes, Raphael’s School of Athens and Poussin’s Landscape with Numa Pompilius and the Nymph Egeria. These are examined through their connections with a network of classical sources, especially Plato’s Protagoras, Ovid’s Metamorphoses and Cicero’s De oratore. Habinek and Reyes see the paintings and the classical sources to which they may be related and compared as elements in an interrelated system that sustains their own intellectual activity as scholars. Their method exemplifies its own thesis, that education, scholarship and intellectual history are collaborative, multimodal and culturally embedded activities that depend on a wide range of external resources. In the Raphael painting, ancient philosophical thought is associated with a series of individual thinkers, each envisaged as an independent source of knowledge in a tradition that reaches into the present of both painter and viewer. In Plato’s Protagoras, on the other hand, a collection of the fifth century’s most prominent thinkers gather at a single place and time in a context that at least holds out the potential for ‘knowledge construction as a situated, embodied practice’ (p. 228). But in that dialogue the refusal of both Socrates and Protagoras to adopt their own mode of inquiry to that of the other means that their interaction fails to create a system capable of achieving more than either individual might in isolation – ­­ironically so, given that this is the method that Socrates ostensibly espouses. Poussin’s painting, for its part, defies association with any single ancient model and so any single determinate interpretation. But it also evokes a number of potential associations – textual, iconographic and musical. For the viewer and the scholar, interpretation is a matter of deploying the resources of art, literature and scholarship in order to establish the potential meaning of an object that exists in a complex nexus of association with a wide range of cognitive artefacts. Unlike the Raphael, Poussin’s painting ‘is not an archive of facts or propositions, but a participant in the recalibration of a cognitive system that embraces as many viewers as avail themselves of its affordances’ (p. 232). A wider nexus of comparison is then created by means of the introduction of Cicero’s De oratore, a dialogue on the education of the ideal orator. This is a work in which an initially isolated authorial voice recreates the experience of dialogue to minimise the cognitive disadvantages of solitude, as imaginary conversations between real historical individuals construct an argument whose nature differs significantly from that of first-person exposition. As in the Protagoras, individually situated and group-based or dialogic forms of cognition exist in tension, in a written work that perhaps reflects the collaborative nature of writing in Cicero’s own day and certainly exemplifies the role of the written text not simply as an inert mechanism for the storage of information, but ‘as a tool for generating the ability to speak well’ (p. 237). According to Habinek and Reyes, De oratore ‘contains within

36

dou glas c ai rns

it the divergent approaches to cognition that characterise The School of Athens, Numa and Egeria and the Protagoras of Plato’ (p. 238). Habinek and Reyes read their sources in a way that might potentially inspire readers of this volume (and the series to which it belongs) more generally. If it is true that human cognition involves complex and multifaceted relations between brain, body, other human beings, the external, physical environment and a wide range of objects and technologies, then this must always have been true, for as long as there have been human beings. If that is the case, then we can expect the distributed nature of cognition to inform all the cognitive artefacts that the ancient Greeks and Romans have bequeathed to us. This will be true whether or not the ancient modes of thought that lie behind those artefacts explicitly reflect positions – as indeed many of them do – that we might regard as in some sense recognising that fact. This, in turn, will mean that an understanding of the distributed nature of cognition should inform all that we do, as students of classical culture, in our attempts to understand and elucidate that culture. A comprehensive history of distributed cognition in classical antiquity will thus always be beyond our grasp, and this volume is merely the beginning of a project that will necessarily remain partial and unfinished. But if we read the chapters in this book (and in its companion volumes) dialogically and comparatively, exploring the ways in which the chapters can be used not only to shed light upon but also to raise questions about each other, we shall surely acquire a sense of the rich variety of ways in which notions of cognition as embodied, embedded, extended and enactive not only leave their traces in classical culture and thought, but also serve as a source of models and methodologies by which we can interrogate and explicate our ancient sources.



physic al sciences 37

3 P h y s i ca l S c iences: P t o l e my ’ s E x t e nd ed Mind Courtney Roby

Introduction Not much is known about Claudius Ptolemy’s life, beyond what can be deduced from the astronomical observations recorded in his Syntaxis mathematica (better known as the Almagest), made between 127 and 141 ce at Alexandria, and the inscription he seems to have set up at nearby Canopus in 147.1 Otherwise he remains a nebulous figure, neither filling his own books with colourful anecdotes of his professional life as did the physician Galen, his near contemporary, nor enjoying the post-mortem attribution of fabulous biographical details as did other towering figures of ancient science like Pythagoras or Archimedes. The afterlife of his work was, however, extraordinarily robust. He became the author par excellence in astronomy and harmonics, to the extent that the works of his predecessors were largely obliterated in the wake of Ptolemy and his commentators. Through the extraordinary breadth of his surviving texts persists a recurrent line of interest in putting the phenomena observed in the world to work in scientific theorisation, most often of a highly quantified kind. In particular, he is persistently interested in the natural and artificial instruments through which we perceive the world, in terms of how the natural ones may be aided and the artificial ones refined so as to provide the most reliable data. Ptolemy’s inquiry into the mechanisms through which scientific observations are actually performed might be considered a second-order investigation with respect to the natural systems that are the explicit subjects of his texts. Yet such an investigation is mandatory for the precisely calibrated observational work Ptolemy recommends to his reader. Far from simply listing the procedures and apparatus of scientific investigation, Ptolemy engages deeply with questions about the kinds of cognitive work they require and in turn perform. The body is presented as a composite of sensory and motor systems together with an array of faculties of judgement, all complexly intertwined and none functioning alone. The eye and ear are not mere relays for  1

For the little information about Ptolemy’s life and a good capsule guide to his works, see Folkerts 2002; Jones 2008.

38

c ou rtney ro b y

sensory data to a separate cognitive faculty, but are themselves an intrinsic part of the process of assessing geometric diagrams, harmonic tones, celestial bodies and other sensible phenomena. The eye may draw as well upon data mediated by computational tables, and Ptolemy describes the eye in action, moving through such tables to seek data or perform calculations. The sensorimotor feedback loops are then closed by further bodily activities, such as an eye moving over a geographical table of latitudes and longitudes that then guides the map-drawing hand, or the fine-tuning performed by the hand of a musician or harmonicist in response to aural feedback, with the manipulations of astronomical and harmonic instruments during the observational process acting to refine the data that will yield new tables. Ptolemy thus conceives of the body’s own sensory and motor apparatus as intrinsic to the cognitive processes that yield scientific observations. Indeed, he portrays those cognitive processes as extending even beyond the bounds of the body itself, to include observational instruments and data tables constructed so as to take on analytical tasks. Such external instruments are not merely convenient but vital to the scientist’s work: while the body’s sensory and motor faculties have their own special abilities to identify observable phenomena such as correct harmonic intervals (or to produce them by playing an instrument or singing), nevertheless they can be fooled. Artificial instruments assist the senses and hands in their observational work by ‘straightening’ (kanonizein) and so ratifying the data. Such instruments can also work as placeholders, memory for small amounts of data, for example by setting swivelling latitude and longitude markers on an astronomical instrument or a set of bridges on a stringed harmonic instrument. Larger amounts of data are stored in, retrieved from and manipulated by a variety of data tables designed to organise information for a spectrum of quite specific tasks. In many cases, the organisational scheme is designed to allow numerical operations to be carried out more or less automatically by the hand or eye, without referring them to any separate calculating faculty in the brain.

Ptolemy’s psychē Ptolemy’s work On the Criterion of Truth (henceforth Criterion) interrogates the faculties that the mind uses to assess the accuracy and precision of logical arguments and observed phenomena. The Criterion is generally supposed to be one of his earliest works, suggesting that the mechanisms of cognition he describes there can be thought of as foundational to the later works.2 Ptolemy sets the stage for his ­­analysis of the psychē’s faculties with an elaborate three-part analogy that compares the mind’s work of assessment to the processes of judging court cases, and to the tools for judging an object’s physical attributes such as magnitude, weight and position (Criterion 3.5–7.6).3 Ptolemy’s analysis is as straightforward as it is detailed, and it will not be necessary here to include a full account.  2  3

The evidence for the place of the Criterion in Ptolemy’s timeline is discussed at Feke 2012: 593–4. References to the Criterion here are taken from Lammert 1961: 3–25, where it appears under the Latin title De judicandi facultate et animi principatu.



physic al sciences 39

Ptolemy’s multi-stage analogy includes not only the matter being judged (which is analogised both to an action evaluated in court and to a physical object), and not only more or less abstract logical means used to settle the question, but also a host of concrete actors and objects in the world. The mind’s ‘instrument of judgement’ (τὸ δι’ οὗ κρίνεται) finds a direct analogy in standards for certifying length measurements (e.g. the cubit), scales, or a ruler used to measure the position of an object. Its ‘agent of judgement’ (τὸ κρῖνον) is compared to the human actors entrusted with carrying out and verifying those measurements: land-surveyors, officials in charge of weights and builders. Although the ‘means of judgement’ (τὸ ᾧ κρίνεται) is matched in the courtroom scenario with the relatively abstract ‘law’ (nomos), its analogues in the measurement scenario are decidedly concrete: physical movements such as the manual manipulation of a ruler along a thing being measured, the bending of a scale, or the act of placing a ruler or stretching a measuring line to find an object’s position. Ptolemy then immediately delves further into the mind’s functional details, arguing that cognition depends on sensory perception, which comprises sensory organs and the impressions (phantasiai) produced upon the intellect (Criterion 5.18– 6.11).4 The ‘instrument’ metaphor for cognitive activity which he employed in the first part of the text is here expanded, further illuminating just how deeply Ptolemy roots it in the physical world. He defines sense organs as bodily instruments (organa) that come into contact with sensible things, yielding an impression through phantasia, which relays that impression to the intellect. The intellect in turn is able to retain and remember the objects of transmission thanks to the faculty of conception (ennoia). The refined version of the courtroom analogy includes more specific pairings of the elements of proving and judging a legal case alongside the psychē’s components. The parties being judged are compared to the objects of sensory perception. For instance, the documentary memoranda involved in the case are compared to the sensory organs, the advocates to sense-perceptions, their speeches to the phantasiai, intellect (nous) to the judge, and so on. The physical concreteness of the functional components and the cooperative distribution of their efforts are once again striking, particularly in the case of the shift from the more abstract ‘law’ to the more concrete ‘memoranda’ (hypomnēmata). Ptolemy’s proposed decision-making apparatus is quite decentralised; even if the psychē’s ‘ruling part’ (hēgemonikon) is meant to serve as a single final arbiter, the process relies on the interoperation of multiple psychic faculties.  4

Phantasia might be generically defined as an image in the mind, but the term has a nuanced and varied history (cf. Huitink, this volume). For Aristotle, it is the capacity for imagination, whereas for the Stoics it was what Long refers to as a ‘sensory “mode of representation”’ (1996: 269). On rhetorical phantasia, see Webb 2009: 107–30. Webb distinguishes between rhetorical and (largely Stoic) philosophical deployments of the word at 115–19. A throughline between the term as it is used in Aristotle’s On the Soul to mean something like ‘imagination’ and by Plato and some Neoplatonists to refer to divinely inspired visions, with some excursions into Neoplatonic ‘mathematical projection’, is drawn in Sheppard 2014.

40

c ou rtney ro b y

Still, though the psychē’s evaluative mechanisms are here separated into a variety of faculties and their reliance on the sensory organs is emphasised, it is not yet clear that Ptolemy’s model of the psychē could in fact be said to be ‘distributed’ in a sense that really incorporates body as well as brain. For that we must turn to a later section of the text, where he begins to discuss how body and psychē operate together. Ptolemy defines the body as consisting of perceptible things like flesh and bone, whereas the psychē is inherently imperceptible, and can only be grasped through the sensory and motor effects it causes in the body (Criterion 11.9–13). The mind-body differentiation here is a functional as well as a structural distinction, as Ptolemy articulates that not only do we think through the psychē rather than the body, but also it is through the psychē that all sensory and motor activities occur (Criterion 11.22–4). Yet though unlike the body the psychē is not directly perceptible, still it is material. Without the body to contain it, says Ptolemy, the psychē would immediately diffuse into its component elements because its structure is so fine, as when water vapour or air escapes from a container. Upon separation from the body, the psychē would no longer exhibit the movements it engaged in when it was a part of the synthetic human being (Criterion 12.1–6). Without the psychē, the body has no capacity for sensation or movement, indicating that those faculties are proper (oikeia) to the psychē. However, the operation of those faculties is dependent on their contiguity (synochē) with the body (Criterion 12.9–12). What does all this indicate about the make-up and operational functions of the psychē? Ptolemy allows two possibilities for its constitution (Criterion 12.13–13.2), underlining the distinction between them with a ‘men . . . de’ alternative. According to one account, the psychē consists of different kinds of material, whose particular qualities shape the parts of the body that contain them so as to make them capable of cooperating with the psychē’s powers. According to the other, the psychē’s material has but a single nature, and its varied powers are the result of differences in the surrounding parts of the body (as the same breath can produce different sounds going through differently shaped bodily organs). Whichever account we choose, Ptolemy insists that the psychē is indeed distributed throughout the body, and that it comprises multiple components either intrinsically or through its multifaceted interoperation with the body. The operation of sense and thought depends on the elemental construction of their apprehending agents, as different combinations of elements are more or less susceptible to transmit the movements of the psychē (Criterion 19.23–20.2, 20.9–13). Those movements will differ from one sense to another, and hence different bodily organs are better able to transmit different information. Touch, being the most material sense of all, extends wherever flesh and blood are found, whereas the others are restricted to parts of the body that are moister and more penetrable (Criterion 20.16–20). Sight and hearing are particularly closely related to intellection, and are concomitantly located nearest the brain. The psychē’s sensory and motor faculties operate through diverse types of motions that occur throughout the body, including the sensory organs, liver and heart (Criterion 21.20–2). The head is of special importance for Ptolemy’s psychē; he asserts



physic al sciences 41

that the brain is the location of many important faculties ranging from intellection to control over the reproductive seed, and says that indeed one might choose to call this the hēgemonikon or ‘ruling part’ of the psychē (Criterion 21.1–6).5 But in a remarkable departure from the mainstream philosophical and medical discussion of the hēgemonikon, he suggests that one might alternatively define a hēgemonikon as the thing in charge of any function. The psychic powers so defined would each be in control of their proper function, so that hēgemonika (Ptolemy uses the plural here) would be distributed throughout the body (Criterion 21.10–12). Ptolemy’s theorising about the psychē does not end with the Criterion. The subject re-emerges with particular vigour in the Harmonics, where he picks up on many of the focal themes and problems of the Criterion, adjusting and refining them along the way.6 The Harmonics draws an elaborate picture of the psychē and its faculties, matching each to specific virtues, domains of human activity and mathematical structures in music and astronomy. The details of that system need not concern us here.7 The important thing is the very real connection Ptolemy draws between the psychē and harmonic structures of musical harmonies and celestial bodies’ recurrent motions, thus situating the operations of the psychē in an extraordinarily broad and deep cosmic context. This psychological-celestial excursion may seem like a puzzling conclusion to a work which otherwise strongly emphasises balance between the logical and the empirical, but it fits easily within the contemporary system of mathematical astrology to which Ptolemy subscribed (and contributed, through his Tetrabiblos).8 The cognitive system Ptolemy describes in the Harmonics is predicated on the presence of a faculty of ‘attunement’ (to hērmosmenon) which is especially visible in the human psychē and in the motions of celestial bodies, because they are nature’s most highly perfected and rational entities (Harm. 3.4.25–9). The faculty of harmonia is likewise correlated with reason and proportion, and hence confers orderliness (eutaxia) upon those who commit to mastering its theory and habituating themselves  5

The notion that a centralised hēgemonikon was responsible for the decisions of rational creatures and the drives to action of all living things came into its own in Stoic philosophy during the Hellenistic period (cf. Gill, this volume), but its bodily location and mechanisms of action were the subject of ongoing debate during Ptolemy’s lifetime. For Stoic ideas about the hēgemonikon and its relationship to the psychē, see Long 1996. The physician Galen’s On the Opinions of Hippocrates and Plato offers sustained discussion of the hēgemonikon and the debate over whether it was located in the brain or the heart (including witheringly sardonic commentary on his cardiocentrist opponents). For the text, translation and commentary, see De Lacy 1978. For discussion of questions in the text of particular relevance here, see Tieleman 1996; Rocca 2003; Gill 2007; Schiefsky 2012. For further comparison of philosophical and medical ideas about the relationship of body and soul from the Hellenistic period to Galen, see von Staden 2000.  6 The relationship between the psychological theory Ptolemy espoused in the Criterion and the one that appears in the Harmonics has recently been perceptively analysed in Feke 2012. The Greek text of the Harmonics I refer to here is Düring 1930: 2–111.  7 For insightful analysis of Ptolemy’s ambitious comparisons, emphasising how they function as models, see Barker 2006a.  8 On Ptolemaic astrology and its context, see especially Taub 1993. 

42

c ou rtney ro b y

to its practice (Harm. 3.4.4–9). Even the faculty of reason is not isolated from the world. Reason functions to organise auditory evidence as much through manual practice and sensation as through theory (Harm. 3.3.38–42): διά τε τῆς θεωρητικῆς τῶν συμμετρίων εὑρέσεως παρὰ τὸν νοῦν, καὶ διὰ τῆς χειρουργικῆς αὐτῶν ἐνδείξεως παρὰ τὴν τέχνην καὶ διὰ τῆς παρακολουθητικῆς ἐμπειρίας παρὰ τὸ ἔθος. through theoretical discovery of proportions by means of intellect (nous), through their manual demonstration by means of skill (technē), and through experience in following them by means of habit (ethos). ‘Habit’ here is matched with ‘nature’ (physis), and as Barker notes, specifically to the Aristotelian idea of a natural inclination of things toward whatever constitutes excellence for them, without the explicit guidance of reason.9 Though the harmonicist ultimately seeks to express harmonic relationships quantitatively, he cannot reach that goal without the concerted cooperation of the sensory organs and the manual operations on material instruments that produce the auditory data he works with. Indeed, the fluid bond between theory and practice is what allows scientific investigations to access mathematical truth. Mathematics, defined by Ptolemy as the one exact science (epistēmē) that encompasses all the rest, encompasses not only theoretical contemplation of its objects, but also demonstration and practice (Harm. 3.3.44–8). The parent discipline of all exact sciences, it is built on a foundation of bodily sensation and habituation. Sight and hearing take pride of place in a cognitive system that is capable of scientific insight precisely because it is rooted in concrete habit and practice.

Textual Instruments The engagement Ptolemy envisions between cognition and the world thus capitalises on a generously embodied model of sensory and motor activities, ultimately involving the whole body through the widely distributed sense of touch. Indeed, the continuity between theorising mind and observed phenomena goes even beyond the boundaries of the body, thanks to an array of observational and informational instruments essential to scientific work as Ptolemy envisions it. The sensory organs serve as a clearing-house for data, but when nuanced distinctions need to be made, artificial instruments can augment the body’s own sensory tools. For example, the unaided eye may judge a figure to be a straight line or a circle even when it merely approximates that shape, but will make more accurate determinations if instruments like a ruler or a compass can be conscripted to help with  9

Aristotle Ph. 2.1, 2.8; Metaph. 12.9; Eth. Nic. 10.7; Barker 1989: 372 n. 30. For a more detailed examination of the relationship between Aristotle’s physics and epistemology and Ptolemy’s, see Taub 1993.



physic al sciences 43

the judgement (Harm. 1.1.52–6). Ptolemy compares this process to the way instruments such as the monochord (i.e. the single-stringed version of the harmonic kanōn) can refine and structure the ears’ perception of harmonic relationships. Really precise scientific observational work will require the acquisition of ‘some rational criterion via the proper instruments’ (τινος πρὸς ἐκεῖνα κριτηρίου λογικοῦ διὰ τῶν οἰκείων ὀργάνων, Harm. 1.1.53–4). More particularly, instruments such as the kanōn serve to ‘straighten’ (kanonizein) the data provided by the senses, rendering it suitable input for developing harmonic, astronomical and other scientific theories (Harm. 1.2.2).10 Once sensory input has been distilled through instruments into precise and accurate data, a new challenge presents itself: organising that information in such a way that it can most readily serve as input for new operations of observation and calculation. Ptolemy’s geographical, harmonic and astronomical texts are dominated by lengthy data tables (kanones), which Ptolemy describes being used to mediate between the world and the observer in a similar way to other types of instruments.11 Thanks to their carefully designed organisational principles, the tables function as textual ‘instruments’ in their own right, used to test and ratify previously discovered information as well as to guide the discovery and integration of new material.12 The term ‘data’ may suggest information detached from the world where it was originally obtained, especially since we often now use black-boxed automated computational systems to process numerical information. For Ptolemy’s reader, the computational work is less seemingly transparent, and Ptolemy offers copious instructions for the best way to carry it out. Usually the tables follow exhaustive analysis of the mathematical theory behind the motion of celestial bodies, map projection schemes or harmonic relationships (the astronomical Handy Tables, being rather light on theory, is exceptional in this regard). Thus the principles that ground the derivation of the data have already been made transparent. Even in the Handy Tables, Ptolemy often pairs explanations of how the values are obtained through a method involving diagrams (grammikōs) with a second method that operates through numbers alone (arithmētikōs). The tables are used for both, as Ptolemy’s instructions point the reader toward the appropriate column for each stage of the calculation. The diagrammatic version of the solution uses numbers looked up in the table to correctly position and scale diagrammed lines indicating astronomical features such as the eccentres and epicyles, while the numerical approach relies on a more complex array of arithmetical manoeuvres to perform operations on numbers rather than diagrams. 10

On the kanōn, see also Barker 2006b: 192–229; Creese 2010: 68–72, 283–99, 304–19. On the relationship between instruments, diagrams and tables in Greek harmonics, see Creese 2010: 22–80. 12 Many digitised manuscripts of Ptolemy’s works including data tables can now be viewed online through their holding libraries. Links to selected digitised manuscripts of the Almagest, Harmonics, Geography and Handy Tables will be found in the Appendix. 11

44

c ou rtney ro b y

Hutchins’s work on the use of navigational charts helps to clarify the significance of the operations performed by Ptolemy’s data tables. As Hutchins notes, navigational charts present spatial information analogous to features of the world, but they additionally serve as computational devices in their own right. Hutchins (1995: 61) describes them as more akin to a coordinate space in analytic geometry than to the sort of simple map I may produce to guide a new acquaintance to my office . . . a navigational chart is an analog computer. The charts Hutchins has in mind are used to solve navigational problems using techniques designed specifically for use with the chart. While those same problems could also be solved symbolically (this is the kind of work that went into constructing the chart in the first place), the chart enables a different cognitive approach to the problem at hand. The navigator who uses the chart finds the information he needs organised optimally for the task at hand, readily available for navigation without the user having to re-perform the symbolic calculations. The chart is, as Hutchins points out (1995: 96), an artefact that ‘crystallises’ knowledge and technique in its own physical structure. Like the astrolabe, another kind of ‘analogue computer’ useful in Western navigation, the navigational chart is a ‘durable external representation’ of knowledge that is difficult for a purely mental representation to render precisely. Ptolemy’s tables, like the tabular instruments Hutchins describes, are designed as ‘durable external representations’ of knowledge which can interoperate with a human user. They perform a role like that Hutchins elsewhere asserts for such ‘blends’ of conceptual and material structures, altering the balance of cognitive input and computational output so that the ‘blend’ produces considerably more complex computational results for a given cognitive effort (Hutchins 2005: 1562). Putting these external representations to work can involve the body in various ways. Ptolemy’s Geography is not itself principally designed to transmit maps, though there has been considerable discussion of the possibility that ancient versions of the text might have included pictorial maps, as the medieval and later copies often do.13 Ptolemy explains that maps are tricky to transmit accurately, being subject to accidental alteration, so that the geographer is better off trafficking in tables of latitudes and longitudes. Using some projection scheme, these tables can then be reconstructed into maps and turned into a visual experience by the reader. Ptolemy analyses several different projection methods, which require balancing mathematical complexity with the resulting map’s precision and accuracy. The projection method he recommends is more accurate, but disallows some graphical tricks that make producing a map using a competing method seem easier, like proceeding ‘by shifting the ruler along and juggling it back and forth, 13

See, for example, Dilke 1985; Brodersen 1995; Talbert and Brodersen 2004; Talbert 2008; Mittenhuber 2009. The text of the Geography referred to here is Grasshoff and Stückelberger 2006: i. 50–470 and ii. 478–920.



physic al sciences 45

as long as just one of the parallels had been drawn and divided up’ (ἀπὸ τῆς τοῦ κανονίου παραγωγῆς τε καὶ παραφορᾶς, ἑνὸς μόνου τῶν παραλλήλων γραφέντος καὶ διαιρεθέντος, Geog. 1.24.28.2–4). That approach will no longer work for his recommended method, which involves drawing the meridian lines so that they curve around the centre meridian. While Ptolemy strongly endorses the more complex and precise approach, he allows that some readers might instead prefer one that replaces mathematical with manual difficulty, having been ‘swayed to the more convenient one by its easiness’ (ἕνεκεν τῶν ἐπὶ τὴν προχειροτέραν αὐτῶν ὑπὸ ῥᾳστώνης κατενεχθησομένων, Geog. 1.24.29.3–4). In fact, both approaches to drawing the map recall Hutchins’s descriptions of certain actions a ship’s plotter carries out as he uses a navigational chart to adjust the ship’s trajectory.14 The plotter manipulates a set of dividers and a protractor on a navigational chart with a scale bar to make ad hoc adjustments to the ship’s path. Hutchins (2010: 442) concludes from the complex of bodily actions and verbal utterances performed by the plotter that he is in fact engaged in a multimodal, temporally extended enactment. According to Hutchins (2010: 443), the temporal extension suggests that a visual/motor memory of an activity performed in the subjunctive mood a few seconds in the past could somehow combine with current visual/motor perception to produce visual/motor anticipation of activity projected to take place a few seconds in the future. Seen in light of Hutchins’s work, the ‘ruler-juggling’ Ptolemy mentions acquires a new significance: the cartographer engaged in ad hoc manipulations of his drafting tools is responding flexibly to the ‘subjunctive mood’ of his unfolding enactment rather than wallowing in desultory indecision. Ptolemy’s aim to assert superior precision and reliability for his preferred method leads him to downplay not only the ruler-juggler’s ‘subjunctive’ adjustments, but also the fact that a cartographer who chooses Ptolemy’s method has to carry out precisely the same sort of adjustments as he draws the curves of the meridian lines, which likewise must be estimated as a smoothed concatenation of arcs. The ruler (kanonion) that has to be juggled around to make the map using Ptolemy’s first method represents just one possible type of interaction between body and instrument. Elsewhere in this section, Ptolemy describes how the viewer’s visual experience will vary depending on the positioning of the eye with respect to the meridian. In this case he is describing a method of projecting latitudes and longitudes on to a planar map (pinax) that will maintain the greatest fidelity to the way they would be arranged on a sphere (Geog. 1.24.10.1–7): Ἔτι δ’ ἂν ὁμοιότερόν τε καὶ συμμετρότερον ποιοίμεθα τὴν ἐν τῷ πίνακι τῆς οἰκουμένης καταγραφήν, εἰ καὶ τὰς μεσημβρινὰς γραμμὰς καταλάβοιμεν τῇ 14

The plotter’s operations are described in detail in Hutchins 2010.

46

c ou rtney ro b y

φαντασίᾳ τῶν ἐπὶ τῆς σφαίρας μεσημβρινῶν γραμμῶν, ὡς τοῦ ἄξονος τῶν ὄψεων διήκοντος ἐν τῇ θέσει τῆς σφαίρας, διά τε τῆς πρὸς τῇ ὄψει τομῆς τοῦ διχοτομοῦντος τὸ μῆκος τῆς ἐγνωσμένης γῆς μεσημβρινοῦ καὶ τοῦ διχοτομοῦντος αὐτῆς τὸ πλάτος παραλλήλου, καὶ ἔτι τοῦ κέντρου τῆς σφαίρας, ἵν’ ἐξ ἴσου τὰ ἀντικείμενα πέρατα ταῖς ὄψεσι καταλαμβάνηται καὶ φαίνηται. We could make the drawing of the world on the map more similar and proportional, if we were to comprehend the meridian lines using the phantasia of the meridian lines on the sphere, with the axis of the visual rays extending through to the place on the sphere through the intersecting cut toward the eye that bisects the length of the meridian of the known world, and the parallel that bisects its width, as well as the centre of the sphere, so that the opposite bounding points would be grasped by the visual rays and perceived equidistantly. Here the cognitive capacity for phantasia, the eye and the map itself collaborate for a maximally accurate viewing experience. Ptolemy’s model here relies on an ‘emissionist’ theory of vision, in which objects in the world are seen by means of a cone of visual rays or ‘flux’ projecting from a point in the interior of the eye.15 While this visual theory perhaps strikes us as strange because it relies on the idea of rays projecting outward from the eye rather than light entering the eye, something interesting happens if we adapt to that strangeness, by inverting the spread of the visual rays into the visual fields we speak in terms of today. In particular, Ptolemy is concerned about the shape of the ray-spread’s impact on the globe, which he notes depends on where the eye is positioned and how it is oriented with respect to the lines painted on the globe. The viewer’s experience will change when viewing a given point from the side rather than head-on, so Ptolemy here stipulates that the specific viewing situation he is trying to replicate on the planar map is the one where the axis of the cone of visual rays hits this specific intersection of latitude and longitude head-on. Ptolemy articulates the dependence of visual experience on the eye’s location and orientation with respect to the viewed point, anticipating recent research on vision by O’Regan and Noë. They assert that ‘seeing is a way of acting. It is a particular way of exploring the environment’ (O’Regan and Noë 2001a: 939). Their focus on the ‘sensorimotor contingencies governing visual exploration’, such as the dependence of the visual field on the deformations of the eye as it moves in its socket (O’Regan and Noë 2001a: 941), is particularly resonant with Ptolemy’s account. Though he uses a very different language, Ptolemy too is concerned with the effects of sensorimotor contingencies on our explorations and subsequent descriptions of the world. 15

The ‘emissionist’ theory is opposed to the ‘emanationist’ theory developed by Empedocles and later favoured by Democritus (and subsequently by Epicureans), who supposed that atomically thin layers of objects are projected as images (eidōla or simulacra) into the eye; see Rudolph 2016: 44–52. On ancient theories of vision, see also Smith 1988, 1999; Lehoux 2007; Netz and Squire 2016.



physic al sciences 47

The eye that views the map is not modelled reductively as a point, but remains a spatially extended organ of sensory experience; hence the experience will depend on the organ’s orientation. A map that will offer a well-controlled experience of the latitudes and longitudes Ptolemy is trying to transmit and project accurately will need to be set up in a way that accounts for the sensorimotor contingencies of the embodied experiencer. The integration between body, mind and paper instrument that Ptolemy envisions can in fact be so deep that the hand and instrument seem to function together seamlessly, without referring cognitive activity to a separate calculating centre in the brain. The case for such a connection is made particularly clearly when Ptolemy explains how a table of numbers can be turned into a graphical map. A carefully designed table can be laid out in such a way that it creates a nearly transparent interface between itself and the user (Geog. 2.1.4–5): Προειλόμεθα δὲ τάξιν τοῦ περὶ τὴν καταγραφὴν εὐχρήστου πανταχῇ ποιούμενοι πρόνοιαν, τουτέστι καθ’ ἣν ἐπὶ δεξιὰ ποιησόμεθα τὰς μεταβάσεις, ἀπὸ τῶν ἤδη κατατεταγμένων ἐπὶ τὰ μηδέπω τῆς χειρὸς ἐκλαμβανομένης. Τοῦτο δὲ γένοιτ’ ἄν, εἰ γράφοιτο τά τε βορειότερα πρότερα τῶν νοτιωτέρων καὶ τὰ δυσμικώτερα τῶν ἀπηλιωτικωτέρων, ὅτι πρὸς τὰς τῶν ἐγγραφόντων ἢ ἐντυγχανόντων ὄψεις ἄνω μὲν ἡμῖν ὑπόκειται τὰ βορειότερα, δεξιὰ δὲ τὰ ἀπηλιωτικώτερα τῆς οἰκουμένης ἐπί τε τῆς σφαίρας καὶ τοῦ πίνακος. We settled on an arrangement for drawing, giving forethought to convenience in every way; that is, one in which we make our movements toward the right with the hand shifting from the things already set down to those not yet established. This would come about by drawing the northern elements before the southern ones, and the western before the eastern, because in drawing or visual encounters it is established for us that northern things are on the upper side of the world, and eastern things to the right, both on a sphere and a planar map. Ptolemy’s preferred organisational scheme capitalises on the fact that a typical Greek or Roman viewing a map can be assumed to share a common spatial orientation, where north is up and east is right. The availability of this assumption can be compared to Hutchins’s remarks (2005: 1569) about the importance of cultural conventions for conceptual models, which are easily overlooked but are crucial to the models’ efficacy. The table is designed to structure the map’s creation so that the hand, working more or less autonomously, simply scans along the page inserting whatever features arise. If the hand were not enabled by the table to function this way, the map would instead first have to be visualised as a whole, and the drawing organised separately, for example from large-scale features such as continents down to small-scale ones such as towns. Certainly it is possible to draw a map using this second approach; it might well even seem more natural. However, Ptolemy is more interested in devising a maximally foolproof way of transmitting geographical

48

c ou rtney ro b y

information, which means among other things attempting to reduce the human map-drawing apparatus down to its most essential element, the hand. The hand and eye will often be called upon to work together, whether a data table is being used to transform existing information into a new form or to guide exploration and collection of new data. The hand-eye link shows up with particular vividness in the star catalogue section of the Syntaxis mathematica, whose information is designed to accompany and guide a journey of the eye from one star in the constellation to the next. The exact layout of the tables differs somewhat between the surviving manuscripts, but for the most part the columns proceed, from left to right, by indicating each star’s number in the constellation, the description of the visual feature of the constellation it corresponds to, its celestial longitude and latitude, and its magnitude according to Ptolemy’s five-level classification.16 The numerical progression at the far left enforces visual tracking along a certain path; the eye must not be allowed to rove willy-nilly through the constellation. The column, which correlates each numbered star to a feature of the imaginary picture that the constellation represents, confirms the pathing enforced by the numbering system, while using the mnemonic of the image to ease the way through it. The latitude and longitude then recontextualise the stars, situating them against a mathematicised skyscape, where they can be used for tasks such as tracking the movements of the planets. An initial glance at the complex, close-packed array of numbers in Ptolemy’s star catalogues and other numerical tables evokes wonder at how such tables could have proved useful. Particularly when the books containing the tables were made of parchment and each page thus presented a non-trivial expense, the tables were often compacted to occupy minimal space. Even information-processing aids, like the marginal labels for each constellation that mark the star catalogues in some manuscripts, can only improve legibility to a limited extent.17 While later copies of the works on paper, both manuscript and printed, often introduced more white space, this tactic still does not render the tables’ contents easily interpretable. Yet they were unquestionably useful to astronomers and astrologers alike for centuries; they were recopied, epitomised and expanded in countless instantiations. What could transform these cramped arrays of numbers into a useful tool for the astronomer? The likeliest answer seems to lie in the astronomer’s expert eye. The tables are inscrutable to most of us because we are not used to navigating them, but to an astronomer trained in making the associations between table data and celestial phenomena, the path through the numbers would have been familiar. While this would be difficult to test empirically since today even professional astronomers are not trained in using such tables, a comparable phenomenon has been tested on the 16

On the contents of the catalogues in different manuscripts, with plates showing a variety of table layouts, see the meticulously researched Peters and Knobel 1915. For a broader array of tables from several of Ptolemy’s works, see the Appendix. 17 For these, see Peters and Knobel 1915, plate D.



physic al sciences 49

visual processing of geometry problems (Epelboim and Suppes 2001). Subjects with different levels of geometrical proficiency were asked to solve a geometry problem using a diagram while their eye movements over the diagram were tracked. While the novices’ focus roamed aimlessly over the space of the diagram and the supervening verbal statement of the problem, the experts’ eyes repeatedly traced a path which was essential to solving the problem but not drawn in the diagram, and at the same time uttered words confirming the operations their eyes were carrying out. The same kind of expert vision could apply to Ptolemy’s astronomical tables; the astronomer’s eye is habituated to find the correct path through the complex spatial arrangement. The astronomer’s expert operations on the table might also be compared to Hutchins’s observations of expert navigators at work. The astronomer’s acculturation to astronomical data tables, and the geometrical and arithmetical operations associated with them, creates the opportunity for ‘the coordination of low-level perceptual and motor processes with cultural materials to produce particular ­­higher-level cognitive processes’ (Hutchins 2010: 434). Hutchins’s plotter and navigator have undergone extensive training and practice that render their charts and tools capable of yielding precise high-level information through apparently lowlevel engagements with the tools and available information. Likewise, Ptolemy’s astronomer has through his training acquired an experienced eye whose confident movements through the data table and over the celestial bodies represented in the table are stabilised by the conceptual expertise and experiential comparative capacity inculcated into him.

Astronomical Instruments The movement of eye and hand on a careful path through a tabular instrument mirrors analogous movements of observation and adjustment that those parts of the body carry out on a physical observational instrument. Ptolemy describes several astronomical instruments in turn, ranging from the relatively mechanically complex armillary sphere to a sighting device with slits designed to measure parallax.18 For the most part he provides minimal details about the craft processes involved in constructing astronomical instruments; he devotes most of his attention in this respect to the features that must be precisely adjusted so the device will function smoothly. The nesting rings of the armillary sphere, for example, must be manufactured in such a way that they fit closely together but still spin easily relative to one another; the plaque that can be used to measure the angle of the sun from its zenith along the meridian has to be precisely positioned using shims, and the rods of the parallactic instrument have to be able to rotate freely. However, these details 18

See, for example, Syntaxis mathematica 1.1.64–66.4 on the equinoctial rings, 1.1.66.5–67.13 on the plaque, and 1.1.403.1–405.17 on the parallactic instrument. On the instruments of ancient Greek astronomy, see Aujac 1993; Evans 1999; Taub 2002; Evans and Carman 2014. The Greek text of the Syntaxis referred to here is Heiberg 1898–1903: i.1, 3–546; i.2, 1–608.

50

c ou rtney ro b y

are relatively sparse, particularly compared to the more exhaustive attention their analogues receive in Ptolemy’s Harmonics.19 He devotes more minute attention to the operation of the devices, which he describes from a decidedly embodied perspective. The set-up and operation of the armillary sphere are obviously dependent on the specific location of the observer in space and time. On a more local level, one must wait and orient oneself correctly so that the phenomena of interest will be visible. If the sun is going to be used as the orienting object, the observer must wait until the sun and moon are simultaneously above the earth, then turn the device’s outer ring so it marks the sun’s position by its intersection with the ecliptic ring, and then rotate the ring through the poles so that the intersection point faces the sun relative to the observer’s perspective. The observer thus adjusts the instrument to reflect his particular position in space, from which he looks out to see the particular instantiation of celestial phenomena determined by that position. The degree to which observation is body-bound becomes still more apparent as Ptolemy moves on to describe a more general case of a similar problem, where the sun is replaced by another star of interest. In that case, the outer ring of the instrument is first adjusted to the star’s hypothesised position on the ecliptic ring, and then rotated so that ‘when one eye is affixed’ (τοῦ ἑνὸς τῶν ὀφθαλμῶν παρατεθέντος) to a face of the outer ring the star will look like it has been ‘glued’ (kekollēmenos) to the ring’s inner and outer surfaces. Once this act of situated viewing has been accomplished, the inner ring is then rotated so that the target object (Ptolemy suggests the moon as an example) can be sighted through the innermost ring’s sighting holes while the other star is being sighted as previously described (Syntaxis mathematica 1.1.353–4). The more general problem is solved in essentially the same way as the sun-­­ specific case (as becomes clear near the end of the passage), but here the interoperation of different parts of the body with the instrument is made more explicit. When the astronomer rotates the armillary sphere’s rings to the target objects’ positions, he uses the instrument to ‘bookmark’ that information so that he no longer needs to keep track of it. The calibration itself involves not only the information the astronomer previously calculated and stored mentally, but also the bodily process of shifting his eye position so that the star appears ‘glued’ to the instrument’s sights. The adjusted instrument thus embodies the information that represents the astronomer’s position and visual experience as well as the target of sighting. In relocating this information to the instrument, the astronomer and instrument become what Clark and Chalmers term a ‘coupled system’ whose ‘behavioral competence will drop’ precipitously if the instrument is removed (Clark and Chalmers 1998: 8). The astronomer’s work thus requires some tricky coordination, as he juggles his sighting on the instrument and the manual adjustments he might be making to it, or parses an astronomical table while making calculations by hand. In addition, 19

On the differences between how instruments are presented in the two works, see Creese 2010: 68–72.



physic al sciences 51

these activities might be transferred to the auditory and verbal domain while observations were ongoing, while using an assistant or slave as a social form of cognitive supplement, but they would still require considerable coordination of multiple cognitive systems. Indeed, the translation of a naturally ‘one-dimensional’ auditory information flow into the ‘two-dimensional’ visual field offered by the table would make navigating some of the astronomer’s tasks considerably more difficult, even as it freed up his eyes to focus purely on the instrument. The cognitive challenge of trying to replace visual inspection of a purpose-built table with a verbal recitation of its contents recalls Jacob’s assertion that a map provides a visual synopsis of information that would otherwise have to be delivered sequentially. If the map’s contents must be expressed in language, the map will turn into an itinerary, ‘an exploration of space unwinding in time without ever having to be organised into a global image’ (Jacob 2006: 23–4). Since such an itinerary would need to be narrated as a whole rather than consulted ad hoc like a map, it quickly exceeds the ability of its audience to keep track of all the information. An astronomer having the tables of the Syntaxis read out to him in order to save his eyes for the instrument would encounter similar problems of memory capacity, travelling a linear informational path of numbers expressed to three sexagesimal places. Indeed, these problems might even be worse for the astronomer, since he would have to correlate that narrative with one he was producing as his eye moved along its own spatial paths between adjacent stars, their intersections multiplying just as those in Jacob’s map do. It is thus not the mere presence of both elements that makes observational operations, such as those of the astronomer,  a ‘coupled system’, but the particular way the human’s bodily competencies interlock causally with the instrument’s affordances to yield a highly productive cognitive system. The close causal relationship suggests the ‘active externalism’ Clark and Chalmers (1998: 9) contrast with prior models of ‘passive externalism’.

Harmonic Instruments The analytical harmonicist and performing musician will find themselves beset by comparable challenges of coordinating instruments and information. Indeed, the processes of adjusting instruments and making observations are far more prominent in the Harmonics than in the Syntaxis mathematica. The harmonicist will principally use purpose-built instruments such as the monochord or its multiple-stringed analogues rather than performance-oriented instruments such as the lyre or kithara. Specialised harmonic instruments are designed with calibrating mechanisms, such as the measuring-rod (kanonion) or the so-called kollabos, to make sure the instrument is helping to correlate arithmetical harmonic relationships and the evidence of the ears.20 Ptolemy recounts attempts to construct harmonic instruments from a wide range of materials, from hollow bronze dishes to the stringed instruments which 20

On the kollabos, see especially Harm. 2.16. On the different needs served by the harmonic kanones and performance instruments like the aulos, lyre and kithara, see Harm. 2.12.1–20.

52

c ou rtney ro b y

ultimately proved more standardisable.21 Strings present problems of their own; the strings of ancient musical instruments were generally made of gut, the elasticity and thickness of any piece of which might vary considerably from others. Despite this, of all the problematic materials available, stringed instruments were best suited to serve as instruments for demonstrating harmonic ratios. Ptolemy praises the kanōn in particular for its precise tuning, its ability to make visible the resonating lengths of string associated with particular tones, and the fact that it can easily be spot-checked for disruptive inconsistencies in the string.22 The monochord kanōn can be used to conveniently and accurately check the ratios of the concords, associating quantitative values with the audible relationships between the notes of the fourth, fifth, octave and double octave. For other types of harmonic investigations, such as judging Archytas’ claims about the tetrachords, it is desirable (though, as Creese notes contra Ptolemy’s claim, not strictly necessary) to introduce multi-stringed kanones. Still other operations are most easily tested on the helikōn, whose angled strings are bridged analogously to the mesolabe used in mathematics to find multiple mean proportionals.23 The kanōn is an effective tool for harmonic demonstration because the division of its strings by bridges matches visible proportions with audible harmonic relationships. Indeed, it is a ‘material anchor’ that functions in Hutchins’s strong sense, where ‘material structure represents conceptual structure’ (Hutchins 2005: 1572). The kanōn enjoys the ‘cognitive stability’ Hutchins demands of a powerful material anchor, both because it is materially stable enough to ‘maintain its system of constraints while being subjected to mental (or physical) manipulation’ and because the harmonic relationships it reveals are ‘embedded in conventional (culturally shared) well-learned and automatically applied internal mental structure’ (Hutchins 2005: 1573–4). The harmonic relationships Ptolemy describes extend from individual notes to composites of ‘familiar genera’, up to ‘modes’, different divisions of the scale that were tightly bound in Greek culture to an array of emotional and cultural associations.24 Though the monochord kanōn is quite suitable for harmonic investigations, Ptolemy concedes that it is not the right choice for every context, and in fact that it is the worst instrument of all for playing actual music. Though he is not 21

On the advantages of stringed instruments over these alternatives, see Harm. 1.8.1–30. Harm. 1.8.24–30. For more detailed discussion of the monochord kanōn’s structure and harmonics, see Creese 2010: 304–24. 23 On the multi-stringed instruments (the kanones of eight and fifteen strings, and the helikōn), see Creese 2010: 325–9, 334–9. On the potential for using the mesolabe for this application, see Creese 2010: 196–206. On the mesolabe more generally, see Taub 2008. 24 Ptolemy describes these composite structures throughout book 2 of the Harmonics; see in particular section 1 for the familiar genera, sections 7–11 on the tonoi, and sections 14–15 for the numerical definitions of the familiar genera and tonoi. Plato describes the cultural associations of the modes in the Republic (398b6–400e3), and Ptolemy addresses the associations between concords and states of the soul in book 3, sections 5–7. For a recent experimental study of the emotional resonance of the modes, see Straehley and Loebach 2014. 22



physic al sciences 53

principally concerned with the context of musical performance (a rather different animal from harmonic demonstration), Ptolemy’s rejection of the monochord kanōn as a performance instrument provides insight into the challenges of manual coordination that plague the performing musician. The monochord’s principal disadvantage is its single-stringed simplicity, which translates into operational complexity, since the performer has to be tuning with one hand and plucking with the other at all times. The monochord thus ‘is deprived of the best aspects of the hands’ work – I mean things like accompaniment, trills, ascending and descending progressions, prolonged tones and in general the intertwining of notes ­­transposed in order’ (ἀπεστερῆσθαι τῶν ἐν τῇ χειρουργίᾳ καλλίστων – λέγω δὲ οἷον ἐπιψαλμοῦ, συγκρούσεως, ἀναπλοκῆς, καταπλοκῆς, σύρματος καὶ ὅλως τῆς διὰ τῶν ὑπερβατῶν φθόγγων συμπλοκῆς, Harm. 2.12.36–8). The monochord’s precise production of a single tone, which makes it so useful for disciplining auditory data, is extremely limiting compared to the myriad techniques a musician’s skilled hands are capable of when paired with a more accommodating instrument. The musician is gifted with bodily tools that receive harmonic information as well as producing it; while his hands move over the instrument in the complex cascades Ptolemy describes, his ears are receiving feedback from the instrument, ready to relay instructions back to the hands to make any necessary adjustments should the tones drift from their correct intervals. As he plays, the musician is entangled in the same kind of ‘embodied multimodal experience’ Hutchins observes in the navigator’s operations, comparing input from ear and hand and adjusting his manual actions accordingly (Hutchins 2010: 431). This is only possible because of the cooperation between the motor activities of the hands, the sensory activities of the ears, and the internal perceptive faculty that ‘all but screams as it clearly and unwaveringly recognises the fifth’ (μονονοὺ κέκραγεν ἐπιγινώσκουσα σαφῶς καὶ ἀδιστάκτως τήν τε διὰ πέντε συμφωνίαν, Harm. 1.10.45–7). This memorable image is deployed in the context of the harmonic debate between the hyper-­­ theoretical Pythagoreans and the hyper-empirical Aristoxenians, where Ptolemy himself advocates a middle path.25 Ptolemy’s moderating solution requires an ongoing feedback loop between calculated action and informed perception; instrument, hand and ear are all recruited as crucial links in the chain of perception and action that constitutes both musical performance and harmonic demonstration. Ptolemy has indeed already introduced such a cooperative cognitive system earlier in the Harmonics, comparing the singer’s vocal tract to an aulos. The singer is able to adjust his tone appropriately ‘since our hēgemonika, with their native musicality, marvellously and readily find and grasp, like a bridge, the places on the windpipe’ (τῶν ἐν ἡμῖν ἡγεμονικῶν τῇ συμφύτῳ μουσικῇ θαυμασίως ἅμα καὶ προχείρως εὑρισκόντων τε καὶ λαμβανόντων, ὑπαγωγέως τρόπον, τοὺς ἐπὶ τῆς ἀρτηρίας τόπους) where the distance to the resonating air outside is in proportion to the desired harmonic 25

For this lengthy and bitter controversy, and Ptolemy’s mediation between the two sides, an ­excellent guide is Barker 2006b: 14–32.

54

c ou rtney ro b y

interval (Harm. 1.3.85–8). Though the hands are replaced as the bodily instrument by the vocal folds and larynx, singing thus involves the same sort of feedback loop between the body and an explicitly decentralised cognitive faculty that guides the body’s physical adjustments. Ptolemy shows us the work of the astronomer who shifts his gaze and hand position between textual and physical instruments to trace the relative positions of the stars, the harmonicist who manipulates his kanōn to close in on the most precise quantitative characterisation of harmonic relationships, and the lyre-players and singers who modulate their hand and voice to reach the stable points of harmony their perceptive faculty ‘screams’ for. Each is a seasoned explorer of his chosen scientific domain, mapping its territory using techniques that strongly resemble Hutchins’s ‘enacted representations’ (2010: 433), which are dynamic, integrating memory for the immediate past, experience of the present, and anticipation of the future. They are multi-modal, in the sense that they may involve the simultaneous coordination of any or all of the senses and any modes of action. They are saturated with affect. Mind, eye, hand, instrument and text cooperate within a space that blends the cognitive and the material, sensory experience and theoretical knowledge, internal faculties and external tools, integrating the scientist into the world he explores.

Conclusion The central role Ptolemy ascribes to sensorimotor functions in carrying out precise scientific observations, and the necessary refinements observational instruments provide, point to a decidedly distributed view of the scientist’s cognitive work. The cognitive processes he depicts in the Criterion as materially enabled and distributed throughout the whole body appear to extend even beyond the body in the scientific activities he describes in other works. Indeed, so robust are the fusions Ptolemy envisions between brain, body and calculating or observational instruments that it appears plausible to think of the processes he analyses in terms of extended mind theory. The ear and eye team up with the kanōn to reconcile and ratify the auditory and quantitative evidence for harmonic ratios. The expert eye, perhaps aided by a placeholding hand (and certainly by a precisely calibrated placeholding instrument), sweeps through a complex table of celestial locations, plucking out the ones needed to perform a new calculation. Such deep engagements with instruments strongly suggest a model of scientific observation and calculation which is not merely ‘situated in a wider system of scaffolding’, but in which that ‘scaffolding in part compose[s] or constitute[s]’ those mental processes (Rowlands 2009: 54). We may additionally ask whether Ptolemy also suggests ideas about cognition compatible with sensorimotor enactivism. Certainly he is concerned in the Geography and Syntaxis mathematica with questions of how the eye engages actively with instruments. The eye’s engagements range from navigating complex data



physic al sciences 55

tables, to positioning itself over maps so that the visual rays impact the image without deformation, to ‘attaching’ itself to the right point on an instrument for observing the stars. Likewise, in the Harmonics Ptolemy strongly emphasises the skilfully nuanced manipulation of stringed instruments by a harmonicist or musician to produce precisely tuned harmonic ratios or richly varied musical performances. Mastery of the complex sensorimotor feedback systems these instruments entail is vital to their correct operation, and must be obtained through habituation and practice rather than explicit instruction through texts. Ptolemy’s emphasis on exploring the environment through skilled manipulations of body and instrument suggests that his cognitive model might be situated somewhere on the enactivist spectrum. At the same time, in both the Criterion and the Harmonics Ptolemy seems to oppose the idea that these sensorimotor and cognitive functions are in fact contingent on a specific bodily form. We may recall that in the Criterion, he remains agnostic on the subject of whether the psychē’s faculties vary throughout the body because it is composed of different materials in different bodily locations, or because it is affected differently by travelling through bodily structures of different shapes, though its own material remains homogeneous. The suggestion at the end of the Harmonics that the harmonic proportions reflected in our various psychic faculties are themselves mirrored in the periodic motions of celestial bodies requires a certain indifference to the specific contingencies of the human body. This appears incompatible with an extreme enactivist viewpoint, such as Noë’s, which insists that ‘to perceive like us it follows that you must have a body like ours’ (Noë 2004: 25). On the other hand, Rowlands argues that sensorimotor enactivism does not absolutely demand such a strong parallel between bodily structures and cognitive capacities (Rowlands 2009: 57).26 Had Ptolemy addressed this question more explicitly he would perhaps not have insisted on one interpretation or the other. As we have seen, his theories of cognition accommodated multiple viewpoints and fluidly changed over time as new evidence appeared, in keeping with the eclecticism of his philosophical views. Ptolemy’s scientific work explores the full reach of his known cosmos, but never from a viewpoint detached or distant from the world he observes. Rather, we are constantly reminded that this work is the product of observations and interventions with the physical world made not by a bodiless analytical faculty, but by a human equipped with a wholly ensouled body, as well as a host of external instruments. The intimate linkages between each component of the system, from the material psychē to the celestial motions it contemplates, form a complexly distributed cognitive system capable of precise and varied exploration of the world it inhabits.

26

For further consideration of the tension between views of embodied and embedded cognition which posit a strong linkage between the details of bodily form and function and those which admit of a ‘body-neutral’ approach, see Clark 2008a.

56

c ou rtney ro b y

Appendix: Selected Digitised Manuscripts of Ptolemy’s Works with Data Tables Work

Library and shelfmark

URL for digitised version

Bibliothèque nationale de France, Grec 2390

(last accessed 22 February 2018)

Bibliothèque nationale de France, Grec 2391

(last accessed 22 February 2018)

Almagest

Almagest (Gerard of Biblioteca Nacional de Cremona’s Latin España, MSS 10113 translation)

(last accessed 22 February 2018)

Almagest (JudeoBibliothèque nationale Arabic translation) de France, Hébreu 1100

(last accessed 22 February 2018)

Harmonics Vatican Library, Barb. gr. 257

(last accessed 22 February 2018)

Bibliothèque nationale de France, Grec 2451

(last accessed 22 February 2018)

Bibliothèque nationale de France, Grec 2452

(last accessed 22 February 2018)

Vatican Library, Barb. gr. 163

(last accessed 22 February 2018)

Bibliothèque nationale de France, Grec 2423

(last accessed 22 February 2018)

Bibliothèque nationale de France, Grec 2493

(last accessed 22 February 2018)

Bibliothèque nationale de France, Grec 2390

(last accessed 22 February 2018)

Geography

Handy Tables



d i stri buted c ognit ion an d in f o rm at i o n t ech no l o g i es 57

4 Di st r i but e d C o g n i t i o n and the D if f u s ion of In f o r m a t i o n T e ch n o l o g i e s  in the Roma n W orld Andrew M. Riggsby

This chapter has a somewhat different focus from the majority of the essays in this volume. Accepting the truth of the distributedness (in various senses) of cognition, those other essays use that truth to shed new light on a variety of what might be called, in a very broad sense, ancient philosophies of mind, both folk versions and more intellectualised ones. Though beginning from similar assumptions, this chapter does not proceed in that archaeological direction. Rather it argues that modern awareness of distributed cognition is crucial to the causal explanation of historical phenomena that are less theoretical. It is a history that uses the concept, not a history of the concept. (In this respect it is closest to the contributions of Meineck, Roby and, in parts, Short to this volume.) In particular, I suggest that distributed cognition is important to understanding when and how Romans used certain scribal techniques of information organisation. 1

Some Definitions The easiest way to explain the scope and aims of this chapter may simply be to gloss several words and phrases in the title. ‘Information technology’ is of course a potentially very broad field, but I mean it here in a fairly abstract sense. Instead of concrete objects, such as the book or the abacus, I will focus on two conceptual devices that I hope will be exemplary – the table and the multiply-nested list. I mean both of these as terms of art, and I will discuss the specific definitions I have in mind below. By ‘Roman world’, I mean principally the Latin-speaking world, rather than all areas that happened to have been under Roman political domination. Though there seem to be broad similarities between ancient Mediterranean societies in information handling, the distribution of specific technologies tends to be language specific. Conventionally, if somewhat arbitrarily, I also limit myself to the pre-Christian empire. Finally, the chapter suggests that ‘cognitive’ factors play 1

I must thank the several anonymous reviewers of this volume for multiple helpful comments along a number of different lines and Douglas Cairns for the opportunity to participate in this most interesting conversation.

58

a ndrew m. ri g g s b y

a significant role in whether or not these technologies are developed and used in particular contexts. Despite my use of the word ‘conceptual’ above and the formal character of the definitions below, I do not treat the tables and lists as mental abstractions. In fact, a more distributed model of cognition explains their use patterns; they are more successful in environments where there is more ‘scaffolding’, with various material, kinetic and social forms to support them. Even when it is limited to its informational senses, the usage of the word ‘table’ can be quite diverse, potentially referring to almost any two-dimensional display of information. I have no quarrel with any of these usages, but I do want to stipulate that I will be talking about a very specific subset of this broad category. For the purposes of this chapter, I will restrict ‘tables’ to matrices in which individual columns and rows consistently represent identifiable variables. They need not literally be labelled, but it should be possible to provide labels after the fact. So a filled-in crossword puzzle or ‘magic square’ grid is not in this sense a table. The rows and columns do not mean anything. Additionally, for reasons to be discussed at the end of the chapter, tables should be used or designed to be accessed along both axes, if not necessarily in equal measure. So, for instance, a telephone book is not, in this sense, a giant, two-­­ column table (or even a set of them). Rather, it is a list of annotated names.1 Given this second criterion (‘designed to be accessed’), there will be occasional borderline cases, but in general it is not difficult to identify a table in this restricted sense. Moreover, both criteria will also eliminate texts which gather repetitive types of information, in which – accidentally and/or because of the aesthetics of ­­arrangement – certain bits of information tend to line up. So, for instance, the following odd-looking formatting (for which there are surprisingly close Roman parallels in actual use):     $  2  .  05     $  1  .  45     $  2  .  99 Nested lists are more straightforward in one sense, but trickier in another.2 On the one hand, a basic definition is not hard to express. I am thinking of lists that collect other lists, each of whose entries gives rise to another list. For instance the table of contents of a complex law code or certain kinds of maths and science books:

 1

I should point out that a very similar distinction has been independently arrived at, in different contexts and from a variety of different disciplinary perspectives, by Robson 2004: 116; Marchese 2013: 56; Gierdien 2009: 18–19.  2 The topic of lists in general in antiquity, or even just the literary contexts which rely on catalogues, is an enormous one and beyond the scope of this chapter. For a cognitively oriented way into the topic, see Minchin 2001a.



Di stri buted Cognit ion an d In f o rm at i o n T ech no l o g i es 59

  chapter 1      section 1      section 2      section 3   chapter 2      section 1      .  .  . There is a main list (of chapters), and nested into it are lists of the sections. As in this example, nested lists are often numbered (or otherwise indexed) today, and this will also be the case for all the Roman examples here. When I say ‘multiply-nested’, I mean not just that there is nesting, but that the lower-level lists correspond to each other. So, in the example above, chapter 2 is divided into the same numbered sections as chapter 1, not (for instance) lettered paragraphs. As a result, these are close relatives of the table; they could be rearranged into two- or three-dimensional matrices, albeit perhaps with some blank spaces.3 On the other hand, while the definition of a multiply-nested list is clear, there is an evidentiary problem. As far as I know, no complete multiply-nested lists survive from the Roman world. In fact, we do not even have any fragments of the size of the English example just given. Instead, we can detect them only through the way individual references are referred to by combinations of the index numbers. So, for instance, we might extrapolate the structure of the Texas Penal Code – four levels of nesting – from a reference of the form ‘31.03 (e)(4)(F)’. As it will turn out, tables and nested lists operate in very similar ways. I will concentrate my discussion on the former, then turn at the end of the chapter to an account of the relationship between the two forms.

The Basic Problem Neither of these devices has been studied as such by classicists. Occasionally a scholar commenting on one of the very few examples will note that tables are quite rare in that particular context, but scholars do not appear to question the naturalness of the device in general. This assumption has, however, been checked to some extent by modern information scientists. Marchese (2013) approaches ancient tables as part of the broader history of ‘data visualisation’. The cognitive psychologists Novick and Hurley (2001: 202–3) also include a very brief historical sketch of ancient tables in the context of their mostly experimental work on various diagram forms. Neither paper directly addresses the nested list, but Novick and Hurley do also examine the closely related ‘hierarchy’ (their term for what might also be called a tree-diagram). Both these studies look more broadly than I do across time and space (as well as taking a broader view of ‘table’ in the case of  3

In principle, these nested lists could be represented by ‘tree’-diagrams, but those appear not to have been invented until late antiquity at the earliest; see Piggin 2015.

60

Andrew M. R i g g s b y

Novick and Hurley) and find a variety of tables dating from the second millennium bce through the Middle Ages and from a number of different cultures. Both then go on to infer that tables really are universal devices, at least in the literate world. At a minimum, tables do not seem to be the product of particularly unusual cognitive capacities (Novick and Hurley 2001: 202). The reasons for this are not hard to find. The same scholars point out that tables and related devices offer several cognitive advantages to users. One such advantage is the ability to capture and express information not just about individual objects, but also about relationships between them (Novick and Hurley 2001: 168–9; Hurley and Novick 2010: 278; Marchese 2013: 35). It is in this context, for instance, that even the lack of information (that is, blank space in a table) can become significant. Another cognitive advantage is the way that tables encourage novel exploration of data sets (Marchese 2013: 35, 56–8; cf. Wainer 1997: 3). Tables, which can be read in many directions, both isolate data points by creating discontinuous spaces in which to store them and free them from the fixed order of, say, a narrative. However, these advantages are not specific to tables. On the one hand, as Novick and Hurley (2001: 159) had already pointed out, they are (at least) properties of diagrams more generally (see also Giere 2002; Tylén et al. 2015). On the other hand, Goody (1977: 80–1) noted that mere listing already supports at least a similar analytic function, by freeing individual objects from narrative or syntactic context and so making them available for easy re-organisation. The salient feature of Giere’s graphic diagrams, then, is not their graphic/non-linguistic character, but the way both diagrams and tables/lists dispose both linguistic and graphic symbols on the two-dimensional ‘page’. The problem with this very neat account, I will now argue, is that the underlying historical claim is misleading. Tables can indeed be found across the pre-modern world, but their distribution across time and space is sparse and highly irregular. That is, while there are certainly clear examples of tables at many times and places, the more striking fact is their near total absence in between. Even if the capacity to grasp and use tables and related diagrams is universal, actually doing so is a particular, contingent development, much like the invention of written language. In particular, what I want to claim is that (a) tables (and nested lists) are nearly absent in the Latin-speaking world, and (b) when they do appear, it is in highly restricted use-contexts.4 Novick and Hurley give no classical Latin (or, for that matter, Greek) examples. Marchese’s (2013: 38–41) extended examples leap from very well-established cuneiform ones (for which see Robson 2003, 2004) to the chronological tables of Eusebius (early fourth century ce) and his Latin translator Jerome (late fourth century ce). In between, he mentions only (in addition to some clearly non-tabular  4

I treat these historical/philological questions at much greater length in chapter 2 of Riggsby (forthcoming). In the present chapter I attempt to sketch just enough of that argument to establish the facts my cognitive argument seeks to explain. Conversely, Riggsby (forthcoming) touches only briefly on the cognitive issues raised here.



Di stri buted Cognit ion an d In f o rm at i o n T ech no l o g i es 61

chronologies) the inscribed lists of Roman magistrates and military victors known today as the Capitoline Fasti. Even though Marchese does not make much of possible tables, it is worth looking at both of them in a little more detail, since they turn out to be good examples of non-tabular Roman practice. The Capitoline Fasti (Inscriptiones Italiae 13.1.1) included two different lists. One list records, year-by-year, the chief magistrates of the Roman state. For most years this means two ‘consuls’, but at various points the consuls were temporarily superseded by individual ‘dictators’ (a formal, constitutional office) or by larger boards of various sorts. Spread over a number of panels, then, we have two columns of names plus marginal notes every ten years telling the time from the founding of the city. This basic framework is interrupted at numerous points not only by years with different magisterial structures, but also by years in which one consul was replaced, and by the occasional interpolation of other information. One might still describe this as a table, albeit a clumsily constructed one, but the deeper problem is the distinction between the two main columns. There is none. The more senior colleague, that is, whoever was first to get the number of votes required for election, is normally listed first, but this is not always the case (Taylor 1951). (Variations seem to have to do with manipulation of prominence for political effect.) In terms of the formal definition offered at the start of this chapter, this is not a table because the columns are not meaningfully distinguished. Functionally, the display does not encourage data-exploration, nor does it say much about the relationships between items. It is essentially just a list of Roman magistrates, with the spacing between names in ‘normal’ years only for easy reading and aesthetic neatness. Similar but even stronger arguments hold for the list of military victors. These include more diverse data that would certainly lend themselves to a tabular ­presentation – victor, victor’s office, enemy, date – but in fact they are not spaced out so as to create columns. Moreover, the victor’s name begins each entry and the date always ends the line, but to maintain this pattern, the victor’s office is sometimes forced (by available line length) on to a second line of text, jumping over the date. A certain formal symmetry is preserved, but at the cost of unpredictable spatial organisation more generally. This is clearly not a table; it is another list (in this case a list of lists). I have treated these two examples at some length because, as I suggested, they make several important points about Roman practice more generally. First, just because we might be inclined to organise information in a table, it does not mean that the Romans did. Second, we should not be misled by Roman interest in symmetry and straight lines. These may be aesthetic preferences, not functional ones. We might compare here the tendency in many inscribed Roman lists to delay the final letter of shorter names simply to right-justify the column. One cannot simply look at a document to see whether it is a table; one has to read it. Third, Romans frequently resisted the use of blank space to maintain tabular organisation (this, for instance, is why their calendars and many of their accounting documents do not fall into tables). I will not go into other individual cases, but instead simply point to a number

62

Andrew M. R i g g s b y

of other circumstances in which Romans might have used tables, but in fact did not, often substituting simpler lists as in the cases above. So, for instance, consider Roman mathematical ‘tables’, grammatical paradigms, calendars, accounting documents and property registers which overwhelmingly adopt list form. (I will return to the main exceptions below.) While lists and even numbered lists are plentiful in the Roman world, ­multiply-nested ones are not. Despite the fact that modern texts of many classical literary authors segment them by precisely a system of numbered sections within sections, there is no trace of the practice in antiquity. And despite the elaborate subsectioning of modern legal codes alluded to above, Roman statutes appear to have been divided, at most, into a single series of capita (‘chapters’). Similarly, though modern accounts of Roman theatres assert that seats were identified in this fashion (e.g. section A, row B, seat C), the evidence for this appears to rest on modern forgeries. Before I turn to the ‘exceptions’ – that is, to the cases in which Romans did use these devices – I want to offer some explanation of what seems to be the ‘normal’ case. Why did Romans avoid tables, despite their utility and despite the fact, as we will see below, there were at least a few examples that show they were in some sense available? I suspect that for any reader of this volume, understanding either a table or a multiply-nested list is as transparent as the basic act of reading itself. Yet, of course, we know that literacy must be acquired individually. Moreover, as Harris (1989) has argued, it is really that acquisition that needs to be explained, not its absence. This typically requires either compulsory education or very strong individual motivation to learn specific skills (or both). A couple of different lines of research help flesh out the parallel between the general case of literacy and the specific cases of information literacy with respect to these particular devices. On the one hand, Hurley and Novick (2010) have shown that representations like tables, networks and trees operate with a number of ‘construction conventions’, and that if these are violated, the representations are harder for readers to use. Many of these conventions are probably not taught as such – for example, in networks and tables respectively, objects are represented by nodes or by cell contents and relations are represented by connecting lines or by row/column structure. Still, experiments show readers rely on these conventions in interpreting these diagrams. This illustrates some of the content that must be learned to use the devices properly, above and beyond any innate cognitive capacity. On the other, it has proved difficult to teach computers to extract tables from documents in which they are embedded, especially across different contexts, and especially without some knowledge of the underlying content in addition to formal features (Hu et al. 2001; Zanibbi et al. 2004; Deckert et al. 2011). Tables have cognitive benefits, but also up-front cognitive costs. Thus, these two lines of research converge on a particular observation: tables (and presumably related devices) are harder to use than they look. Although these devices are not generally the topic of much direct attention in our culture, they are nearly ubiquitous in daily life, including early educational situations in which there is some opportunity for incidental instruction (Novick and



Di stri buted Cognit ion an d In f o rm at i o n T ech no l o g i es 63

Hurley 2001: 202–4). This suggests that today a ‘virtuous circle’ develops regarding tables. Tables are broadly understood, and so widely used in many contexts. This in turn creates an environment in which individuals are frequently exposed to them in intelligible contexts, thus enabling induction of their standard properties. That is, the authors and readers of tables create each other. In the Roman world, by contrast, we might expect a vicious circle. Few people create tables, so few people learn how to read them. If few can read them, there is no motive to create them. We can see some evidence for this hypothesis in the case of the one grammatical ‘table’ we know of that actually comes in tabular form. The first-­century bce polymath Varro included such a table systematising the various inflected forms of the adjective albus, in his treatise On the Latin Language (10.43–4). In this work, he does not simply give the table, but instructs the reader how to construct it for herself, starting from the example of a game board to illustrate the form of the requisite matrix. Then he warms up with a numerical example that is neither a conventional multiplication nor addition table. That is, he clearly expects that his reader does not know what a table is in general, and that she is not familiar with their particular uses in either grammar or arithmetic. This describes exactly the kind of environment where the absence of tables would be self-reinforcing.

Why the Outliers? Even if my argument up to this point has been generally persuasive, it is still very much incomplete. As I have already acknowledged, Romans do use both tables and nested lists. Varro’s grammatical table, and perhaps one or two others, are highly intellectualised exceptions. Another type of exception comes from more utilitarian, and highly delimited, contexts. The restriction to specific contexts in itself responds, at least in part, to the claims above. It might make sense that virtuous circles could arise locally in particular places, even if there were not enough of those locations to then create a critical mass for the broader society. In this section I will argue that there are other predictable factors that make the uses of tables and nested lists more or less likely in specific contexts. In particular the environments in which they appear all involve multiple types of cognitive ‘scaffolding’. First, take note of the main utilitarian contexts in which we find tables. In theory, Roman land-surveyors creating new colonies measured and distributed land by creating a grid of squares identified by their distance from each of two perpendicular roads.5 In other words, they created a grid with x and y coordinates. We have enough evidence in particular cases to confirm that these theoretical procedures were carried out in practice. We also have a pair of military duty rosters, which cross-index individual soldiers and individual days to keep track of assignments.6  5

For the process and the evidence, see Dilke 1971. P.Gen. Lat.1ver, column V = Fink 1971, no. 9; O.Claud. (= Bingen et al. 1997) no. 308. The former is on papyrus, the latter on a large pottery fragment, both conventional media for ephemeral documents.

 6

64

Andrew M. R i g g s b y

This kind of document was not normally intended for preservation, so these two examples could easily stand in for a much larger number. Nested lists appear almost only in the organisation of large government archives, both in Rome itself and in Roman-controlled Egypt.7 My use of ‘scaffolding’ is, I think, fairly conventional in distributed cognition contexts.8 By this I mean the array of resources in the physical environment (including one’s own body) which are used to assist or even to constitute cognition. I am convinced that the stronger claim of constitution is correct, at least in a significant number of the cases in which it is advanced, and I think it is valid for much of what I describe below. However, for my historical purposes the weaker (assistance) claim suffices, so I will not attempt to distinguish between the two. In the present case, I want to look at three particular kinds of scaffolding: external material objects, physical activity by agents and social context. The objects involved include the writings that make up the record-keeping technologies, but also their physical substrates (for instance, tablets) and modifications made to the entities that are being organised (for instance, surveying marks). The bodily actions in question are all closely tied to the objects just listed – the marking and movement of them. There is a considerable modern literature on the detailed role of gesture in various forms of reasoning and communication.9 In these ancient cases, we simply do not have the specificity of evidence to follow out the same kinds of paths, but we can infer certain general types of activity from the artefacts themselves. Discussion of the social context faces some of the same limitations. We cannot know precisely the circumstances in which these tables were used, but we can say something about the kinds of interactions that must have occurred. In both these cases (action and social context), detailed reconstruction is impossible. I think, however, that we can infer enough to justify suggesting that cognitive mechanisms better known from modern experimental work may be brought into the discussion, even though there are challenges in their application to a distant historical context. As already noted, the material and action aspects of the process overlap significantly, so I will largely treat them together. One of the two ‘normal’ cases of table usage is surveying in the context of colonisation or other distributions of land. The basic parcels of land are identified by their coordinates within an x, y grid system. Of course, here, the grid is not a pure mathematical abstraction, but it maps directly to the earth itself which has been literally inscribed with marks in parallel fashion.10 The lines separating cells in the table of ownership correspond to paths, ditches, marker stones and other literal lines on the ground. Here the reality directly models the information-organising device, and that reality is created not  7

Examples of the various categories include: CIL 16.10 (discharge records), CPL 157 (birth registry), Zollgesetz der Provinz Asia (= Engelmann and Knibbe 1989), line 4 (senatorial decree).  8 I mean to follow in the tradition of Clark 1997 and its successors.  9 e.g. Nathan 2008; Kirsh 2010; Williams 2013. 10 Cf. Clark 2008b: 141 quoting Brooks on the world as its own best model.



Di stri buted Cognit ion an d In f o rm at i o n T ech no l o g i es 65

by a single act of imagination, but by repetition of the specific procedures of the survey process: using the cross-shaped device known as a groma to sight guide-posts along perpendicular axes and measuring equal distances along those axes with chains and rods.11 The other case of tables that may be conventional is that of the duty rosters. Though overall literacy rates in the Roman world were low, the army relied on a variety of documents for its daily operations, as well as other additional types of document that digested these into records of longer periods. Most of these documents, and various kinds of unit rosters, were clearly mere lists. Still, I would suggest that we can see amidst those lists an evolutionary path that gave rise to the tables. If we look at the numerous examples that survive, we can see that even the basic listing often gave multiple bits of information in each entry (that is, in each line) – not just the name of a soldier, but sometimes also his rank, pay-grade and/or place of origin.12 Since the items to be recorded were highly conventional, the information in question naturally fell into near-columns much of the time. Moreover, over time the character of individual documents changed. The lists were given verbal and symbolic annotations, which naturally extend out to the right or left of the main column of names. Some kinds of information, notably duty assignments, were likely to change frequently and so to be an add-on item rather than part of the original draft. Moreover, this information would need to be added to most or all names, so the horizontal extensions would be constrained to follow more or less the lines established by the initial list (rather than wandering up or down the page). Not only, therefore, did the initial lists develop a more genuinely two-dimensional form, but the simple process of filling in the available spaces, both horizontally and vertically, led to a table-like structure.13 Now, this process would not transform the common unit rosters into duty rosters. If we were to force a tabular reading of most of these rosters, they cross-index individual soldiers not with a series of dates, but with a more diverse set of categories (implicitly set at one time or timelessly): rank, duty, origin. Still, the assignment is the one of these most likely to be changed and is found at the outside edge of the core columns. Simple re-use of the same document without erasure would produce something very much like the tables we in fact have. A more direct ancestor of the duty rosters might in fact be found among some rarer documents preserved in the vicinity of one of the full rosters. There we have several lists of names with numerals appended, apparently to show what watch individual soldiers would take (O.Claud. 309–36). These are on small, disposable ceramic fragments, and therefore do not lend themselves to expansion, but if an officer happened to start on a 11

Frontinus, De arte mensoria 12–15 Campbell. Most extant examples are collected in Fink 1971. 13 It is perhaps not irrelevant that Varro’s instructions include both a pre-made substrate (the game board) and action-oriented instructions for construction, rather than simply writing the thing out. Partly this may reflect caution about copyists, but his intuition of how to solve that problem is perhaps significant. 12

66

Andrew M. R i g g s b y

larger piece, it would grow into almost exactly the form of the attested duty rosters. It may not be an accident that the two duty rosters arrange the names vertically, not horizontally, thus seemingly following the form of all the simpler rosters. I stipulated at the beginning of the chapter that accidental alignment of elements did not necessarily constitute a table; this case suggests, however, that repeated coincidence of that sort could still be an important causal step in their development. Public documents arranged by nested lists include records of military discharges, births and senatorial decrees. In none of these cases do we have even a fraction of the original database, but we do have reference numbers on documents copied from the database which could be used to track down the original. These all happen to take triple-index form such as the year/tablet/entry-within-tablet system used in an Egyptian birth register. (I will speak of these indices as numerical, though in fact years are referred to by eponymous magistrates.) These state records are very large documents and they were continuously, or at least regularly, updated. Notably, their numeration is not as abstract as, say, a chapter/section/subsection system. The tablets are treated as physical objects in their own right. The Latin term I have translated as ‘entry’ (pagina) seems properly to mean not a data point, but a field on the larger physical support, hence its English derivative ‘page’, that is, a concrete location in the world. And in the one case we can understand clearly, the cycle of tablets resets at the beginning of each year; the time period itself is abstract, but here the reference is actually to an identifiable set of tables apparently stored together. As in the two tabular cases, the structure is created, even without reflection, by the act of inscribing particular entries into the register as they come in and periodically filing registers away. In all these cases, as I suggested, there is a close connection between the material implementation and the physical actions of the various agents involved in producing and using the records in question. Somewhat separate from this are the features of the social context that go into the same use and production. That social context in all three areas (that is, around the surveys, duty rosters and government records) is quite similar. These are all government documents of various sorts, but they are all essentially internal rather than being ‘public’ in a broader sense. A private individual might hold a discharge certificate or receipt for birth registry or even read an inscribed decree or law that included reference coordinates, but she would never actually be able to lay hands on the original for verification; that task would be in the hands of the same clerks who did the original filing. Similarly, a colonist might know the coordinates of his individual assigned plot (as if they were a proper name), but would not have access to the manuscript registers which relied on such numbers for their internal organisation. (The formae or maps were more publicly displayed, but they seem to have been used more for symbolic purposes than for informational ones.) The duty rosters are rare enough that their usage is less clear, but one of them is clearly written in multiple hands, suggesting that it is meant for the use of some commander, rather than for public posting. The other includes both ancient erasures and frequent use of dots, apparently as some kind of checkmark, both suggesting some kind of active and official use. Moreover, the former is



Di stri buted Cognit ion an d In f o rm at i o n T ech no l o g i es 67

part of a longer papyrus roll and the latter is written on a ceramic fragment; both media strongly suggest the same kind of internal role, not as documents intended for general circulation even within the unit. These three contexts pre-empt the ‘vicious circle’ problem that I raised above. In the cases in which use of these devices is well attested, we can see that they circulated among a relatively small number of individuals who were both producers and consumers of the information form. But we should also note that within those circles, multiple users would have interacted with the same documents. Here I stress that I mean materially the same documents, not merely texts of similar content. I have mentioned the multiple hands on one duty roster. Most of the government archives are large enough operations that more than one clerk would probably be adding records to them, even over fairly short periods of time. The same might be true of the land-holding records, though that would have been less of a practical necessity. This interactivity then means entries are necessarily discontinuous and we have already seen that this is crucial to the function of both tables and lists. Different hands automatically distinguish different bits of information. In some cases, though this is more speculative, interactivity might also direct the user in more specific ways. Much as the form of the annotated list encouraged spreading in a second dimension, the spaces left available by one writer could have helped determine where the next entry went. At any rate, the general picture seems clear. The few cases of consistent use of tables or nested lists correspond to circumstances in which there is dense scaffolding of various sorts to support them. Such an explanation is all the more plausible when we consider some of the cases in which Romans did not use those tables or nested lists. These include most of the cases mentioned above for which Roman and modern practice diverge. So, for instance, the content of would-be arithmetic tables and grammatical paradigms is fixed and finite (and, in fact, fairly small). They do not have the repeated use of the military documents. Nor do they have any natural material analogue, as in the case of centuriation. The calendar is a less intuitive case for users of the modern version. Nearly all our surviving Roman calendars post-date the reforms effected by Julius Caesar.14 At this point, it was possible to create a single prototype ‘perpetual’ calendar to be inscribed on stone. The calendar was a list of the days of each month in order, each accompanied by various other pieces of information such as religious festivals to be celebrated. As presented in stone inscriptions, this is a slightly longer, but still not enormous, and entirely static object. Finally, theatre seating is known to have made some use of listing; in various places in the Roman world, we know entrances, sections, rows and other elements of the whole were numbered. Still, they appear not to have been numbered in nested fashion. The array of seats in even a small theatre is considerably larger than any of the above data sets, but it is again entirely static; the structure of the theatre does not change and so there is no action-scaffolding. There is a material infrastructure that is reminiscent of the centuriation case, but 14

On Roman calendars (Fasti), see Rüpke 2011.

68

Andrew M. R i g g s b y

its relation to a hypothetical nested list would be more complicated than that between surveyed land and a table of holdings. Obviously in the general case, a physical layout cannot correspond directly to a more abstract, textual organisation in the way it can to a graphic like a table. But even beyond that, the nesting creates a particular problem. A given seat can represent itself, but it is also an element of, say, the row and section of which it is part, meaning it would have to play multiple simultaneous roles in any direct projection. The world is not, in such a case, the simplest model of itself. Now, I have made this argument in terms of the objects themselves, but in fact it does all hinge to some extent on the uses to which they were put. One could in principle imagine contexts in which theatre management might want to engage with the structure in a more intensively interactive way – say, sampling seats to check for repair or to give out prizes to ticket-holders. A less publicly focused calendar, one that had more interest in market cycles or even individual projects, might well have been interactive enough to generate the more sophisticated information structures. But there is no evidence of any of this happening. More positively, it is probably relevant to the case of the mathematical and grammatical data that, if they have a ‘normal’ context, it is in primary education, where they would most commonly have been dictated by the teacher to his students. This reproduces a basically linear structure. So certainly there is a considerable degree of historical contingency in all these cases, but in the evidence we have, there is a strong correlation between the presence of multiple forms of scaffolding and the use of tables and nested lists.15

How This Works At the beginning of the chapter I cited experimental research on the question of what tables and related devices are good for. In the section previous to this, I observed that the instances of these devices in the Roman world are largely restricted to contexts in which the table or nested list could be shown to be part of a complex distributed system. In this section, I would like to connect those two lines of argument more closely. How, I ask, do these particular kinds of distributed systems achieve the kinds of results we might expect of them? To do so I will turn to modern research on cognition and diagrams. Ronald Giere (2002: 228) begins with the following observation drawn from work on artificial ‘neural’ networks:16 [W]hat such networks do best is recognize and complete patterns in input provided by the environment. The generalization to human brains is that humans 15

I alluded in my definitional section to ‘borderline cases’ between tables and lists. It may be interesting that these typically come from working accounting documents, but ones that don’t seem to be part of multi-party interactions. 16 He is referring to the seminal work represented in part by Rumelhart et al. 1986.



Di stri buted Cognit ion an d In f o rm at i o n T ech no l o g i es 69 recognize patterns through the activation of prototypes embodied as changes in the states of groups of neurons induced by sensory experience. But if something like this is correct, we face a big problem. How do humans do the kind of apparently linear symbol processing required for such fundamental cognitive activities as using language and doing mathematics?

The answer, he argues, is that we offload linear aspects of processing to various kinds of external representations, diagrams in the first place, but also other forms of writing and language more generally. Kirsh (2010: 446–7) notes as an important particular case of this kind of process the fact that external representations are available for physical rearrangement. This avoids a great deal of expensive calculation and reduces search to a visual, pattern-matching scan of the resulting representation. Moreover, he notes, the physical persistence of physical objects (including marks on a page) means that it is possible to focus on one part of a complex system without degrading the stability of the whole. Your mind eventually ‘loses track’ of details of complex representations; the page does not. And in fact this kind of persistence is important beyond the question of rearrangement (Kirsh 2010: 443–4). Safer long-term storage is of course the classic benefit of writing over human memory, but we should also attend to the value of external representations for their use as short-term storage in active processing. Though it is not part of Giere’s or Kirsh’s arguments, we should note that the features they point to – linearisation, rearrangement, persistence – are very closely tied to the specific benefits of tables that I pointed to earlier. I stressed in my definition of ‘table’ the division between rows and columns. This is valuable precisely because it reduces complex relational comparison to a visual scan. More specifically, the kind of structure that allows for this is not merely linear, but requires a whole array of linear structures to be maintained at once. Both the linearity and the simultaneity are problems from a purely internal cognitive point of view, but they can be easily handled by an appropriate graphic representation. Both duty rosters were either literally rearranged (by erasures) or composed in stages. Thus the compiling officer(s) were ‘trying out’ configurations of assignments by inspecting the roster as written, rather than by either pure imagination or trying to physically move soldiers around. Given the purposes of this volume, I have focused on the distributed processing effects of these written documents, but it is worth at least noting their value simply in projecting stored information through time and space, as in conventional accounts of literacy. We have multiple accounts of the property records of centuriation being consulted long after they were first recorded. In fact, in the best attested case, records were consulted at Rome, even though the land in question was in Campania to the south, and over a century after the fact.17 Similarly, the various schemes of filing by nested lists served to authorise individual documents (such as military discharge certificates) by reference to central registries in places like Rome and Alexandria. 17

Granius Licinianus p. 10F.

70

Andrew M. R i g g s b y

Finally, I would like to come back to the relationship between tables and nested lists. At the beginning of this chapter I linked them in terms of their formal similarity. Any table can be converted, albeit with some loss of information, into a list of lists. Conversely, the nested lists here could all be converted to matrices, typically in three dimensions, if one were allowed to add in appropriate ‘blank’ spaces.18 Equating the actual practice with a purely abstract formulation, however, is a risky move in this kind of context. All it really justifies is a hypothesis of similarity in practice. The extensive parallels between tables and nested lists documented in the body of the chapter, then, are important in part just to justify that initial move. Yet this should not be too surprising, since, in fact, it is not only the table that performs the isolation and linearisation just discussed; the nested list clearly does both as well. It differs from the table in not sharing the latter’s inherently comparative nature, and in fact we see that it is used primarily to link individual bits of data to large collections. Tables are used to manage smaller data sets more intensively.

18

This is therefore a special case; other kinds of trees (say, those describing family structures or the elaborate divisiones of rhetorical theory) are more asymmetric and thus do not lend themselves to this kind of treatment.

Plate 1  Detail from the Pronomos Vase. Attic red-figured volute krater by the Pronomos Painter, c. 420–400 bce (Naples, NM 81673). Credit: akg-images/ Album/Oronoz.

Plate 2  Red-figure chous or oinochoe fragment, c. 430 bce. American School of Classical Studies at Athens: Agora Excavations (P. 32870).

Plate 3  Raphael, School of Athens, Stanza della Segnatura, Vatican Palace. Photo credit: Erich Lessing/ART RESOURCE, New York.

Plate 4  Nicolas Poussin, Landscape with Numa Pompilius and the Nymph Egeria, Musée Condé, Chantilly. Photo credit: Harry Bréjat/ RMN-Grand Palais (Domaine de Chantilly).



ma s k as min d t o o l 71

5 M a s k a s M i nd T ool: A M e t h o do lo g y o f M a teria l Eng a g ement Peter Meineck

I live in the facial expression of the other, as I feel him living in mine. Maurice Merleau-Ponty In many ways the Greek dramatic mask embodied the very idea of ancient theatricality: it was closely associated with the rites of Dionysus; always worn by the performers of tragedy, comedy and satyr drama; highly effective at establishing an instant performative frame; and able to project easily recognisable affective states. In ancient Athens to act without a mask would have been unthinkable. Yet today the Greek dramatic mask and how it functioned in performance are widely misunderstood. In its most reductive form it has come to symbolise drama itself, such as the use of its comic and tragic incarnations as the logo for the actors’ union in the United States. In the West the mask is now widely regarded as a passive instrument of disguise and most often associated with the inactive neutral face of the mime. In this chapter I will explore how the mask operating within the spatial conditions of the ancient theatre acted as a superb focaliser and a highly effective distributor for the way in which spectators perceived and reacted to affective and sensorimotor information. I also attempt to explain some of the thorniest ontological problems of the mask, including the question of its duality as both object and face, and its associated uncanny or ‘magical’ properties. Ancient commentators noted that drama had the power to ‘move the soul’ (psychagōgia) and was a highly emotional and compelling experience.1 I will argue here that this was due in large part to the cognitive qualities of the mask. I also revisit some of my earlier research into mask functionality, which was primarily based on neuroscience (Meineck 2011), and apply a theory based on material engagement to explore how the mask significantly enhanced bodily distributed cognitive mechanisms. To understand the cognitive functionality of the mask we need to ask not what the mask was but how it was – that is, to move away from arguments about what the mask may have represented, and instead approach it as an enactive object in its own right. The methodology I employ to try to achieve this is Material  1

Pl. Minos 321a; Isoc. Evagoras 2.10, 2.49; Arist. Poet. 1450a33; Polybius 2.56.11.

72

pet er mei neck

Engagement Theory (MET), developed by Lambros Malafouris (2013) in the field of cognitive archaeology. Malafouris is concerned with understanding the ‘signs, lines and traces’ that can be found in different periods within the archaeological record and developed MET as an interdisciplinary analytical framework based on several influential existing theories of distributed cognition. In particular, MET places material ‘things’ (artefacts, objects, materials and material signs) in cognitive equilibrium with brains, bodies and environments, across permeable mental boundaries. The theory is subdivided into three main areas: the extended mind, the enactive sign and material agency. In this chapter, I will apply each of these to certain features and properties of the Greek dramatic mask to better understand how it functioned in performance and how it contributed to the total experience of ancient drama. Greek dramatic masks did not remain the same throughout antiquity; as the theatre changed in physical form and cultural function, so did the mask. Thus, the large stone Hellenistic performance spaces with their high stages and vast seating areas were populated by large masks with fixed expressions, gaping mouths and often elaborate headdresses. This form allowed the audience to perceive the actor behind the mask and provided recognisable character typologies. Here, I am concerned with the mask used in the theatre of Aeschylus, Sophocles and Euripides, and it was quite different. This was a whole-face mask no larger than the human head, with realistic features rendered in stiffened linen or wood. It was attached to the performer’s head by a small cap with realistic hair. When worn, the actor’s own face was completely obliterated by the painted eye sockets (the actor looked out of small pupil holes) and a small mouth aperture. There is absolutely no evidence from this period that the mask was equipped with any kind of sound production device as is still commonly believed. The form of the satyr mask was essentially the same as the tragic, with equine ears and slightly exaggerated features.2 The comic mask used by Aristophanes and his peers differs in form from its tragic counterpart in that it was a whole head and not a face mask and had more grotesque bulbous features. However, it operated in much the same way as the tragic mask, but as part of a more frenetic and fast-paced physical performance style. It was worn with a padded costume that also exaggerated the human form for comic effect (Varakis 2010). My model for the form of the tragic and satyr mask is the Pronomos Vase (see Plate 1), which was manufactured in Athens c. 420 bce and depicts the cast of a satyr drama and three actors gathered around Dionysus and Ariadne.3 The vase is named for Pronomos, the famous Theban aulos-player, who sits in pride of place below the god. In particular, the mask of Herakles, held by the actor on the left in Plate 1, is an excellent example of a classical Greek tragic mask.

 2

Wiles 2007 includes good illustrations of much of the available fifth-century evidence for masks. See also Hart 2010.  3 An excellent study on the Pronomos Vase is Taplin and Wyles 2010.



ma s k as min d t o o l 73

Mask as Extended Mind Tool The first plank of MET, extended mind, is a familiar concept of distributed cognition but well worth restating. Malafouris (2013: 67) posits that ‘the content of a mental state is in part determined by elements of the external world, and thus human cognitive skills cannot be studied independently of the external environment’. Extended cognition must therefore include exchanges between people, ­­artefacts, environment and time. Malafouris uses the example of Mycenaean Linear B tablets and goes beyond the deciphering and translation of the writing system to emphasise the human and material interactions. In this way MET helps us learn more about ancient labour, social practices and the communicative pathways, both verbal and textual, that indicate informational exchange in both existing and emergent cultural practices. Malafouris describes (2013: 79) this methodology as ‘a shift from the micro level of semantics to the macro level of practice – Linear B is no longer seen as a disembodied abstract code; now it is seen as a situated ­technology . . . encompassing reciprocal and culturally orchestrated interactions among humans, situations, tool use and space’. Theories of cognitive embodiment have tended to emphasise the role of bodily sensorimotor experiences as part of a feedback loop with the surrounding environment. Therefore space is not a fixed external realm but a cognitive element possessing its own agency. As Kirsh (2013a: 2) has stated, ‘the actual environment we live in must be a superposition of dozens of enactive landscapes, each one with its own set of prediction generating elements and attention drawing features.’ The environment within which the mask operated was the Theatre of Dionysus on the southeast slope of the Acropolis in Athens. It was not yet the imposing stone structure of the late fourth century but a frontal theatron in the form of wooden stands accommodating around 5,000 people looking over a large playing space adjacent to a religious sanctuary. In this open-air environment, the most prominent feature available to the optic array of the spectator was a view of the sky, then the distant Attic hills and a panoramic view of the southern city. Barthes (1985: 79) commented on the phenomenology of an open-air performance, writing that ‘the open-air cannot have the same image repertoire as the dark theater: the latter is one of evasion, the former of participation’. Just as the mask was an enactive element within the total experience of the drama, so was the environment within which it participated. Thus, as any practitioner who has staged a play in the open air quickly learns, the external environment cannot be ignored but must also be a participant in the life of the drama. This is especially relevant when considering the view of the ancient audience as research on the different cognitive effects of personal space has indicated that distal or ‘ambient extrapersonal’ space can enhance feelings of spirituality, deepen contemplation, promote cognitive dissonance, and in extreme cases lead to discombobulation.4 It  4

On the cognitive attributes of theatrical sky space, see Meineck 2012. For research on pilots and space as an embodied concept, see Previc 2006.

74

pet er mei neck

seems clear that early Greek theatre spaces were all situated with this kind of spatial engagement in mind, thus the theatron was more than the ‘seeing-place’ for the play, but a grandstand for a view that expanded the mind. How then did the mask operate in this vast open-air environment? As the available evidence indicates that they were not much larger than the actor’s head and there was no sound reinforcement, what were the cognitive benefits of masking the performers? In a theatre of dance, movement and song it seems counter-intuitive to retain the mask as a performative tool if its only function was as a vestige of an older ritual tradition or a symbol of the god. One obvious advantage of the open air was that the play of shadows generated by the natural light from the southerly facing performance space would have helped make the mask’s features seem somewhat animated,5 but there were also several other important advantages to performing in a mask in this type of space.

Mask as Frame One of the most immediate attributes of the mask is its ability to create an instant theatrical frame, in that the act of wearing it before spectators creates a call to watch a performance. This is simply demonstrated by an exercise I undertake every year in New York’s Washington Square. A participant is selected and told to stand before the monumental arch at the entrance to the pedestrian plaza. Observers watch what happens as tourists and residents walk past – nothing. The same exercise is restaged, except this time the participant wears a mask. Instantly the passers-by stop and gaze and many will wait for several minutes, watching the masked figure and expecting something performative to happen. This exercise demonstrates that the simple act of wearing a mask before an audience instantly transforms its surrounding environment, in this case from pathway to performance space, and activates the dynamic extended mental relationship between performer and spectator. This is not only because of what the mask may represent, but also its uncanny abilities to stimulate our face recognition systems and how it operates as a multimodal enactive sign, attributes I will explore in more depth below. The point I wish to make here is that the act of wearing the mask transforms the environment within which it is worn.

Gaze Direction The spatial organisation of the fifth-century Athenian theatre, with a large audience (5,000 or more) gathered on wooden benches in an open-air space, meant that audibility was an important factor in how the plays were staged. To be heard the actors would have had to face out when speaking, a traditional form of acting still practised today in large theatres and in opera houses, although the by now common use of microphones has brought changes. Up until quite recently, in most  5

See Halliwell 1993.



ma s k as min d t o o l 75

theatres if an actor wanted to be heard they needed to ‘cheat out’, or find creative ways to always be sending their vocal energy into the auditorium and not upstage or into the wings. This was magnified when performing in a Greek mask as the relatively small mouth aperture dictated that to be heard the actor was required to face the audience. Additionally, from the performer’s perspective, the mask offered no peripheral vision, and the performer could only look out through the very small pupil holes. Another common misunderstanding about classical masks is that they had large, dark, open eye holes, but the material evidence found on vase paintings and relief sculpture contradicts this. If we examine the masks of the Pronomos Vase (Plate 1) we can see the distinct sclerae (whites) and pupils – these masks had eyes. This meant that the actor was also visually engaged with the spectators rather than the other performers in the playing area. The effect of the mask on the staging of the plays may be indicated by the surviving texts where arguments (agōnes) between two characters are written as individual speeches following on from each other with no interruptions, and the more rapid single line-by-line dialogue (stichomythia) is quite rigidly structured and often involves certain words being repeated by each speaker. Thus, the mask dictated a particular formality of verbal delivery and a frontality of blocking (stage movements) known today as the ‘split focus’. This term is used to describe a style of vocal delivery where the actors face out towards the audience to speak even when other characters or events are alongside or behind them. In the Athenian classical theatre, to turn away from the theatron while speaking would have rendered words inaudible. We know from other masked performance traditions, such as Noh and Topeng, that to turn fully away from the audience for more than a few seconds was to disengage the masked performer from the action, and in effect be offstage. This direct frontal engagement may have been reflected in the predominantly frontal arrangement of Athenian theatre space in the fifth century, as recent archaeological evidence for the ikria (wooden benches) seems to suggest (Papastamati-von Moock 2014). This comports with the references we find to acting in the Athenian theatre within the comedies of Aristophanes, where characters addressing the audience call it ‘being in front of the theatron’ (Peace 735).6 For the spectator, one striking visual effect of the theatrical en face mask, with its realistic eyes, would have been the feeling that these masks were staring directly at you. The fact that the ancient representations of classical masks all depict eyes with prominent sclerae may also be significant. Humans are unique in that our eyes have prominent sclerae and therefore we use the gaze direction of others to try to gain insight as to their thoughts and intentions. We also find it most disconcerting to  6

Perhaps an example of this kind of frontal performance is shown on the ‘Oedipus’ vase of the Capodarso painter dated to 330 bce (Museo Archeologico Regionale Paolo Orsi, Syracuse 66557). Here an elderly figure usually identified as the herdsman from Sophocles’ Oedipus Tyrannus faces directly to the front with a raised arm and seems to be telling his story. See Taplin 2007: 91–2, and, for an opposing view, Billing 2008. Csapo (2010: 68–9) sees this scene as reflective of messenger speeches in New Comedy, where the actor frequently faced the audience while gesturing to the characters behind him.

76

pet er mei neck

be stared at, or to look into somebody else’s eyes, unless we are on intimate terms with them. This may have been magnified in ancient Greece, where concepts of baskania (the ‘evil eye’) and extramissive vision could have created a scopic regime where to look was equated to touching.7 With this in mind, it is notable that on the Pronomos Vase and many other mask representations from the fifth century the gaze directionality of the mask is emphasised, with other figures in the painting responding with looks of absorption or surprise. With this in mind it is notable that in many representations of the god of the theatre, Dionysus, on vase paintings he is depicted en face – gazing out frontally, something of a rarity in Greek art. Eyes were also painted on wine cups that were used to drink to the god at the symposium, another Dionysian performance event. When the drinker lifted the cup up to the mouth, the face was covered, and the large open eyes painted on the cup faced out at the onlooker, as if they were masked.8 Therefore it may well have been that the eyes of the classical Greek mask were far more than just decorative, but played an important part in creating the attentional and cognitive interface between actor and audience, one that could be highly affective and deeply personal, as if the mask operating within this vast space was actually looking right at you, one spectator among thousands. Being stared at by an en face mask in the frontal conditions demanded by the performance space would have been quite an intense and deeply personal experience. Thus, the ability of the mask to provoke the human predilection to track the gaze direction of others helped foster a kind of mimetic intimacy between spectator and performer, which in turn enhanced the mask’s ability to then act as a projection device for deep affective engagement – a key element in Athenian drama’s reputation for provoking extreme emotional responses.

Mask, Body and Emotions By incorporating face-recognition, eye-tracking, behavioural and brain imaging studies with the ancient material evidence for masks I have shown how the mask could seem to change its emotional expression and be utilised as a powerful mimetic affective device (Meineck 2011). However, if we draw only on findings from these types of brain-based facial recognition studies and apply them to the mask we immediately encounter a Cartesian problem. A study by De Gelder (2009) found that 95 per cent of social and affective neuroscience studies focused on faces, while of the remaining 5 per cent most used scenes and auditory inputs. The limitations of these studies are telling: most people on encountering a disembodied head in the world would be shocked to say the least. Just as we engage with the face as part of the entire body in the world, so we engage with the mask as an integral part of the performer’s total body schēma. Additionally, externalist theories  7  8

See Cairns 2005: 138. On the connections of the frontal gaze with Dionysus, see Csapo 1997: 256–8. On the frontal face in archaic vase painting, see Korshak 1987.



ma s k as min d t o o l 77

of embodied and embedded cognition as well as enactivism all posit that human thoughts are partially constituted by components of the environment outside of brain, bone and skin. In this respect Malafouris and Noë are quite correct to warn of the epistemic power of brain imaging and suggest that we adopt a critical perspective towards neuroscience (Noë 2009: 164–5). Yet within the broad field of neuroscience, new areas have developed, such as social and cultural neuroscience, that accept the brain’s cultural, epigenetic and experiential metaplasticity, as well as different models of distributed cognition. As far back as 1996, neuroscientist Joseph LeDoux, who worked on the emotions from a predominantly computational perspective, noted (LeDoux 1996: 29) that ‘the opportunity for bodily feedback during emotional reactions to influence information processing of the brain and the way we consciously feel is enormous’. Furthermore, Antonio Damasio (1994: 118) has stated, ‘it is not only the separation between mind and brain that is mythical: the separation between mind and body is probably just as fictional. The mind is embodied in the full sense of the term, not just embrained.’9 There is now broad agreement that the body acting within its environment is a fully functioning part of human cognition and the face is an integral part of the entire bodily schēma presented to the observer. Therefore, to study the mask as only a static object detached from the body of the actor, the physical environment within which it was perceived, and the non-verbal communication that occurred in the cognitive space between actor and spectator, is to run the risk of misunderstanding it almost entirely.

Emotional Masks Most of the face-only studies examined by De Gelder were concerned with the emotional qualities of human faces and how they were processed by observers. Many of these studies were built on Ekman’s theory of basic emotions, which posits that there are a minimal number of facially displayed affective states and that these are universally understood within all human cultures (Ekman 1999).10 Ekman has also suggested that there are also ‘discreet’ emotions which are distinctive but do not necessarily require an evolutionary explanation. It could therefore be argued that it is ontology and not phylogeny which is responsible for commonalities in certain emotional experience.11 Ekman has also stated that each emotion is not a single affective or psychological state but rather a family of related states, which he calls a ‘theme’. These themes are phylogenetically influenced and the variations within them a result of social experience. According to Ekman, this permits variability, cultural differences in expression and fluidity of display.12  9

See also Damasio and Carvalho 2013. For discussion of the influence of Ekman’s theories, see Plamper 2015: 147–63. 11 Ekman and Cordaro 2011. 12 For further discussion on Ekman’s theories related to ancient masks, see Meineck 2017: 86–8. 10

78

pet er mei neck

Ekman’s idea that emotions occur as part of themes of related states goes some way in recognising the fluid and enactive qualities of human affective cognition. In this sense emotions are not singular signifiers of emotion that easily fall into neat typographies, but embodied conveyers of a multiplicity of feelings, reactions, responses, physical and mental states. Furthermore, these perceived signifiers of emotion do not occur on the face in isolation from the rest of the body. With this in mind it should be acknowledged that affective brain imaging studies, though informative, are also limited, as the participants’ movements are restricted and their sensorimotor capabilities impaired by the sensory deprivation of the fMRI scanner and the sterile environment of the laboratory. Research on whole-body perception can provide a better understanding of the mask as an embodied object. Several findings stand out (Baylor: 2009; De Gelder 2009; De Gelder et al. 2010): first, that the body is far better at communicating affective states over long distances than the face, which is only effective when in close proximity to the observer. The orchestra of the Theatre of Dionysus had to be able to accommodate the fifty-strong choruses of the dithyramb, and although recent archaeological surveys of the theatre have suggested a capacity for the auditorium of around 5,000 (as opposed to the old assumption of up to 16,000), possibly less (Papastamati-von Moock 2014), this is still much larger than all Broadway or West End theatres. Therefore, the mask becomes a useful tool in communicating emotional states, not only due to its capacities for quickly creating attention and accentuating cognitive projection (which will be discussed below), but also for enhancing the spectator’s awareness of body movement and gestures within the expansive open-air performance space. The use of the mask in this environmental context also shifted attention away from facial identification (gender, age, ethnicity, status, etc.), which is not particularly relevant for the rapid processing of the expressed affective state. Thus, bodies can transfer action and behavioural responses more effectively than faces over longer distances. There is also a marked difference in the social communicative roles of the body and the face. When we focus on the face we are prone to infer an individual’s mental state, but attention to the body focuses on individual or group action, an important consideration when watching choral-based theatre where collective movement formed a large part of the total experience. Bodily displayed expressions can also provide emotional context and some studies have shown that people are often confused as to how to emotionally categorise a disembodied face displaying surprise (De Gelder et al. 2010). When observing the same face in its corporeal context, emotion recognition becomes far higher. Additionally, although the cortical neural networks involved in face and body processing overlap within the fusiform gyrus, the area of the brain most closely associated with facial processing, viewing whole-body expressions utilises a broader network including the cortical and sub-cortical motor areas. Therefore, the evidence suggests that watching the whole body in motion is far more cognitively engaging than only looking at a face (Peelen and Downing 2005). As the mask emphasised the body, so the body’s movements were distributed



ma s k as min d t o o l 79

through theatrical space. The Greek theatre was predominantly a form of dance drama, especially by the choruses, and the kinaesthetic power of the moving chorus to non-verbally communicate emotions should not be underestimated. Chorality was embedded in Athenian culture, and the audience themselves were all expert dancers – for them, experiencing the dance was far more than an aesthetic, purely visual relationship; it was also a somatosensory, empathetic one. The chorus has been a problematic and strange aspect of the Greek theatre for many modern scholars,13 but in antiquity to stage a drama was to be awarded a chorus. There was not even a prize for actors until the 440s, and Eric Csapo has pointed out that, although dramatic choruses are regularly depicted from about 490 bce on, we find no images of actors on vase paintings until the last quarter of the fifth century bce (Csapo 2010: 23). Fifth-century drama was dominated by the chorus, and the chorus was masked. We know little of the choreography of the ancient dramatic chorus and can only imagine the overall effect of twelve to twenty-four highly trained masked dancers moving together in the orchestra, so here, rather than generally explore the many distributed qualities of dance (Jola and Reason 2016), I wish to briefly focus on another important aspect of movement communication in performance – gestures. Hutchins (2006: 376) has described gestures as properties of a system of interaction, distinct from the cognitive properties of the individual who participates in the system. Gestures form bodily enacted distributed systems of interpersonal communication and, just as the mask heightened spectator attention to bodily displayed affective states, it also enhanced the perceptions of gestures.

Gestures In every culture humans gesticulate when they speak, people who have been blind from birth use gestures, though they have never seen them, and both infants and primates use gestures alongside basic vocalisations to communicate.14 Gestures are a universal human communicative system and although every culture has developed its own distinctive language of gesture, all use hands and arms to communicate. In a recent study participants were asked to communicate with each other either by developing a brand-new vocalisation system or with a new gestural system. Overwhelmingly it was the gestural system that was preferred as a basic form of easily communicable semiotics, even over a combination of gestures and vocalisations (Fay et al. 2014). Gestures are also embodied in that they create and extend the cognitive process beyond the skin and out into space. They have been linked to both the facilitation of abstract thought and the engagement of the brain’s visuospatial systems in that gestures help our thoughts move beyond 13

On problems interpreting the chorus, see Goldhill 2007: 45–80; Gagné and Hopman 2013; Billings et al. 2013. 14 On gestures, see McNeill 2015 and Goldin-Meadow 2005. On gestural systems from the ­perspective of distributed cognition, see Wheeler 2013.

80

pet er mei neck

basic cognitive ­­ functions as we ‘grasp’ at thoughts and ‘reach out’ for new ideas. As Novack and Goldin-Meadow (2016: 652) have stated, ‘because gestures are abstracted representations and are not actions tied to particular events and objects, they can play a powerful role in thinking and learning beyond the particular, specifically, in supporting generalisation and transfer of knowledge’. The visuospatial aspect of the gestural process has been shown to activate the pre-motor sensory areas of the brain involved in the contemplation of movement (Cabrera et al. 2017; Clark 2013). In this way, perceived gestures enable another kind of kinaesthetic empathy, that is, the ability to be affected emotionally by watching the movements of another, and help place concepts into the visuospatial realm.15 Gestures assist us to envision other places and another’s emotions, and both physically performed and simulated or imagined gestures influence human reasoning and problem-solving (Alibali et al. 2014: 150). Contextual information from ancient objects and texts can suggest a typology of ancient gestures.16 In addition, the lyrics of Greek drama contain many overt references to gestures, particularly in choral songs. A fine example can be seen in the choral entry of Aeschylus’ Libation Bearers (22–5). Sent from the House to bear libations    Heavy hands, beating hard.    Cheeks marked with crimson gashes, Nails plough furrows, fresh and deep. We have many portrayals of funerary rituals in Greek art which depict the kind of beating of heads and chests described in the choral song. Aeschylus also describes tearing at cheeks until they bleed, something impossible to show on a mask, but that can be powerfully communicated by a clear and easily recognisable gesture. A cross-cultural survey of living masked theatre traditions such as Indian Kathakali theatre, Japanese Noh drama and the European Commedia tradition would show the centrality of gestures and the precision with which they must be performed. In Noh, there is an intricate movement system known as kata that is learned by the apprentice performer. For example, the kata called shiori denotes grief and involves the masked performer slightly lowering the head and drawing up one or both hands over the eyes; one hand can sometimes be used to point off into the distance to locate the cause of the grief in distal space (Ishii 1994). Kata in Japanese Noh and Kabuki theatre can be likened to the schēmata of Greek dance, described by Plutarch as when the movement is ‘held’ for a moment. The idea of schēmata or ‘fixed poses’ can seem static, but Lloyd Llewellyn-Jones has shown how a fluidity of movement between gestures and dance steps might have operated in the ancient Greek theatre (Llewellyn-Jones 2005). We see something of this in 15 16

On kinaesthetic empathy, see Reynolds and Reason 2012. For an excellent survey of Roman gestural systems, which also includes a good deal of material on Greek gestures, see Corbeill 2004.



ma s k as min d t o o l 81

Greek vase paintings where masked performers are depicted with pronounced gestures with great attention to detail on the rendering of hands and feet.17 The way in which the features of the mask were rendered also played an important part in helping to accentuate the perception of gestures. In my earlier paper (Meineck 2011), I noted the ambiguous facial expression of Greek masks that can be seen on the Pronomos Vase (Plate 1). This helped the spectator project imagined affective states on to the mask, which seemed to change its expression. The mask’s affective ambiguity may have also contributed to the distribution of body movements. A 2007 whole-body perception study (Van den Stock et al. 2007) showed participants angry and fearful faces and combined them with both congruent and incongruent angry and fearful body postures. The results showed that the bodily displayed emotions biased the identification of the facial expressions. This might be explained by the influence of other contextual information that has been shown to influence the perception of facially displayed emotions. For example, observers can be primed by emotional words, the observers’ own emotional state, or any particular cultural biases to the subject under observation (Kleinsmith and Bianchi-Berthouze 2013). However, for understanding the power of the mask, one finding of this study stood out: when the participants were asked to rate how fearful or angry the faces were, the emotion displayed by the body posture proved most influential, and this influence was the greatest for the faces that were the most ambiguous (Van den Stock et al. 2007; Van den Stock and De Gelder 2014). Beatrice de Gelder has suggested that ambiguous faces are processed in areas ‘more dominantly dedicated to body perception’ (De Gelder 2009: 3475). Clearly defined gestures and movements extended out into the shared cognitive space between actor and audience were essential to this process. In sum, the mask did not ‘distance’ the viewer, as is sometimes claimed,18 but instead effectively extended the attentional and affective reach of the performer.

The Enactive Mask Malafouris argues that material things are not merely representational but fully enactive within their respective environments. This concept of the enactive sign rejects object passivity and instead places things as part of the active cognitive process. By applying this model to the mask, we can move beyond questions of representation and mask semiotics to further explore the how of the mask. This methodology is useful in asking a fundamental question about the cognitive life of the mask: what makes the enacted mask seem so uncanny and compelling in performance? One ‘uncanny’ attribute of the mask is its seeming ability to resemble the actor who is wearing it after a certain amount of time. This is captured by the painter 17

For example, the depiction of a dancing chorus Maenad on an Attic red-figure bell krater, c. 460– 450 bce, Ferrara, Museo Archeologico Nazionale, inv. T173C. 18 Jones 1962: 234; Calame 1986. On mask ‘distancing’, see Wiles 2007: 287.

82

pet er mei neck

of the Pronomos Vase who depicts the three actors (but not the chorus members) with faces that strongly resemble the features of the masks they are holding. This was noted in the 1930s by the classical scholar Pickard-Cambridge (PickardCambridge et al. 1968: 187) who called it ‘melting’, in that the fixed features of the mask come to resemble the real features of the actor who is wearing it, even though the mask itself does not physically change at all. This strange aspect of the mask may be related to how it forces the viewer to pay more attention to the movements of the actor’s body and, as noted above, how facial recognition is actually closely connected to whole-body perception. It may be that we unconsciously recognise people by their body movements even more than their faces. This was demonstrated by the experiments of Johansson (1973), who filmed people moving with light diodes attached to their joints and then showed films of only those few light dots moving. When shown still photographs of the light sequences observers were unable to recognise anything, but when they watched the diodes in action they could easily identify the movements being performed, and if they knew they person who had been filmed, they were able to accurately identify the subject. It has certainly been my experience working with masks in rehearsal that the mask comes to quickly resemble the actor wearing it. When the same mask is used by another actor in a different production that mask seems to take on the features of the new actor. What we may be experiencing here is what the ancient painters of the melting mask/faces were also visually articulating – that in greatly enhancing the spectator’s attention to the body the mask can seem to meld with the actor, adding yet another level to the complexity of the mask in performance. This uncanny aspect of the mask may have been exploited in antiquity where actors doubled in roles, changing masks to adopt different characters. Were the audience aware at some level of the actor behind the mask because of his movements, and might this have had an impact on how the roles were presented? One tantalising example occurs at the end of Sophocles’ Philoctetes, a play where Odysseus uses the young son of Achilles to trick the embittered Philoctetes to go to Troy to fight for the Greeks. Odysseus has mentioned that he will come in disguise to aid the ruse and the small role of the merchant, who enters about halfway through the play, is usually seen as Odysseus hiding his identity from Philoctetes. At the end, when it seems as if Philoctetes will never acquiesce, Herakles appears to command his philos to go with the Greeks to Troy. The only actor available for this role was the one who also played Odysseus. So, in working within the three-actor rule and the masking conventions of the Athenian theatre, is Sophocles also exploiting the mask’s capacity to enhance movement recognition and tacitly implying that this ‘Herakles’ might actually be another trick of Odysseus? Does the mask’s inherent ambiguity here help project a brilliantly ambiguous end to the play? One way to think about the ambiguous qualities of the mask is to relate it to the way in which people engage with caricatures. The Greek dramatic mask was a schematised image of a human face and therefore processed in a way that is cognitively analogous to the way we perceive facial caricatures. Research into the facial recognition of caricatures has shown that less is often more when it comes to recog-



ma s k as min d t o o l 83

nising a face. For example, Brennan (2007) experimented with a schematic system where the face of a famous person was rendered as an exaggerated line-drawing caricature and set alongside an accurate line drawing of the real famous face. It was found that subjects strongly preferred the caricature face, which tended to be identified far more quickly than the realistic rendering of the same face (Benson and Perrett 1994). As Gombrich (1982: 112) noted, writing on what he termed the borderland between caricature and portraiture, ‘we generally take in the mask before we notice the face’. Gombrich meant that a ‘mask’ in this sense could be a caricature, or even a photograph, where the sitter’s own emotional state of mind and facial expression might communicate something entirely different to a viewer experiencing it within a different contextual frame.

Uncanny Enactivism That masks have the capacity to beguile the viewer was famously demonstrated by Richard Gregory’s ‘hollow mask’ experiment (Gregory 1997).19 Here a plastic joke shop mask is rotated and the inside of the mask, which we know to be concave, magically transforms into the convex face of the mask. However hard we try, we cannot disable this visual disruption, even though we know that we are observing the back of the mask. In effect, our face recognition systems override our conscious mind. Gregory surmised that this was an evolutionary countermeasure, intended to protect us from predators by creating the face of any potential threat from the merest of schematic information. This is why some people think they see the face of Jesus on a tortilla or a burnt piece of toast, and how most can identify age, gender and ethnicity from the scant visual information provided by the so-called ‘Mooney Faces’, low resolution two-tone images of faces that can seem at first like random ink-blots.20 With the addition of movement the mask becomes far more dynamic in its ability to compel and project emotions. These kinds of uncanny qualities provoke an important question: is the mask a representation of a human face or when worn does it become a surrogate face? Or to put it another way: does the mask serve a distinct semiotic purpose communicating meanings that are deciphered by the spectator, or is it part of the physical body of the character inhabited by the actor? MET advances that we do not need to make these types of distinctions; instead, what is most significant about the cognitive life of things is ‘their ability to be, at the same time, mental and physical’ (Malafouris 2013: 80). This is what Malafouris (2013: 79) describes as ‘a shift in the level of description from the micro level of semantics to the macro level of practice’. MET achieves this by making a patent distinction between a linguistic sign and a material sign. A linguistic sign is representational and ephemeral, and operates on the principle of equation, whereas a material sign, or a ‘thing’, is actual and 19

The experiment can be viewed at (last accessed 23 February 2018). 20 For an illustration of ‘Mooney Faces’, see Meineck 2011: 134, fig. 11.

84

pet er mei neck

physical, and ‘can be touched, carried, worn, possessed, exchanged, stored, transfigured or destroyed’ (Malafouris 2013: 95). Material signs also can incorporate a multiplicity of changing qualities; in the case of the mask this could be a variety of equally important affordances including religious, social, political and affective. This is because the mask as a material thing functions foremost as an active cognitive participant and not as part of a static sign system of symbolic equivalency. The mask, then, is an enactive object and acts as an ‘action-constitutive force’. This is borne out by the mask’s ability to immediately create a frame for performance – the simple act of wearing it before an audience instantly transforms its environment and activates the dynamic extended mental relationship between performer and spectator. One of the major stumbling blocks in understanding the ancient mask in performance has been the perception that it acts only as a representational object. For example, Foley (1980) described the mask of Dionysus in the Bacchae as displaying an unchanging ironic smile and acting as the sign of ‘pervasive doubleness’; Vernant envisioned the mask as a symbol of transformation (Vernant and FrontisiDucroux 1988); and others, such as Pickard-Cambridge (1968: 171–4) and Jones (1962: 44), saw it as a theatrical prop or a stage tool, used to aid audibility and help the audience see the performers in a large theatre space.21 Frontisi-Ducroux (1991) did make the suggestion that the mask projected the notion of disconnection, but she envisioned this as a part of a wider sign system of ritual behaviour. More recently, Wiles (2007) took a broad anthropological view of the mask and was primarily interested in its representation in ritual. However, he also approached mask iconography from the perspective of functionality, which brought valuable new insights into what the vase painters and sculptors were communicating about the mask to the viewers of their art works. Wiles also set the ancient mask alongside modern mask performance practices, which helped us to understand masks within the enactive context of the practitioner (Wiles 2007: 4–7).

Mask as Material Anchor If we place the mask within the distributed cognition framework of MET as an enactive sign then we can reconsider it from a different perspective, one that is analogous to the way in which the mask was cognitively processed in antiquity. It then becomes evident that the mask provided a superb material anchor for nuanced cognitive projection. Kirsh (2009: 2310) maintains that this kind of enactive projection provides the foundation for sense-making. Yet, as he points out, projection is not perception – it goes beyond perception to allow us to see what may not be physically present, but what could be. Also, projection is not imagery, which requires no physical property but relies on working memory; rather projection builds a mental scaffolding on an existing material anchor. Kirsh (2009: 2312) calls this the ‘project-create-project cycle’ – we project possibilities on a material 21

On Jones’s impact, see Wiles 2007: 275–7.



ma s k as min d t o o l 85

anchor to attempt to understand it, and these projections augment what we actually see. That mentally projected augmentation is then externalised, which ‘simultaneously changes the stimulus and makes it easier to project even deeper’. Such an externalisation may be a doodle, a map, a tool or, in this case, a mask. Kirsh posits that these externalisations free up working memory and increase mental power. In the same way, the mask increases cognitive absorption and affective engagement and was a superb material anchor for projection. This type of facial cognitive projection, combined with the evolutionary mechanisms revealed by the hollow mask experiment, can also help explain the uncanny qualities of the mask and its seeming ability to merge its features with those of the wearer. In this sense the worn mask, as it is being actively projected on by the spectators, has crossed the ‘uncanny valley’. This refers to a graph constructed by the Japanese roboticist Masahiro Mori (1970). Mori showed how people are attracted to human-like features, especially the face, of things like stuffed animals and simple cartoons. He calculated that this attraction starts at around 50 per cent of human likeness. The attraction level continues to grow until about 95 per  cent, when people might marvel at the likeness, but are still fully aware that they are looking at a non-human form. However, according to Mori, at around 96 per cent everything changes, and the upward curve of his ‘likeability’ graph suddenly plummets as people become disturbed by a likeness that seems ‘too real’, and yet is not real at all. This is the ‘uncanny valley’ – across it, at 100 per cent on the graph, is a healthy human, and within that dip Mori placed ‘nearly real’ repellent human facsimiles, such as corpses and zombies. Crossing the ‘uncanny valley’ has become an obsession within the CGI film and video game industries, yet to date no attempt has been fully successful, despite James Cameron’s announcement that he had crossed the uncanny valley with the special effects of the 2009 movie Avatar (he hadn’t: the ‘Navi’ characters were rendered in blue because realistic skin tones defeated the film makers). But what many seem to have missed, in the quest for a realistic simulated human, is that Mori himself proposed that the simple Japanese Bunraku puppet had already crossed the valley. Mori placed it at around 98 per cent on a scale of moving representations and stated (1970: 34): ‘Although the puppets’ body is not humanlike, we can feel that they are humanlike owing to their movement. And from this evidence I think their familiarity is very high . . . movement is a sign of life.’ What many seem to have missed about Mori’s original theory is that crossing the valley is not a matter of replication – that way lies the depths of repulsion – but of a material anchor, the enactive process of cognitive projection, conceptual integration and surrogacy. Mori’s Bunraku puppet, like the Greek mask, is a schematised version of a human: just enough of a familiar, concrete form with which to blend an unfamiliar phenomenal domain (the uncanny ability of the puppet to seem human or the mask to change its emotions). Therefore, the material anchor provided by the mask acts as a familiar input projection into conceptual ‘blended space’ (Hutchins 2005; Fauconnier and Turner 2008). The process of projection, in the case of the mask and the inference of an affective state (the mask looks happy or sad etc.), produces

86

pet er mei neck

a new emergent structure, such as emotional engagement, empathy or narrative understanding. Thus, the mask seems to have the ability to transform its appearance in real time, though the physical features of the mask have not changed at all.

Mask as Surrogate Face Another facet of cognitive projection is surrogacy, which is particularly relevant to the mask. Clark (2010) has demonstrated how humans are highly likely to construct ‘surrogate situations’ to aid with mental processing, and prefer to embody their cognitive process with pencil and paper, lines drawn in sand, gestures, models and the like. These surrogate material structures act to allow attention on essential features while subduing material specifics. Clark cites the study by DeLoache (1991) that found that young children preferred symbolic representations (‘models’) over realistic representations. This study had two- and three-year-old children watch a scale model of a toy as it was placed under a model chair and hidden from view. The children were then required to find the real toy under the real chair in the next room. The more realistic the model the less well the children performed in discovering the real hidden toy. When the experiment was repeated with the children watching the models from behind a glass wall their success rate increased. The more schematic the representation or surrogate was, the more the children could successfully process abstract relationships between a model and a real target object. Clark points out two key features of surrogate situations: they bring attention to important elements by diminishing specificity, and they relax temporal constraints on reasoning. Surrogacy also allows religious artefacts to transcend basic biological perception and penetrate abstract cognitive domains (Clark 2010: 26). The Greek mask was a quasi-religious object in its own right and although the origins of the dramatic mask are murky, as are most Athenian institutional aetiologies, there is material evidence of masks being used in ritual contexts from sites such as the Sanctuary of Artemis Orthia near Sparta, dating from the seventh century bce.22 In the sixth and fifth centuries bce, the mask became closely associated with the cult of Dionysus: for example, in some vase-painting images a hanging disembodied mask denoted the presence of the god.23 Furthermore, the plays we know from the fifth century were all performed at either the City Dionysia or the Lenaea, important theoric festivals in honour of the god. We may also have a glimpse of what happened to the masks after the performance: a fragment of a lost satyr play by Aeschylus, Theoroi (or Isthmiastae), depicts a chorus of satyrs dedicating images of themselves (quite 22

See Carter 1987. Pausanias (8.15.3) relates how a priest of Demeter in Pheneos in Arcadia wore a mask of the goddess. Masks continued to be used in various rituals throughout the classical period, such as the Eleusinian rites, and were not the sole preserve of the theatre. See Loucas 1989 and Napier 1986: 30–44. 23 A good example is an Attic red-figure stamnos, 470–450 bce, found in Vulci and attributed to the Villa Giulia Painter, depicting a sacrifice to Dionysus. The presence of the god is marked by a bearded mask hung en face on a column. British Museum, London, 1856,0512.14.



ma s k as min d t o o l 87

probably masks) at a temple of Poseidon, suggesting that perhaps dramatic masks were dedicated to the god when their service to the play was over (Aesch. fr. 78a–c Radt). With all this in mind it is possible to perceive that the stable material structure that was the dramatic mask was a powerful conceptual processing projection board – a surrogate face, a character that was not the wearer, a communicator of affective states, and the presence of the god. This was a heady conceptual blend that made watching the mask in performance truly compelling.

Mask as Material Agent The third plank of Material Engagement Theory is material agency. Malafouris rejects the concept that agency is deeply entrenched in anthropomorphism and has challenged this notion with the suggestion that objects possess their own agency. Malafouris cites Knappett (2002): ‘agency comes to be distributed across a network, inheriting in the associations and relationships between entities, rather than in the entities themselves.’24 Knappett’s example is a marionette, which he describes as both ‘iconic’, in that it bears a visual reference to the thing it represents, and ‘indexical’, as it does not fully represent a human and is obviously controlled by the puppeteer, since either the puppeteer himself or at least the puppet’s strings are visible. The puppet’s activation in performance is a result of the convergence of icon and index. Knappett also refers to the uncanny qualities of the puppet, which he describes as ‘sympathetic magic’ (the object visually resembles the referent) and ‘contagious magic’ (in this case the contact with the puppeteer). Yet he also notes that, although the puppet ‘dies’ once the strings are cut, the puppet has not entirely lost its agency. I would suggest that the puppet has less agency when it is viewed as inanimate by the spectator (although some sense of agency is still there), but far more when it has been perceived in action and is then observed at rest. Perhaps this is one reason why the fifth-century dramatic masks were retired and dedicated after the performance. An example of mask material agency in practice can perhaps be discerned from the text of Euripides’ Bacchae. In that play, the audience watch an actor play the young monarch, Pentheus, wearing a mask that must have resembled a young man. At the end of that play (1235–85) we learn how Pentheus, while spying on the women of Thebes and their rituals to Dionysus, was mistaken for a wild beast and torn apart by the Bacchants. Pentheus’ mother, Agave, enters with the disembodied head of her son on the end of her thyrsus – her Bacchic staff. This was more than likely the same mask which had previously been worn and animated by the actor playing her son (Foley 1980: 121). This mask’s continued agency as the head of Pentheus, the symbol of Dionysus, and an object of the theatre would have been very powerful indeed. Like Knappett’s example of the puppet, the Greek mask was iconic in that it resembled a face. It was also indexical because the observer would have been 24

Cited in Malafouris 2013: 127.

88

pet er mei neck

aware that the mask was also a material object worn by an actor. Its ‘magic’ lay in this melding of agency and object where sign and referent are indistinguishable. Knappett also notes that this can be a powerful or potentially dangerous phenomenon and comments on instances of possession as an example.25 Possession is also a factor that has been attributed to acting in a mask: are actors somehow taken over by the mask in performance, and do they forget themselves in a kind of ecstatic state? While we have no textual evidence of this in antiquity, modern mask training books do contain anecdotes of some actors having this kind of experience. While they do not totally lose themselves (they remember their lines and blocking), some actors have described feeling ‘taken over’ when working in a mask. Sears Eldredge, a mask teacher and practitioner, describes this phenomenon (1996: 8): It [the process of transformation] starts with the external object of the mask, a physical image that activates a strong answering reflection in the actor’s imagination. This reflected image seems to be a conflation of embryonic mental, emotional and physical states and actions. Once such a conflation occurs, a character has become manifest in the actor. Fifth-century vase paintings depict this sense of conflated material agency of icon, index and symbol (the mask as a cultural marker of the spirit of Dionysus and later of the theatre itself). A fragment found in the Agora in 1996 depicts a satyr mask in the hand of a chorus member (see Plate 2). The other hand of the actor is rendered with open palm and extended fingers, like a gesture of sudden astonishment. The mask’s eyes gaze up at where the actor’s face, now lost, would have been, and it seems to have come alive – it is both a mask and a living head at the same time. The other figure on the fragment is the aulos-player, who has lowered his pipes and gazes directly at the mask in what appears to be open-mouthed astonishment. Like Knappett’s severed marionette, the mask, though disembodied, is depicted as still having a form of its autonomous magical agency. The Pronomos Vase (Plate 1) offers multiple views of the mask at different stages of use. Some of the chorus members relax and hold their masks carefully; others are contemplating them directly as if preparing for performance. One performer is fully masked and dances almost out of the border painted on the bottom edge of the scene. Mask and actor have merged, and what is depicted is a satyr, not a masked actor playing a satyr – the transformation is complete. David Wiles (2007: 28–33) has astutely pointed out how the two seated chorus figures at the far extremes of the scene both hold their masks and look out to the other side of the vase, which depicts a non-theatrical scene of ‘real’ satyrs dancing. In fact, whenever we see masks depicted in the fifth century the artist seems far more interested in depicting the agency of the mask than in creating any kind of symbolic representation. This idea that the mask has its own uncanny agency seems to have been a trope of vase painters who were primarily interested in the ‘liveness’ of the mask. 25

On masks and trance states in Japanese and Balinese masked drama, see Coldiron 2004.



ma s k as min d t o o l 89

One episode of Rod Serling’s groundbreaking 1960s TV show The Twilight Zone is simply entitled ‘Masks’. In it, a money-grabbing group of family members come to New Orleans on the night of Mardi Gras to be at the bedside of their dying wealthy relative. Before he dies and the will is read, he forces them all to wear masks until midnight. As the clock strikes twelve, he dies and they remove their masks, only to discover that their own faces have become the same as the disfigured masks they wore. This episode compels because many cultures hold that the mask retains a primal power and can possess the wearer. This uncanny aspect of the mask may be a result of its cognitive instability as a prosōpon – an object ‘before the gaze’ of the spectator, and a term that was used to denote both a mask and a face, a linguistic polyseme that may have been exploited in the Bacchae. This instability is a key to what made it so compelling – when the mask was enacted, it was perceived as both a biological face and an object, and it seemed to change its expression, even though the spectators knew the features of the mask were fixed. Perhaps this is why Aristotle in Poetics (1450b20) made a point of stating that the production of theatrical opsis (the visual) with its enthralling, ‘soul-moving’ effects was properly the art of the skeuopoios – the mask maker or the ‘maker of things’.

Mask as Fetish One way to approach this thorny aspect of mask agency is to take Malafouris’s suggestion and apply Appadurai’s theory of methodological fetishism. This removes the focus of agency from the anthropomorphic to the material and is the process of hypothesis formation, which uses metaphoric logic instead of inductive logic. Thus, the familiar is projected over the unfamiliar to seek an explanation. Methodological fetishism maintains that the meanings of things are inscribed in their forms, uses and trajectories. Therefore, ‘from a methodological point of view, it is the thing-in-motion that illuminates its human and social context – this is agency as ‘intention-in-action’ (Appadurai 2011: 5). Malafouris posits that this is not an internal quality but a part of extended cognition. In this respect, agency is not a fixed property of humans but an emergent and mediated product of the world and its materiality. In these terms, the mask is not a piece of inert matter – a prop or costume piece that spectators and performers acted upon – but something active, with which they engaged and interacted. Yet the mask was not only a thing or an object; when enacted in performance it was part of the total character before the gaze of the spectators. In performance, the mask was forgotten as an object; it was instead the face of Clytemnestra, Electra or Orestes. Perhaps this is the reason the dramatic mask is never mentioned in fifth-century literature, not even in the comedies of Aristophanes where he has some characters metatheatrically dress in Euripidean costumes – they never change their masks. The mask was a part of the total embodied character and yet it remained always uncanny – familiar, ambiguous and strange. It is as if it resided within its own particular ‘cognitive gap’, defying categorisation and unsettling normal mental processes. By applying material agency to the Greek mask, we can perceive it as a

90

pet er mei neck

­­ performing enactive object and see how its concrete schematised features allowed for abduction and abstraction. In performance it was enlivened by the movement of the actor, by the play of back and top ambient light shadows, and by both confusing and engaging neural facial-processing perceptual networks. The mask operated as a causative object in its own right, conflating signifier and signified and transcending being a mere symbolic representation (as it has become in the West today). Perhaps most importantly, the mask reflected an ambiguity of agency: not only were its emotional visages rendered as deliberately ambiguous to allow for spectator projection, but it also remained ambiguous as to whether the actor controlled the mask or the mask controlled the actor – hence the mask’s association with possession and trance states.

Conclusion It would not be unreasonable to claim that the development of institutional drama in Athens was nothing less than a major development in the way in which people engaged with narratives as a new form of praxis – that is, narrative articulated by emotional live action. The enactive qualities of the mask played a vital part in the effectiveness of what was a brand-new performance form. This is not to imply that there were no performance forms that enacted emotionality prior to Athenian drama – the tears of Odysseus listening to the songs of the bard Demodocus in book 8 of the Odyssey indicate the affective power of Epic, and the chorus of Delian girls in the Hymn to Apollo tell us how they could assume the voices of characters to articulate the narrative content of their songs (Homeric Hymn to Apollo 161–4). Additionally, Aristotle tells us of the performance of ‘little plots’ at festivals, and how humans naturally enjoy mimicking each other and performing stories. Athenian drama developed out of the rich Hellenic traditions of bardic performance, choral dance and song, and a culture of ritual procession and theoric festivals and possible influences from early dramatic traditions in the Peloponnese and Sicily. But none of these earlier performance forms seem to have afforded the opportunity to experience actors embodying characters and interacting in mimetic situations before a live audience on such a large scale. The mask was certainly not the invention of Athenian classical drama: masks have been identified in Neolithic Greece and remained a part of cult and ritual throughout antiquity. But the dramatic mask was a new development which emerged out of these older traditions, as was the form of the drama itself and the space where it was presented. This was the dramatic revolution propelled by the extended qualities of the iconic, indexical and symbolic mask, distributed by an enactive feedback loop through actor, space and audience. A major practical challenge in communicating enactive and affective narratives to a large audience in an expansive open-air performance space was attention and intimacy. In antiquity, the mask helped negate the distractions of the open air by creating an immediate frame for performance, focusing the spectator’s foveal vision, and by letting its inherent uncanniness grab and hold the audience’s atten-



ma s k as min d t o o l 91

tion. Furthermore, its seemingly magical ability to change its emotional aspect and the way in which it enhanced the affective communication abilities of the body helped facilitate an emotional intimacy between actor and audience in a large performance space, an affective intimacy whereby the individual spectator could relate personally to the situations presented by the enacted narrative and yet still be part of the collective experience of attending a cult festival in honour of a god. Of course, Greek drama was a Gesamtkunstwerk, combining plot, lyrics, music, dance, props and the mask to propel its narratives, but it is hard to see how it all would have worked in its environment without the affective, uncanny and compelling cognitive capabilities of the mask. The application of theories of distributed cognition helps us to see that the mask was fundamental in facilitating this new development in the performance of narrative. The Athenians enshrined this in their own mythical aetiology which recognised a mythical figure named Thespis who stepped out of the chorus to become the first ‘answerer’ (hupokritēs) by in some versions masking his face with white lead and colouring his lips red to assume his new role, and in other sources by donning a linen mask.26 Thus the mythical moment when drama was born was envisioned as the first time an actor performed in a mask or masked his face. As drama proliferated and changed during the Hellenistic period, so did the mask. Performance spaces became much larger, the chorus diminished and actors became famous as professional touring companies emerged.27 The mask became much larger, more stylised, far less ambiguous, and more representational as a clearly identifiable character type. Also, the mouth and eye apertures became larger so that the spectator would always be aware of the actor behind the mask.28 Yet it remained the quintessential symbol of the theatre in the ancient world, well into the Roman period, where representations of masks are found on wall paintings, mosaics, vases and sculptures – an acknowledgement of the centrality of the mask to the whole enterprise of ancient drama. Athenian drama, with its strange otherworldly music, shocking narratives, kinaesthetic collective dances and deep emotionality, was nothing if not an absorbing, dissociative and moving experience across a multiplicity of sensorimotor and distributed cognitive processes. At the centre of the entire phenomenological t­­ heatrical experience was the mask – a material object that enabled enacted, embodied and extended mimesis and helped change forever the way in which people experienced ­­ the performance of narrative.

26

Csapo and Slater 1995: 89–102. For a detailed analysis of the sources for Thespis, see West 1989. On the proliferation of drama in the fourth century, see Csapo et al. 2014. 28 For example, two small Greek terracotta roundels in the form of theatrical masks, one male and one female, from the first century bce, now in the Metropolitan Museum of Art, New York, 1999.316a, b. 27

92

willia m mich ael s h o rt

6 E m b o di e d, E x t e n de d a n d D is trib u ted C og nition i n   R o ma n T e c h nica l Pra ctice William Michael Short

Introduction Moving beyond the information-processing model of cognition characteristic of the classical computational theory of mind in philosophy, cognitive psychology, and ‘good old-fashioned artificial intelligence’ (Haugeland 1985: 112), the ­second-generation cognitive sciences emphasise the central – and even constitutive – role that the body and environment play in human thought.1 Theories of embodied cognition posit that reasoning in abstract as well as in concrete domains is underwritten by conceptual structures and cognitive processes arising from our sensory and motor interactions with the world. These structures include image schemas and their metaphorical projections.2 Some theorists claim that cognition in fact always happens at the interface of brain, body and world.3 Andy Clark and David Chalmers have articulated one of the strongest forms of this ‘transcranialist’ claim in their extended mind hypothesis, according to which the mind is literally composed of both neurological and environmental components.4 Theories of distributed cognition similarly highlight the ways in which human beings tend to offload cognition on to structures in their physical and social surroundings, taking advantage of whatever affordances the environment may provide for structuring and coordinating action or information.5 One aspect of this ‘new science of the mind’ (Rowlands 2010) that has received less attention, however, is how such embodied, extended and distributed cognitive processes cooperate in shaping behaviour in differing cultural contexts. In this chapter, I explore several ways in which members of ancient Roman society exploited the social and, above all, physical environment as a scaffold for  1

See Rowlands 2003, 2010; Gallagher 2005; Boden 2008; Shapiro 2010. See esp. Lakoff and Johnson 1980, 1999; Lakoff 1987; Johnson 1987; Kövecses 2005, 2006.  3 e.g. Varela et al. 1991; Thompson 2007b; Thompson and Stapleton 2009; Stewart et al. 2010.  4 Clark 1997; Clark and Chalmers 1998; Chemero 2009; Menary 2010; Arnau et al. 2014.  5 Through what has been called ‘cognitive niche construction’: see Clark 1997; cf. Hutchins 1995, 2001; Estany and Martínez 2014.  2



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 93

cognition in their technical practices, where ‘technical’ is meant to indicate practice within defined arenas of expert activity. Focusing on memorisation and (to a lesser degree) calculation and on the special challenges these activities presented within a specifically Roman cultural framework, I argue that in Latin speakers’ performance of such activities the environment did not serve merely as the ‘passive background’ against which their cognition unfolded (paraphrasing Malafouris 2013: 72). Rather, it often constituted the functional architecture upon which their cognitive processes were built. I suggest, moreover, that in carrying out some memorisation and calculation tasks, Latin speakers’ distributed cognitive strategies were mediated by the particular embodied metaphorical conceptualisations made available by their language and culture: very specifically, a system of conceptual metaphors, widely detectable in the semantics of Latin, according to which thoughts are conceived as locations and mental activity is conceived in terms of movement in space. This entrenched spatial metaphorics, I contend, provided a sort of motivating context that implied certain ways of utilising the physical environment in support of cognition were more salient and more ‘meaningful’ than others. I conclude by considering some implications of this study both for classical studies and for the cognitive sciences.

Extended and Distributed Cognition in Ancient Memorisation Strategies The human memory faculty consists of several interlocking cognitive subsystems. Generally, cognitive neuroscience recognises four or five different types of memory, distinguished mainly by their temporal persistence: modality-specific sensory memory (retaining information for only several milliseconds); transient short-term memory (including a visuospatial ‘scratchpad’ and ‘phonological loop’ for storing and manipulating visual images or verbal information, lasting about ten to fifteen seconds); working memory (which serves a central executive function for integrating, processing and transferring information between memory stores); intermediate-term memory (storing select information from short-term memory over a period of two to three hours); and long-term memory (where synaptically consolidated memory traces or ‘engrams’ can be stored potentially indefinitely).6 Long-term memory is theorised to consist of several separate, specialised stores for explicit declarative (‘episodic’ and ‘semantic’, ‘know what’) knowledge and implicit procedural (‘know how’) knowledge.  6

Cf. Craik and Lockhart 1972. Atkinson and Schiffrin’s 1968 ‘multi-store’ model of memory consisted only of sensory, short-term and long-term memory. Baddeley and Hitch 1974 later proposed a revision to the model of short-term memory, specifying its constituent subsystems (phonological loop, episodic buffer, visuospatial scratchpad) and incorporating the executive function of working memory. More recently, experimental studies have distinguished between intermediate-term and long-term memory as a function of neural activity in the hippocampus: e.g. Rosenzweig et al. 1993; Parvez et al. 2005; Eichenbaum et al. 2010; see also Cowan 2001 and Kesner and Martinez 2007.

94

willia m mich ael s h o rt

Each of these types of memory exhibits certain definite limitations. Short-term memory is severely constrained in terms of both its capacity and its duration: its capacity is limited by George Miller’s ‘magical number seven, plus or minus two’, although ‘chunking’ – encoding spans or sequences of related information into groupings at a higher level of organisation, especially through salient associative links – can permit the amount of information stored in memory to be quite large.7 Under normal conditions, its duration is between ten and thirty seconds.8 Conscious phonological rehearsing (that is, repeating verbally encoded information to oneself subvocally) may aid retention of information over a somewhat longer period, but the effortful attention required for such rehearsing inhibits most other cognitive as well as many kinds of motoric activity.9 Long-term memory is instead generally capacious and lasting; in fact, it is thought to be unlimited ­­ and some memories may endure for an individual’s lifetime.10 However, voluntary transfer of naturally fading memories from the short- to the long-term store requires special attention and effort. Declarative information stored over the long term may also require periodic ‘maintenance rehearsal’ in working memory and can sometimes be difficult to bring into conscious awareness: searching within, and retrieving information from, the long-term memory store may entail considerable costs in cognitive effort and time – and may nevertheless fail, especially if a subject is fatigued, distracted, inebriated or otherwise impaired.11 The organisation of information in memory – for instance, into conceptual or associative categories, hierarchies, matrices or episodes – as well as explicit encoding (‘I know that I know . . .’) therefore bears crucially on the success of recall.12 At the same time, though more lasting and automatically available, procedural knowledge can be difficult to articulate verbally. This is not to say that human beings everywhere and at all times are equal with respect to memory (or cognitive capabilities more generally). Memory abilities sometimes vary with age: so-called ‘eidetic’ memory, or the ability to recall detailed images after only limited exposure (not to be equated with the ‘photographic memory’ of popular mythology: see Minsky 1986), is usually connected with early childhood, although some adults have an abnormally capacious longterm memory store regarding details of their own lives (‘hyperthymesia’). More generally, the capacity and efficiency of working memory (and thus the ability to consolidate memories long-term) can vary significantly between individuals, at least within certain limits (cf. Conway et al. 2007). Some research indicates that memory  7

Miller 1956; cf. Broadbent 1975 and Schneider and Detweiler 1987. Capacity varies by task and information type, however: cf. Alvarez and Cavanagh 2004.  8 Peterson and Peterson 1959; Cowan 1988; Rubin and Wenzel 1996.  9 See Baddeley 1998: 63–4, 98–101, and Atkinson and Schiffrin 1968 on the effects of phonological rehearsal. 10 Bahrick 1983, 1984; Conway et al. 1992; Martinez 2010. 11 Cermak and Craik 1979; Hasher and Zacks 1979; Baddeley et al. 1984. 12 Tulving and Donaldson 1972; Battig and Bellezza 1979; Hasher and Zacks 1979; Rabinowitz and Mandler 1983.



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 95

differences also exist between the sexes: for example, women have been shown to have superior recall in verbal episodic memory tasks (remembering works, objects, pictures), whereas men have superior recall of symbolic, non-linguistic, visuospatial information.13 Furthermore, a long tradition of scholarship in cross-cultural psychology,14 as well as in anthropology (cf. Lévi-Strauss 1962) and indeed in classical studies, has suggested that the cognitive capacities of members of non-literate and historical societies may have differed greatly from those of modern societies, particularly as regards various aspects of mem­­ory.15 Certainly, as depicted in Plato’s Ion (537a–e), the ability of an ancient Greek rhapsode (a professional performer of epic song) to learn and recall long passages of poetry would seem to far outstrip that of probably any present-day individual, even perhaps one specially trained in mnemotechnics. In Hippias Maior, the sophist Hippias claims to be able to learn fifty names after hearing them only once (285e7, ἅπαξ ἀκούσας πεντήκοντα ὀνόματα ἀπομνημονεύσω), a feat of immediate recall that today we would probably associate only with autism spectrum disorders (cf. Bennetto et al. 1996). If some variation in memorisation strategies can be found at the local, individual level, what are probably innate biological features of the human memory system seem historically to have encouraged certain ‘large-scale’ cultural strategies aimed at the formation, preservation and recollection of memories both individual and collective.16 Maurice Halbwachs and Martin Bommas have stressed how in many traditional societies monumentalisation – the artistic or architectural building-up of the urban landscape that anchors collective rituals to specific places – helps preserve ‘cultural memory’.17 Paul Connerton has noted the importance of material objects and also of the body itself as ‘containers’ for shared memories, the contents of which become recapitulated or, to use his term, ‘silted’ not only in manufactured items like photographs or tape recordings, but also in habitualised bodily performances like handshakes or table etiquette. Meanwhile, Walter Ong, Jack Goody, Ruth Finnegan, Jacques Le Goff and Jan Assmann (among others) have emphasised that the invention of writing has been perhaps the most powerful instrument for exteriorising and thus preserving memory.18 Indeed, writing has 13

Cf. Loftus et al. 1987; Speck et al. 2000; Burton et al. 2004. Luria 1979 stressed the cultural basis of cognition and experimental studies have demonstrated sometimes significant differences in cognition across cultures, in particular in children: see Berry and Dasen 1973; Cole and Scribner 1974; Rogoff and Waddell 1982; Rogoff and Mistry 1985; Mistry and Rogoff 1994. A distinction is often seen between ‘Western’ societies, where memory is object- and self-focused and constructed associatively, and ‘Eastern’ societies, where memory is scene- and group-focused and constructed narratively. For a survey of more recent studies, see Gutchess and Indeck 2009. 15 For the idea in classical studies, see Harwood 1976; Small 1997: esp. 4–10; Minchin 2001a; Calame 2009. 16 Cf. Olson and Cole 2006; Olick 2008. 17 See Bommas 2011. For the ‘spatial writing’ of history in a Roman cultural context, see Kraus 1994; Edwards 1996; Jaeger 1997; Larmour and Spencer 2007; De Sanctis 2014. 18 See, for example, Ong 1982; Finnegan 1988; Goody 1988; Olson 1988; Le Goff 1992; cf. Assmann 2006: 85: ‘Its task was to safeguard and store data that are too complex or haphazard for human 14

96

willia m mich ael s h o rt

come to ­­pervade modern Western conceptions of memory to such a degree that, as Maurizio Bettini (2011a: 3) remarks, memory is now imagined primarily in terms of written or electronic ‘archives’. Already in archaic Roman culture writing was so linked to the preservation of memory that it provided the dominant metaphor for this faculty, as implied by, for example, the derivation of Latin oblivio, ‘forgetting, forgetfulness’, from ob-lew-, that is, ‘smooth (out)’ (as in erasing the writing on a wax tablet or inscribed surface),19 and, perhaps more pointedly, usage of oblitterare (< ob- ‘against, over’ + littera ‘letter, writing’) in the sense of ‘cause to be forgotten’ (as in Liv. 21.29.7, ‘nondum oblitterata memoria superioris belli’, ‘not yet forgotten memory of the previous war’).20 Yet even in their most textual phases, the societies of ancient Greece and Rome were still predominantly oral societies, as Rosalind Thomas (1992) and Thomas Habinek (2009) remind us. Storytelling therefore represented a crucial vector for the preservation of shared memories in antiquity, especially in ritual contexts (cf. esp. Chaniotis 2011b). Greek society’s oral tradition certainly seems to have functioned as, in Eric Havelock’s definition (1963: 27), ‘a sort of encyclopaedia of ethics, politics, history, and technology’: according to Plato’s Ion (537–41), it served as an enduring source of practical knowledge about fishing, medicine, prophecy and horsemanship. Many scholars have shown that certain linguistic features of the Homeric epics in fact seem designed expressly for ease of memorisation.21 Alessandro Vatri (2015) argues that certain characteristics (metre, conciseness, repetition and recapitulation, topicality and thematic organisation) even of Greek literary texts normally not considered part of an oral tradition were explicitly aimed at facilitating memorability. Similarly, Roman culture’s mythistorical tradition offered the behaviour of figures such as Horatius Cocles and Titus Manlius as models for imitation, and the stories that make up this tradition exhibit narrative features such as repetition and symbolic condensation that suggest they were part of an oral tradition in which ‘texts’ would have been memorised for preservation, if not necessarily for performance (cf. Bremmer and Horsfall 1987). But the safeguarding of culturally valorised beliefs and practices was not the only context in which memory would have been a task-critical asset. Many kinds of undertakings required memorisation, and several extended and distributed cognitive strategies were developed in the ancient world to enhance memory. For instance, long-distance trade required the careful inventorying of goods and recording of the prices at which transactions occurred. In archaic Greece, a functionary known as a mnēmōn served this requirement (Gernet 1968; Vernant 1965). memory.’ Writing has produced a number of different strategies for preserving memory: cf. Bolzoni 2001. More generally on the cognitive implications of literacy, see Olson 1996. 19 Cf. also memoriam eradere, delere, oblitterare in the sense of ‘forget’. The conceptual contiguity of memory and writing may also be captured in expressions such as mandare memoriae and mandare litteris. English tends to metaphorise the memory in terms of a ‘container’: see esp. Armstrong 2009. 20 On this metaphor see Bettini 2011b. 21 Esp. Minchin 2001a, 2007; Watson 2001; Mackay 2008. Rubin (1995) argues that features such as formulaic epithets are in fact fully consistent with Miller’s ‘magical number seven’ rule.



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 97

According to a scholiast of Homer’s Odyssey, the mnēmōn was the member of a ship’s crew who ‘memorises how much each thing is worth’ and ‘keeps the cargo in memory on account of not knowing letters’.22 This term was sometimes also used of civic magistrates charged with learning a city’s religious calendar of feasts and sacrifices, or facts related to important juridical matters (see Carawan 2008).23 In these settings, the mnēmōn functioned just as a written text or computerised databank would in later societies: namely, as a passive repository of knowledge, indeed, acting as a human ‘account ledger’ (livre de comptes) (to use the image offered by Gernet 1968: 285) or ‘living archive’ (to use that of Thomas 1992: 19) to be consulted or queried by others. At the same time, Greek mythographers record that mnēmones were also assigned to some heroes of the epic tradition to serve as ‘rememberers’ whose task it was to remind the hero to – or, just as usually, not to – perform a certain action.24 For example, Eresios is reported to have written that Protesilaus was accompanied by a mnēmōn whose job was to remind him not to be the first to set foot on the shores of Troy on account of an omen that ‘he would be killed if he jumped before’.25 Similarly, Antipater of Acanthus is said to have mentioned a mnēmōn whose task it was to remind Hector ‘not to kill anyone dear to Achilles’.26 Even if these mnēmones are an invention of the mythographic imagination – as Cameron (2004: 138) asserts – their stories reveal that Greek society envisioned the human ‘rememberer’ as potentially serving a very different function to a mere ‘archive’: that is, to actively and directly increase the capacity of another individual’s memory. In Roman society, the notion that one individual’s memory could be amplified or enlarged by another’s appears deep-rooted as well. In aristocratic practice, for instance, slaves known as nomenclatores served a memory-enhancing function very much like that of the mnēmones of Greek myth (even if the consequences for memory failure were decidedly less severe) (cf. Plutarch, Quaestiones Graecae 28). Seneca the Younger reports (Epistulae 19.11) that nomenclatores were employed to memorise the names of their masters’ clients and to remind them proactively of the correct forms of address to use in the course of the daily morning greeting.27 Cicero notes derisively (Pro Murena 77) that some candidates for official magistracies would employ nomenclatores to accompany them as they canvassed for votes, thus permitting them to affect some degree of familiarity with members of the electorate by supplying names and titles in an opportune fashion: ‘Why is it Schol. ad Hom. Od. 8.163, ὁ μεμνημένος πόσου ἐστὶν ἕκαστον ἄξιον . . . τὰ ἐγκείμενα κατέχειν διὰ τὸ ἀπείρως ἔχειν γραμμάτων. 23 See also Gernet 1968: 286–7; cf. Ar. Nub. 615–26; Plut. Quaest. Graec. 4. 24 See esp. Bettini 2011a: 33–5. 25 Eustath. in Hom. Od. 11.521, Ἐρέσιος δὲ ἱστορεῖ καὶ Πρωτεσιλάῳ δοθῆναι μνήμονα Δάρδανον Θετταλὸν, δοθέντος χρησμοῦ Φυλάκῳ τῷ πατρὶ, ἀναιρεθῆναι εἰ προπηδήσει. 26 Eustath. in Hom. Od. 11.521, Ἀντίπατρος δὲ ὁ Ἀκάνθιος φησὶ καὶ τῷ Ἕκτορι Δάρητα Φρύγα δοθῆναι μνήμονα, μὴ ἀνελεῖν φίλον τοῦ Ἀχιλλέος. 27 Nomenclator was also apparently used of a slave employed to remember the names of a master’s other slaves, when the size of household slave retinues had grown unmanageably large: cf. Plin. HN 33.26. 22

98

willia m mich ael s h o rt

that you have a nomenclator? In this you are indeed tricking and deceiving people. For if it is honourable that your fellow-citizens be addressed by name by you, it is shameful for them to be better known to your servant than to yourself.’ Latin authors indicate that a similar role was fulfilled by the monitor in oratorical practice, as well as perhaps in the theatre: to stand near to a speaker and to remind him what he should do or say at a certain moment, in the case that his own memory should somehow fail.28 Judging by this evidence, the Roman nomenclator or monitor appears to have played very much the same role that the notebook does for Otto in Clark and Chalmers’ famous thought experiment.29 This figure is called upon not only to store and to produce information on demand, but also to do so in such a way that this information appears effectively to belong to his master (or the orator or actor). Though external to the subject, the ‘rememberer’, like Otto’s notebook, is treated as a simple extension of biological memory (cf. Michaelian 2012). An exaggerated case of this kind of social extension of memory may be that of Calvisius Sabinus. As described by Seneca the Younger (Ep. 27.5), Sabinus had a terrible memory but was surpassingly wealthy, so he trained a group of slaves ‘unum qui Homerum teneret, alterum qui Hesiodum; novem praeterea lyricis singulos adsignavit’; ‘one to memorise Homer, another Hesiod; and then assigned each of the others one of the nine lyric poets’ (for this passage, see also Konstan, this volume). Sabinus kept these slaves ready at dinner parties to prompt him with verses for recitation (‘habebat ad pedes hos, a quibus subinde cum peteret versus quos referret’; ‘He kept these slaves at the foot of his couch, from whom he could request verses to cite’). The role that Sabinus contemplated for these living memory aids thus goes far beyond the ‘convenience’ or ‘shortcut’ (compendiaria) that Seneca suggests it is. Because ‘he believed that he himself knew everything that anyone in his household knew’ (‘ille tamen in ea opinione erat ut putaret se scire quod quisquam in domo sua sciret’), for Sabinus his slaves were not mere cue cards or memory crutches; he trusted them to provide him with accurate information in appropriate circumstances and at appropriate times, and they presumably did so without instigation on his part. In this sense, their memories collectively just are his memories. In historical settings where writing was not widespread, and where other ­memory-aiding technologies – teleprompters, personal digital assistants, iPhones – were still unknown, ‘rememberers’ therefore afforded an adaptive solution to the challenges presented by human memory. As a more or less constantly and immediately accessible, as well as automatically endorsed, source of information, the mnēmōn, nomenclator or monitor would have helped mitigate any potential deficits of learning and recall an individual might experience. Yet this solution to the problem of memorisation could only ever be a partial one, as it failed to address the fundamental problem of human memory. If the living ‘rememberer’ is again similar to Otto’s notebook in constituting an implicitly trusted source of informa28 29

Cic. Div. Caec. 52, De or. 2.24.99; Quint. Inst. 6.4.8–9; Fest. De verb. sign. p. 122 Lindsay. Clark and Chalmers 1998; cf. also Sutton et al. 2010.



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 99

tion, it is dissimilar in that it replicates the very fallibility of its ‘user’, since there is no guarantee that the mnēmōn’s (or nomenclator’s or monitor’s) memory will not also succumb to the inherent weaknesses of this faculty. This deficiency of a strategy that distributes the memorisation task across ‘devices’ of the same kind (i.e. socially, to another individual or group of individuals) is emblematised by so many mythic mnēmones who themselves – ironically, and usually tragically for their heroic charges – suffer bouts of forgetfulness (see Tümpel 1894: 3075) or by the seemingly almost proverbial ‘old slave nomenclator’ who makes up names to give his master when he cannot remember them (Sen. Ep. 27.5). This is probably why mnēmones and nomenclatores were employed mostly in situations where discrete, discontinuous information seems to have been concerned (a register of goods, stage directions, warnings, names) and why a different strategy altogether was developed in ancient rhetoric for learning and recalling lengthy, continuous stretches of discourse. To assist the orator in memorising an oration (or any sequence of utterances or ordered series of items), ancient mnemotechnics developed a sophisticated procedure through which symbolic associations were created between a given structure of physico-spatial reality and the information to be recollected, which could later be recalled during performance through a sort of virtual scanning of mental space: the so-called ‘house’ or ‘palace of memory’. Though usually treated uniformly by modern (and some ancient) handbooks, the Greek and Roman practices actually appear to have differed considerably. In the (less well-attested) Greek system, normally attributed to Simonides of Ceos, the orator seems to have been encouraged simply to visualise the text of the oration in the mind’s eye. In Cicero’s account of the Simonidean method, memorisation thus proceeds by imagining a series of ‘places’ and ‘shaping’ in the mind ‘images’ of the information to be recollected, which are then ‘set in order’ in the imagined places (De oratore 2.354): itaque eis, qui hanc partem ingeni exercerent, locos esse capiendos et ea, quae memoria tenere vellent effingenda animo atque in eis locis conlocanda; sic fore, ut ordinem rerum locorum ordo conservaret, res autem ipsas rerum effigies notaret atque ut locis pro cera, simulacris pro litteris uteremur. And so for those who would train this part of the mind, places must be imagined and those things which they want to hold in memory must be reproduced in the mind and set in order in those places: thus it comes about that the order of the places preserves the order of the things, but the images of the things represent the things themselves; and that we use places instead of a wax tablet, images instead of letters. Some features of Cicero’s description reveal the generically topological character of the visualisation process in the Greek system. First, the imagistic character of the ‘places’ constituting the structural element of the visualisation is left unspecified: use of masculine loci (‘single places’) rather than neuter loca (‘a physical region’) to gloss, presumably, Greek τόποι implies that these were to be understood in quite

100

willia m mich ael s h o rt

abstract terms. Likewise, effingere, a term drawn from the technical vocabulary of moulding in the plastic arts, suggests that the ‘images’ (effigies) were not understood as rich visual images or scenes, because when writers of Latin want to characterise something as being worked out in the mind in a highly detailed and intricate fashion they tend instead to employ metaphors drawn from the language of painting (Short 2012: 115–16). Second, the spatial dimension of the visualisation amounts to little more than a serial ordering of words as they are pictured in the mind. This is made explicit by Cicero’s affirmation that ‘we use places instead of a wax tablet, images instead of letters’ – where a direct correspondence is set up between the ‘mental scratchpad’ and the writing space of a wax tablet. It is also suggested implicitly by his use of conlocare, a term conventionally used of arranging words in a sentence, to refer to the process of ‘setting in order’ the images in their respective places.30 Furthermore, the mental images of the ‘things’ to be remembered are here simply iconic: a visually perceived text is reproduced in mental space in a correspondingly visual form. If, as may be gathered from Cicero’s description, the Simonidean technique for memorising an oration proceeds through simple textual visualisation and is ‘spatial’ to the extent that the words of a written text have a certain topology (i.e. their own textual sequencing), the mnemonic technique as developed in Roman rhetoric appears instead to have made thoroughgoing use of the physico-spatial environment as a device for structuring memory and assisting recall. The general procedure can be reconstructed from the Rhetorica ad Herennium (3.16.29–19.32) and Quintilian’s Institutio oratoria (11.2.19–21): to memorise a speech, the orator was first to ‘construct’ (constituere) or ‘furnish’ (comparare) the representation of some place – the Rhetorica ad Herennium advises, for instance, the image of ‘a house, an intercolumniation, a corner, an arch’ – in mental space (cogitatio). The orator could ‘build and design’ (fabricari et architectari) this mental representation through a purely imaginative process, but the Latin texts recommend modelling it after some real geographical location. Not just any place would do, however; the features of the location had to be taken into account. As Quintilian writes (Inst. 11.2.18): loca discunt quam maxime spatiosa, multa uarietate signata, domum forte magnam et in multos diductam recessus. in ea quidquid notabile est animo diligenter adfigunt, ut sine cunctatione ac mora partis eius omnis cogitatio possit percurrere. et primus hic labor est, non haerere in occursu: plus enim quam firma debet esse memoria quae aliam memoriam adiuuet. They learn places of the greatest possible spatial extent, distinguished by considerable variety, such as a large house divided into many apartments. Whatever is remarkable in it is carefully fixed in the mind, so that recollection may traverse every part of it without hesitation or delay. Indeed, the first challenge is not to 30

Conlocare verba: cf. Cic. De or. 3.171, 3.172; Quint. Inst. 9.4.58, 10.1.4; Fronto, Ep. 4.3.1; Gell. NA 11.13.2.



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 101 get stuck in your progress, for a memory that is supposed to help another ought to be more than certain.

The Rhetorica ad Herennium (3.32) similarly advises that the orator choose a setting that provides ‘the most appropriate arrangement of suitable places’ (‘idoneorum locorum commodissimam distinctionem’), as these ‘places’ will afford a structural anchoring for the memorised text. Once an appropriate location has been selected (taking into account the organisation of the oration), the orator is instructed to walk repeatedly through this location, to carefully ‘learn’ (discere, commeditari) the contours of the physical terrain and at the same time to ‘arrange images’ (conlocare imagines) or ‘attach labels’ or ‘marks’ (mandare notas, signa) to the various features of its mental counterpart, corresponding ­­ to different parts of the oration. Importantly, unlike in the Greek technique, the ‘images’, ‘labels’ or ‘marks’ – that is, schematic or richly elaborated pictures or words – are not of themselves the information the orator wishes to memorise: rather, they are signs intended to evoke previously learned material. That is, each image arranged in the mentally represented ‘place’ is a symbol, an indexical sign that points to something else already committed to memory by the orator. In this way, by later imagining himself walking through this location in the course of delivering the oration, the orator would be able to recall the text of the oration when these signs ‘return’ (reddere) the memory of that text in a sequence determined by their topographical arrangement. Differently from Greek practice, then, where the ‘images’ arranged in ‘places’ apparently correspond to the words of the oration following their linear textual order, in the Roman system these images are symbols re-presenting (in the sense of providing mental access to) other parts of the memory store, and the places constitute a relational structure defining a global address map. In this light, the Roman technique of loci is not so much a method for memorising the text of an oration (which would otherwise need to be committed to memory) as for learning its overall organisation. As a strategy for memorising (the organisational structure of) long sequences of text, the Roman technique of loci thus stands out in construing the generically topological ‘places’ of the Greek system in explicitly architectural and topographical terms.31 But the cognitive process that characterises this technique goes beyond ‘literal interpretation of the term “places”’, as Jocelyn Penny Small (1997: 100) explains it. To begin with, the mental representation of space the orator relies upon during recall is not a literal location. Being modelled upon (the orator’s perception of) physical reality, this representation derives some of its visual and spatial characteristics from a ‘literal’ location, but is itself a figured imagistic space. Furthermore, when the orator goes about ‘arranging images’ within or ‘attributing signs’ to this mental image during the process of memorisation, he sets up a complex semiotic relationship between the location as it exists in his mind and the location as it exists in the world. If, on the one hand, the logical structure of the oration determines, 31

Cf. Carruthers 1990: 16–23; Small 1997: 96–116; Draaisma 2000: 24–30.

102

willia m mich ael s h o rt

in part, the character of this representation (by suggesting what is or is not an appropriate space to be modelled), on the other hand the topographical features of the real location determine the structure of the orator’s memories (by anchoring them to a determined spatial arrangement). Tightly integrating spatial setting and mental representation, the orator brings about memorisation by extending his internal visualisation out into the world, so that his movement in and interaction with the external environment have organising effects on his subjective knowledge. Beyond merely aiding memorisation, this movement and interaction actually constitutes the process of memory formation, yielding exactly the kind of ‘elaborately encoded’ memories needed for long-term retention (Schacter 1996: 44–71). It is worth emphasising exactly what is distributed about this memorisation strategy, however, since it may appear to involve only the orator’s mental image of a place and thus to be compatible with an entirely internalist view of cognition. According to the Rhetorica ad Herennium, the orator does in fact have the possibility of utilising some purely imagined spatial configuration as the basis for his visualisation and of placing his indexical imagines within this space through a purely mental imagistic process. In this case, because the orator would be relying on exclusively neural resources, it would probably be fair to object to any characterisation of this technique as a distributed cognitive process. For precisely this reason, the Aristotelian method of memorisation as described by Small (1997: 87–94) – in which, it appears, stretches of mentally visualised text were arranged sequentially in equally mentally visualised (but sketchily pictured) ‘containers’ or ‘bins’ – ­­probably cannot be considered ‘distributed’. However, this is not the favoured Roman practice. As the Roman rhetorical writers make clear, in their native practice the structural arrangement of an oration is to be memorised, instead, through a process that combines the elaboration of a mental image with the physical activity of moving through space: not only is the mental image to be modelled directly from spatial reality, but it is also the orator’s movement through this space (in the course of which he ‘assigns’ referential symbols to his mental image that correspond to features of the spatial environment) that actually serves as the mechanism of memory consolidation.

Externalising Calculation on to the Social and Physical Environment Arithmetical calculation represented another cognitively demanding activity benefiting from distribution to the external environment in the Roman cultural context.32 As Roman numerals constitute a cumulative (additive) notational system rather than a positional value system (as do Arabic numerals with zero ­­provided), the challenges this presented to members of Roman society in ­­performing ­­calculations

32

See Chrisomalis 2009 for a cross-cultural survey of artefact-use in calculation; cf. also Miller and Parades 1996.



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 103

were not insignificant.33 As Schlimm and Neth (2008) argue, the basic difficulty with Roman numerals is the additional number of processing steps required for resolving symbols into numbers, which can make even simple addition tasks cumbersome and inefficient. Consider, for instance, the problem of adding 14 to 27. Using Arabic numerals, the solution is trivial, as it involves merely summing the numerals in each place (any inefficiency will rest in the necessity of carrying the remainder of the sum of 4 and 7 to the tens-place). Using Roman notation, by contrast, the solution requires considerably more processing steps, since the numeral symbols would need to be converted (mentally) into number concepts and this process is not at all straightforward. In the complex symbol XXVII, the units-value is represented by three characters which require an internal summation (5 + 1 + 1); the tens-value is represented by a digraph which also must be summed (10 + 10); in XIV, on the other hand, the units-value is represented by two characters, whose resolution requires subtraction (5 − 1) and the tens-value is represented by a single symbol. Then the two numbers must be added, and the result encoded in the correct notation. It is theoretically possible that some arithmetical problems could be solved more quickly through hierarchical redistribution and concatenation of symbols (thus eliminating the need to convert to numbers, as proposed by Detlefsen et al. 1975), but this procedure would still have required several steps of decomposition and recomposition and there is no evidence that it was ever used.34 To obviate the challenges that a cumulative notational system presented to mental calculation, several externalisation strategies were used in Roman society. For instance, a system was developed for representing numbers with the fingers (and perhaps additional hand gestures). Augustine reports that the left hand was used for indicating the numbers 1–99 and the right hand for indicating the value of the hundreds-place (Sermones 175.1, ‘nonaginta enim et novem in sinistra numerantur; unum adde, ad dexteram transitur’).35 Literary and archaeological evidence suggests this system was used primarily for indicating integer sums: Pliny the Elder (Naturalis historia 34.33), for example, describes the ancient statue of Janus in the Roman Forum as having ‘fingers shaped so that they indicate the sign of 365 days’ (‘digitis ita figuratis ut CCCLXV dierum nota . . . indicent’), while a tomb relief from Isernia (now in the Museo della Civiltà Romana in Rome) appears to depict 33

See Maher and Makowski 2001; Crosby 1997; Kennedy 1981; Murray 1978; Glautier 1972; Taisbak 1965; Menninger 1977; Turner 1951. 34 Very little evidence at all exists for the practical conduct of calculations using Roman numerals. Some papyri from Roman Egypt present very basic arithmetical problems (e.g. P.Mich. III, 145.3.6, including the conversion of the length of a field measured in σχοινία to a measurement of area in ἄρουραι), but these are all using Greek numerals. A Latin treatise of Marcus Junius Nipsus (second century ce) gives instructions (300.11–301.5) for finding the area of any triangle using Roman numerals in a series of eleven or twelve distinct steps of addition, subtraction and multiplication. The final calculation, involving a square root, is given as a single step, which may suggest the availability of arithmetical tables. See Cuomo 2001: 170–6. 35 Cf. also Juv. Sat. 10.248–9, ‘felix nimirum qui tot per saecula mortem / distulit atque suos iam dextra computat annos’.

104

willia m mich ael s h o rt

two standing figures stretching out their hands to indicate different sums with their fingers, apparently negotiating the price of lodging at an inn. Some authors imply, however, that the system could be used for computation as well, ranging from simple additions – Suetonius (Divus Claudius 21.5) recounts that Claudius would ‘count out the gold coins offered to victors . . . both vocally and on his fingers’ (‘oblatos victoribus aureos . . . voce digitisque numeraret’) – to fairly complex calculations involving recursive sums, as when Augustine leads his congregation in a finger-counting exercise to add all the numbers from 1 to 17 (yielding 153, the number of fish caught by the disciples of Jesus when he returned to the Sea of Tiberius after his resurrection: Serm. 248.5 and 270.7, ‘apud vos numerate . . . si vero computes ab uno usque ad decem et septem et addas numeros omnes . . . sic pervenis usque ad decem et septem, portans in digitis centum quinquaginta tres’). This system would have been limited to sums up to 1,000, however (Williams and Williams 1995; Alföldi-Rosenbaum 1971), and was apparently prone to mistakes even by experts: cf. Apuleius, Apologia 89, ‘posses videri pro computationis gestu errasse, quos circulare debueris digitos aperuisse’; ‘you could seem to have erred in gesturing your calculation, since you opened the fingers which you ought to have formed in a circle’. Calculations could also be performed using the hand or table abacus, or the so-called sand ‘reckoning board’, which could in their most sophisticated form represent any number from 1 to 9,999,999 (cf. Schärlig 2001; Taisbak 1965). The simplest (and probably earliest) of these devices consisted of a small wooden board covered in sand, on which parallel lines were drawn with the finger or stylus; on, or between, these lines the user would then place stone counters to represent the number in each place (cf. Persius, Satires 1.131–2: ‘qui abaco numeros et secto in pulvere metas / scit risisse vafer’; ‘A clever fellow who knows how to laugh at numbers on an abacus and the boundary-lines in the divided sand’). More elaborate were the hand-held counting boards made of bronze, with vertical slots cut for increasing powers of ten usually up to 106 (marked with a mixture of Roman and Etruscan numeral symbols, viz. I = 1s, X = 10s, C = 100s, (I) = M = 1000s, ((I)) ¯ | = 1,000,000s) and round beads for counting.36 = 10,000s, (((I))) = 100,000s, |X Some extant examples have additional columns, probably for weight calculations with fractions (in units of twelfths and thirds).37 Yet these were highly inefficient 36

On a typical Roman abacus, each ‘column’ is actually divided vertically into two sections: the top section would have two beads each counting for 5, and the bottom section five beads counting for 1, so, e.g., 9 would be represented by one active bead in the top section and 4 in the lower section. Unlike the Roman system, where the ‘counters’ in each vertical section were identical and represented equal values (obviously according to their columnar placement), the evidence presented in Netz 2002 suggests that in Greek arithmetical practice, pebbles (ψῆφοι) of different sizes and shapes were used to represent different numbers and place values. 37 Almost all Roman measures of weight and land were based on a division of 1⁄12 (uncia): e.g. as the basic unit of weight, the uncia was 1⁄12 of an as or libra. Used in combination, the two additional abacus ‘columns’ could thus represent any land measurement from the scripulum (a plot ten feet wide by ten feet long) to the iugerum (288 scripula). See Maher and Makowski 2001.



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 105

technologies. A trained account-keeper (calculator) might perform addition and subtraction quickly (as quickly, in fact, as he was able to set pebbles down on the board or move beads on the frame). Calculations of multiplication and division, though, would have been quite laborious and time-consuming, since these operations could be accomplished only through repeated additions or subtractions. Furthermore, since these technologies could represent only a single amount at any given time, complex calculations probably required the support of written note-taking, as suggested by visual depictions like that on the so-called Darius Vase from Canosa in Puglia (probably fourth century bce) in which a figure holds an abacus in one hand and a writing tablet in the other (see Figure 6.1). The Roman senate faced a special challenge in computing vote tallies. Senate procedure permitted any member to express his opinion when voting on a given question. However, because the senate operated under strict time constraints – all sessions had to close at sunset, and if business had not been completed, voting would have to begin anew – in a body possibly as large as 900 members by 45 bce, it would have been impractical for each senator to speak personally at any length (or even at all). Furthermore, though the senate was in theory a body of equals, a hierarchy based on institutionally ascribed and personally achieved authority meant that when voting by roll call, most members would be expected only to give their assent or dissent to an opinion (sententia) already expressed by one of the

Figure 6.1  Detail after an artist’s rendition of the Darius Vase (South Italian red-figure vase found at Canosa, Puglia, dating to about 340 to 320 bce, now housed at the Museo Archaeologico Nazionale in Naples, Italy). The seated calculator manipulates a mounted counting board while holding a writing tablet.

106

willia m mich ael s h o rt

senior magistrates (Ryan 1998: esp. 64–87). Therefore, the practice of ‘voting with the feet’ (pedibus in sententiam ire) was devised as a more expedient method of vote counting. There appear to have been two procedural possibilities.38 In discessio (literally, ‘a walking apart’), the curial space would be divided into two regions (partes) and senators would be invited to indicate their agreement or disagreement with a particular opinion by physically moving to one or the other, as the formula recorded by Pliny the Younger suggests: ‘Those who vote for these things, go into this part; those who vote all other ways, go into that part according to your opinion’ (‘qui haec censetis, in hanc partem, qui alia omnia, in illam partem ite qua sentitis’, Epistulae 8.14.19). During interrogatio, on the other hand, the curia was notionally divided by reference to the spatial positioning of individual senators, the general zone around whom came to stand for a verbally expressed opinion; thus, as Herbert Nutting (1926: 426) explains, other members of the senate could indicate their agreement or disagreement with one or another stated opinion ‘by merely joining the party or man whose speech they supported’.39 (This is in part why no senator would sit near Catiline after his denunciation by Cicero: cf. Plutarch, Cicero 16.4. In isolating Catiline physically, the senators register their social rejection as well as political abandonment of the conspirators.) In whatever procedural context it was practised, ‘voting with the feet’ followed the same symbolic principle. To facilitate the counting of votes, specific areas within the senate house would be treated as corresponding to specific sententiae, through a figurative mechanism in which the physical location where a senator verbally expressed his opinion operated as a signifier metonymically and metaphorically standing for what existed as a purely mental representation – an idea in some senator’s mind. Effectively transposing a privately held belief on to a publicly available feature of the world, this process permitted others to signal their own subjective beliefs by physically moving themselves in relation to this feature. In this way, the problem of efficiently computing vote counts could be solved by, as it were, transforming the floor of the senate into a reckoning board and the senators into pebbles – but perhaps this is not so surprising a development in a cultural context where, as Reviel Netz claims (2002: 327), ‘the abacus is not an artefact; it is a state of mind’. The tallying of votes would then be achieved through gross physical comparison, as a direct outcome of the physical interactions of the group of senators. In other words, it is in the senators’ bodily behaviour that the process of calculation actually takes place. Indeed, the process of ‘voting with the feet’ seems altogether to obviate the need for calculation and for representing the vote in numerical terms (of whatever form). Because the senators themselves functioned as the ‘numbers’, in most situations where the count on each side would not be very close, a simple visual reckoning between places in the senate house would suffice to indicate the outcome. 38

See O’Brien Moore 1935. There has been much dispute over the applicability of pedibus ire to the senate’s different voting procedures: compare Taylor and Scott 1969 with, for example, Scullard 1973: 26. 39 Taylor and Scott 1969: 534–5; Ryan 1998; Lintott 1999: 79–82.



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 107

The Metaphorical Basis of Roman Distributed Cognitive Strategies In the orator’s technique of loci and the senator’s practice of pedibus (in sententiam) ire, the environment serves as an immediate resource for the scaffolding of cognition. But is the way in which Latin speakers made use of the physical environment in such contexts of technical practice simply to be attributed to the kind of spatial thinking Yuri Lotman (1990: 250–3) deemed characteristic of traditional societies? Or is there some dimension of Roman culture that specially motivates the forms their distributed cognitive strategies tend to take when tackling problems of memorisation and calculation? In my view, the basic symbolic principle according to which these strategies unfold, though based in experiences shared by presumably all human beings (namely, that the particular thoughts, feelings or memories consistently evoked by a place come, involuntarily and unconsciously, to be associated with that place), actually emerges from a distinguishing feature of Roman society’s signifying order. Very specifically, I believe that the semiotic mechanism underpinning these ­behaviours – that mental representations are treated as having a physical reality of their own – was conditioned by Latin speakers’ conceptualisation of mental phenomena in terms of an embodied spatial metaphorics, which, though partially paralleled in other languages and cultures, appears to constitute their ‘preferential’ or ‘privileged’ conceptual model for comprehending the mind and its various operations. An abundance of linguistic and literary evidence indicates that for Latin speakers ‘thinking’ – broadly construed to include cognitive operations of all sorts – was in fact a conceptual domain pervasively structured by metaphor (see generally Short 2012). In particular, cognitive processes having to do with the ‘contents’ of mind were construed in terms of a set of systematically interrelated images of movement through space relative to a location. In this system, the metaphorically defined concepts – ‘acquiring’, ‘having’ and ‘relinquishing’ thoughts – represent general categories of mental activity at a high level of abstraction, each image delivering a core conceptualisation that underwrites a whole network of conventional expression in Latin. Thus, as examples 1 to 6 below show, mental operations that involve ‘acquiring’ ideas – in other words, that involve thoughts either entirely new to the thinker, or to which the thinker newly turns his or her conscious awareness – are expressed metaphorically in Latin as movement toward a location in space:   1. ‘qui consilium iniere, quo nos victu et vita prohibeant  / is diem dicam, irrogabo multam’; ‘To those who formulated (literally “entered into”) a plan to prohibit us from vital nourishment, I will name the day and stipulate a fine’ (Plaut. Capt. 493–4);   2. ‘ni occupo aliquid mihi consilium, hi domum me ad se auferent’; ‘Unless I come up with (lit. “occupy”) some plan, they’ll carry me off home’ (Plaut. Men. 846–7);  3. ‘ipse quoque huic sententiae accedo’; ‘I agree with (lit. “approach”) this ­opinion, too’ (Just. Dig. 36.2.12.6);

108

willia m mich ael s h o rt

  4. ‘veniamus ad bonorum malorumque notionem’; ‘Let us consider (lit. “come to”) the idea of Good and Evil’ (Cic. Luc. 128);   5. ‘ne in cogitationem quidem cadit ut fuerit tempus aliquod nullum cum tempus esset’; ‘It is not even possible to conceive (lit. “it does not even fall into thought”) of a time when time did not exist’ (Cic. Nat. D. 1.21);  6. ‘ire in cogitationem iubet et dispicere quid ex hac tranquillitate sapientiae debeam’; ‘[Old age] tells me to consider (lit. “go into thought”) and examine how far I owe this serenity to philosophy’ (Sen. Ep. 26.3). As these examples show, Latin speakers normally talk about formulating plans, agreeing with opinions, considering ideas, conceiving notions and so on as ‘entering’, ‘occupying’, ‘coming to’, ‘returning to’, ‘moving to’ and even ‘falling’ or ‘slipping into’ a location. This metaphor constitutes the entirely regular way for Latin speakers to talk about such experiences, moreover. For instance, Latin expresses the concept of ‘paying attention to’ (that is, ‘acquiring’ something as the focus of mental attention) as, literally, ‘turning the mind toward’ (animum advertere), ‘directing the mind toward’ (mentem intendere) or ‘causing thought to go toward’ (cogitationem conferre ad), where mental attention is construed as directionality of thought toward a location. At the same time, words denoting position in a location (‘being’ or ‘standing in’) regularly structure the concept of ‘having’ an idea in mind – that is, the concept of any sort of mental activity that involves either the contents of an individual’s system of beliefs or thoughts constituting the object of conscious reflection: cf., e.g.,   7. ‘eum defixum in cogitatione esse sensisset’; ‘He realised that he was deep in (lit. “completely fixed in”) contemplation’ (Cic. De or. 3.17);   8. ‘nec mihi in cogitatione tum lex fuit’; ‘Nor was the law in my thought at the time’ (Quint. Decl. min. 270.25);   9. ‘quod eis respondi, ea omnes stant sententia’; ‘Whatever I tell them, they all hold (lit. “stand in”) that opinion’ (Plaut. Curc. 249–50); 10. ‘quamquam in falsa fuerit opinione, demonstrandum erit neminem tantae esse stultitiae, qui tali in re possit veritatem ignorare’; ‘It must be shown that, even though he was mistaken in that opinion (lit. “was in a false opinion”), no one can be so foolish as to be ignorant of the truth in such an affair” (Cic. Inv. rhet. 2.27); 11. ‘adhuc in hac sum sententia, nihil ut faciamus nisi quod maxime Caesar velle videatur’; ‘I still am of (lit. “in”) the opinion that we should do nothing but what Caesar seems most to want’ (Cic. Fam. 4.4.5); 12. ‘qua in sententia et Vergilium fuisse video’; ‘I see that Vergil, too, held (lit. “was in”) this opinion’ (Plin. HN 18.35). Again, this metaphor organises the Latin vocabulary in a systematic fashion. For example, while the ancient etymological tradition derived considerare from sidus, taking the act of mental deliberation as a kind of metaphorical ‘star-gazing’, Greenough



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 109

(1890) has suggested a derivation from sedes (‘seat, location’), which is not only more plausible phonetically, but also provides a motivated explanation for the verb’s meaning: if ‘being in’ means metaphorically ‘thinking over’, then by inference ‘completely being in a location’ (con-) will convey the sense of ‘completely thinking over’. Finally, expressions denoting movement away from a location (‘standing away’, ‘departing from’) consistently convey the concept of ‘relinquishing’ an idea from the mind – that is, giving up some idea that is under current consideration or abandoning some closely held belief, as in: 13. ‘necessario sententia desistunt legatosque ad Caesarem mittunt’; ‘Of necessity they abandon (lit. “stand away from”) this idea and send legates to Caesar’ (Caes. B.Gall. 6.4.2); 14. ‘perterriti Galli, ne ab equitatu Romanorum viae praeoccuparentur, consilio destiterunt’; ‘The Gauls, fearing that the passes were already occupied by the Roman cavalry, gave up on (lit. “stood away from”) this design’ (Caes. B.Gall. 7.26.5); 15. ‘aiunt ipsum sapientem . . . si ita rectius sit . . . de sententia decedere aliquando’; ‘They say the wise man sometimes changes his mind (lit. “departs from his opinion”) when it is better to do so’ (Cic. Mur. 63); 16. ‘ille vir . . . de civitate decedere quam de sententia maluit’; ‘That man . . . preferred to depart from his city than from his belief’ (Cic. Balb. 11). And again the metaphor is systematic. That is why digressio/digressus and egressio/ egressus, for example, are used in the sense of a departure from the idea that forms the main subject of some discourse – a digression being viewed metaphorically as temporary movement away from a certain thought-location. As may be seen, Latin’s metaphors of mental activity are both consistent, in the sense that they organise the figurative meanings of whole lexical fields (rather than belonging to the semantic structure of individual words), and coherent, in the sense that in organising Latin’s vocabulary of mental activity they operate as a system that preserves the inferential structure of spatial motion through metaphorical abstraction. The metaphors, that is, fit together in a structured way, each conveying a mental operation that is logically related to the others. What defines their logical relation is a central – but mostly implicit – metaphorical correspondence between the thought over which the mental operation occurs and the location in relation to which the physical motion occurs, and thus the systematic interrelations that hold in the spatial domain between movement toward, movement from and position in (a location) as schematised bodily experiences, which are mapped to mental activity as metaphorical entailments. Overall, the mappings from the domain of physical spatial experience to that of abstract mental experience that constitute this metaphor system can be represented as in Table 1. In the perspective of Lakoffian conceptual metaphor theory, the consistent as well as coherent metaphorical structuring of Latin speakers’ expressions of mental

110

willia m mich ael s h o rt

Table 1: Metaphorical correspondences in ‘thoughts are locations’ spatial motion

mental activity

the location in space movement toward (the location) position in (the location) movement from (the location)

the thought ‘acquiring’ the thought in mind ‘having’ the thought in mind ‘relinquishing’ the thought from mind

activity in terms of spatial images suggests that these metaphors actually constitute their conceptualisation of this domain. In this theory, metaphor is claimed to be not merely a linguistic or literary device, but a crucial structuring device of the conceptual system.40 Cognitive linguists posit that the clustering of metaphorical linguistic expressions around many (mostly abstract) concepts in fact reflects inherently metaphorical understandings that speakers of a language possess of those concepts. Speakers of a language talk about abstract concepts metaphorically, that is, because they actually conceive of them in terms of other (mostly concrete) concepts. Under this view, the meanings of many concepts are said to correspond to ‘image schemas’, gestalt structures of experience deriving from sensory and motor interaction with the world.41 Metaphors are regular projections or mappings of image-schematic structure that occur as a way of mentally representing abstracta. Moreover, it is the systematic nature of these mappings – that is, that they involve the projection of structured systems of concepts and a defined cognitive topology – that allows people to think and reason meaningfully about experiences that may be difficult to comprehend in themselves.42 It is through the metaphorical mapping of bodily based image-schematic structure on to concepts not directly grounded in experience that human abstract thought is in fact possible. Let me emphasise, then, that in my view it is these metaphors per se that deliver meaning to Latin speakers’ conceptualisation of mental processes. That is, as I see it, it is the image schemas of movement toward, movement from and position in a location (as depicted in Figure 6.2) in themselves that, in being metaphorically projected to the mental domain, constitute Latin speakers’ understanding of ‘acquiring’, ‘having’ and ‘relinquishing’ a thought in mind. Furthermore, while spatial metaphors of mind are identifiable in a large number of the world’s languages and may well be universal,43 these mappings clearly 40

Lakoff and Johnson 1980; Johnson 1987; Lakoff 1987; Kövecses 2005, 2006. See also Hampe and Grady 2005; Lakoff 1987, 1993; Lakoff and Johnson 1999. On the experimentally demonstrated psychological reality of image schemas, see Gibbs and Colston 1995; Gibbs 1994. 42 Cf. Lakoff 1993: 215: ‘Metaphorical mappings preserve the cognitive topology (that is, the imageschema structure) of the source domain, in a way consistent with the inherent structure of the target domain.’ 43 Spatial metaphors of mind are prevalent in English (see esp. Ryle 1949; cf. also Kövecses 2002: 98–101, 2005: 210–15; Lakoff and Johnson 1999: 236–8; Jäkel 1995: 222), but also in many typo41



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 111

Figure 6.2  Schematic images metaphorically underlying Latin’s expressions of mental activity. constitute a sort of privileged model of conceptualisation for Latin speakers. This emerges from the fact these metaphors represent Latin’s most conventionalised, most systematic and most potentially elaborated way of understanding and hence communicating about mental phenomena. Although an exhaustive statistical analysis is probably not practicable, even a cursory comparison with Latin’s other metaphors of mind (see Short 2012: 112–22) suggests that this one system of metaphorical understanding accounts for a larger number by far of conventional linguistic expressions. What is more, it is perhaps the only such metaphor characterised by a strict internal coherence of imagery, in the sense that logically interrelated spatial concepts define correspondingly logically interrelated mental concepts.44 The basic mappings of this metaphor system are also easily extended through the use of Latin’s highly articulated vocabulary of spatial motion, yielding complex metaphorical entailments and thus expressions that are rich in connotations vis-à-vis the mental domain. In constituting Latin speakers’ preferential conceptualisation of mental activity, these spatial metaphors can therefore be expected to play an important role in guiding the kinds of behaviours they elaborate in relation to their own cognitive activity (including, or especially, memorisation and calculation, which stand out as perhaps prototypical processes of knowledge acquisition). Of course, in suggesting this, I would not wish to imply there is any kind of determinative relationship between these metaphors as an aspect of Latin speakers’ language and thought and their behaviour in certain arenas of technical practice. While deeply entrenched in Latin speakers’ conceptual repertoire and indeed a defining feature of the Roman worldview, these metaphors are by no means the only ones Latin speakers had available to them in understanding the mind (see again Short 2012: 111–26). Thus, the operation of the ‘locations’ metaphor in Latin speakers’ conceptual system cannot be said strictly to require any particular way of behaving vis-à-vis the mental domain. Nevertheless, it cannot be coincidental that this metaphor system’s­­ logically distinct languages: Yu 2003: 143–9 reports that Chinese speakers, for example, employ combinations of certain words referring to spatial orientation and motion – left (si), right (you), through (tong), coming (lai), going (qu) – with the generic ‘thinking’ verb xiang to define a whole range of mental processes. 44 For the most part, the other metaphors making up Latin speakers’ overall conceptualisation of mind are ‘one-shot image metaphors’ that afford single images to figurative understanding: see again Short 2012: 136–7.

112

willia m mich ael s h o rt

central mapping – namely, ‘thoughts are locations’ – is effectively operationalised in their strategies for tackling cognitively demanding tasks through the use of the social and physical environment. In Roman rhetoric’s ‘topographical’ memorisation technique and the Roman senate’s ‘spatial’ calculation of votes, this metaphor system actually appears to serve as a ready-made plan for action for extending and distributing cognition out into the world – a symbolic model that picks out salient aspects of experience, orients awareness toward their relation, and even suggests a directionality to this relationship.45 Indeed, the practices of other cultures in similar contexts of cognitive performance suggest that this kind of spatial thinking vis-à-vis memorisation and calculation represents a real cultural choice on the part of Latin speakers: why not, in fact, simply adopt the Greek practice of voting with pebbles, as they had adopted so many other things from this prestige culture? Or why not obviously extend the practice of finger-counting to the other parts of the body for more complex calculations, as for instance many native societies of Papua New Guinea do?46

General Discussion Facing certain challenges presented by the nature of human memory (especially in terms of the limited capacity and rapid decay of short- and intermediate-term memory) and given certain cultural features like the Roman numeral system, Latin speakers resorted to extended and distributed cognitive strategies as a means of supporting and indeed implementing memorisation and calculation in their technical practices. These strategies exploited both the social and the physical environment as a scaffold for cognition: for instance, the difficulty of remembering the names and titles of one’s many clientes could be mitigated through use of a ­nomenclator – a slave whose job was to learn certain minutiae of social life and to ­­proactively remind his master of these details in such a way that the master appeared to know them himself. In circumstances calling for memorisation of a lengthy text, meanwhile, the orator could rely upon a symbolic process in which his movement through the physical terrain or some architectural setting was used to provide structure to his memory. For arithmetical calculation, a number of different body-based and technological devices (finger-counting, the abacus, the sand 45

For metaphors functioning as generic ‘sign-images’ that may take various ‘tropic’ forms when transposed into different representational (visual, behavioural, gestural) codes, see Fernandez 1991, 1986. More generally, this system constitutes a shared context of meaning in which some forms of behaviour will ‘make sense’ more than others. At the same time, these forms of behaviour also, to some degree, warrant or license Latin speakers’ particular ways of thinking and ways of speaking about mental phenomena (namely, in terms of locations) through a sort of semiotic feedback loop. In this sense, the metaphors – as a part of Latin speakers’ conceptual system used both in interpreting and in guiding behaviour – function precisely in the way that Geertz (1973: 93) characterised the operation of cultural symbols, as both ‘models of’ and ‘models for’ reality. On the role of metaphor in motivating many problem-solving behaviours, see Sarbin 1986. 46 See Rauff 2003; Saxe and Esmonde 2012.



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 113

reckoning board) were employed to circumvent the inefficiencies of the Roman numerals. In the Roman senate, where above all a timely and efficient method of tallying was needed, the calculation of votes was implemented as a function of the senators’ bodily positioning in space relative to one another. I have claimed, moreover, that what often motivates the particular (not to say peculiar) form of these practices is Latin speakers’ all-pervasive metaphorical conceptualisation of thoughts in terms of locations. These have only been hints. Nevertheless, they offer several points for general reflection. For scholars of ancient societies, this study suggests the need for an approach that considers not only how universal aspects of human embodiment contribute to abstract conceptualisation in Latin, but also how embodied metaphorical understanding underpins specifically Roman forms of practice. As I have argued elsewhere, the study of metaphor along cognitive linguistic lines becomes anthropologically revealing when it permits us to highlight differences in metaphorical ­­conceptualisation, and above all when patterns of conventionalised figurative meaning in language can be linked to patterns of representation in a society’s symbolic activity at large.47 Thus, the orientational metaphors that Francisco García-Jurado (2000) has shown to be present in the language of Plautus (‘good is up’, ‘good is hot’, and so on), in being grounded in what is probably a universal human experience (the correlation of physical and emotional well-being with bodily uprightness and warmth), are interesting to the degree that they confirm the commonality across societies – including historical ones – of certain experience-based conceptualisations. Likewise, the ontological metaphors that Chiara Fedriani (2014) has shown to structure much of Latin’s grammatical system – ‘states are containers’, ‘events are movements’, ‘experiences are things possessed’ – in reflecting the operation of a very high-order metaphorical conceptualisation known as the ‘Event Structure Metaphor’, known to be widespread across even typologically unrelated languages, help substantiate the universalising claims of some cognitive linguists. However, these metaphors are less relevant to the characterisation of what is unique about Roman culture than where Latin’s metaphors seem to coalesce around concepts in idiosyncratic ways or play out as ubiquitous themes of the signifying order.48 Latin’s ‘locations’ metaphor of thoughts probably also has its grounding in a universal dimension of human embodiment: namely, that we perceive the patterns of thinking we find ourselves engaged in in certain environments as directly linked to those environments. Yet comparison with the ‘same’ metaphor in English (as reflected in expressions like ‘come to an understanding’, ‘arrive at the end of an argument’, ‘take a position’) or in Greek (as manifested in, for instance, uses 47

See in particular Short 2012, 2013a, 2014; cf. Deignan 2003. On metaphorical themes, see esp. Goatly 2007; cf. also Shore 1996 on the role of cognitive models more generally in shaping cultural artefacts. Danesi and Perron (1999: 294) refer to such culturally pervasive cognitive structures as ‘macrosignifieds’, that is, ‘signifieds that underlie the specific forms that various signifying structures assume across the signifying-order’.

48

114

willia m mich ael s h o rt

of προσβιβάζειν, ‘cause to approach, bring near’, in the metaphorical sense of ‘persuade’, or of spatial prepositions like δια-, ἐκ-, ἐν-, ἐπι- and παρα- in certain compound verbs referring to mental activity)49 suggests its sociocultural situatedness is equally relevant. While in English the metaphor focuses narrowly on the conceptualisation of rational thought (the broad categories of ‘acquiring’, ‘having’ and ‘relinquishing’ ideas more commonly figured through images of object possession or manipulation, as these labels suggest: cf. esp. Jäkel 1995), and in Greek it is elaborated linguistically to a far lesser degree (limited largely to prepositional semantics), for Latin speakers it delivers a very basic part of their understanding of the mind – namely, how conscious attentional processes work – and structures perhaps the largest portion of their vocabulary of mind. Indeed, it is probably fair to say that the images of the ‘locations’ metaphors provided Latin speakers with their principal means of comprehending the mind and its various operations. Not only does this one system of metaphorical understanding account for a larger number by far of conventional linguistic expressions than other metaphors, but the internal coherence of the images that make up this metaphorical system – that logically interrelated spatial concepts define correspondingly logically interrelated mental concepts – and the complex ways in which its basic mappings are extended and elaborated indicate it was of paramount importance in Latin speakers’ metaphorical understanding of mind. Frequent comment by authors on the close linkage between physical locations and thoughts implies that space was in fact inextricably bound up with how mental activity was understood in Roman culture. Quintilian (Inst. 11.2.17), for instance, notes the commonly occurring experience that ‘when we return to a place after a considerable absence, we do not simply recognise the place itself, but we also remember things that we did there, we recall the persons whom we met there, and even the unspoken thoughts which passed through our minds when we were there before’. Cicero (De finibus 5.2) expresses a similar notion through the mouth of Piso: ‘I myself, looking upon the old senate house of Hostilius, used to think of Scipio, Cato and Laelius; but most of all, of my grandfather – so great is the power of suggestion residing in places’ (‘tanta vis admonitionis inest in locis’). This notion is stated more explicitly in De oratore (2.358), where Cicero seems to sug49

e.g. Ar. Eq. 35, εὖ προσβιβάζεις μ’; Xen. Mem. 1.2.17, πάντας δὲ τοὺς διδάσκοντας ὁρῶ αὑτοὺς δεικνύντας τε τοῖς μανθάνουσιν ᾗπερ αὐτοὶ ποιοῦσιν ἃ διδάσκουσι καὶ τῷ λόγῳ προσβιβάζοντας. Forms like διανόησις, ἔννοια and ἐπίνοια, where the process of ‘thinking’ is construed as somehow ‘through(out)’, ‘in’ or ‘on’, suggest Greek has a spatial metaphor very much like Latin’s; however, these usages appear to imply an image of thought as static locationality, whereas Latin’s metaphor is based on the image of movement (along a path) in respect of a determined thought-location. At the same time, usages like ἄντικρυς, literally ‘straight’ = ‘true’ (as in Thuc. 8.64, ἡ ἄντικρυς ἐλευθερία), or διορθόω, διόρθωμα, literally ‘making straight’ = ‘correcting’ (as in Plut. Alex. 8.2.1, τὴν Ἰλιάδα . . . διορθώσαντος), and, conversely, σκολιός, ‘crooked’ = “unjust”, or ψάγιος, ‘askew’ = ‘blundering’ – where the metaphor is of spatial linearity – have to do with the ‘correctness’ of thought (see Short 2013a: 143–4) or the ‘intricateness’ or ‘difficulty’ of thought (as when ἀγκύλος, σκολιός, καμπύλος mean ‘wily, crafty’), more than with notions of the contents of the mind, which is the main meaning focus of Latin’s ‘locations’ metaphor.



e mbo died, ex ten ded an d dis t ri b ut ed co g ni t i o n 115

gest that thought is not possible unless somehow spatialised: ‘Forms and objects, like all things . . . require to be situated (sede opus est), for an object cannot be understood without a place (corpus intellegi sine loco non potest).’ Recent scholarship has shown just how much Latin speakers in fact conceived of the mental in terms of the spatial, utilising images of landscape and topography to, for example, provide logical structure to arguments (Leach 1988), describe characters’ mental and emotional states (Vasaly 1993), or represent different stages of philosophical development (Henderson 2006).50 Florence Dupont (1992: 73) claims that ‘The Romans lived and thought in spatial terms’. In this sense, while Latin speakers’ ­­metaphorical conceptualisation of thoughts as ‘locations’ may be based on universal bodily experiences, their privileging of such experiences in understanding the operations of the mind as well as in the elaboration of extended and distributed cognitive strategies constitutes a distinctive feature of their ways of representing, understanding and being in the world: in a word, of their culture. At the same time, for theorists of extended and distributed cognition, this study suggests the need to consider empirical connections between aspects of cognition that are normally treated separately. Indeed, though the different strands of research tied together under the moniker of ‘4E’ theory have some common ground – in particular, their rejection of any view of cognition as abstract symbol manipulation, and their commitment instead to treating the human body’s sensory and motor capacities as directly relevant to forms of thought – to my knowledge, at least, very little attention has been paid to the ways in which cognitive processes that are extended, embedded, enacted or embodied may interact, and especially to the ways in which generally situated cognition may be bootstrapped to embodied semantic knowledge. But if the distributed cognition claim is that human beings naturally take advantage of material artefacts or other structures of the external environment in their thought processes, what different types of artefacts and structures they tend to use in thinking and why different social groups go about utilising different artefacts and structures would seem to be relevant questions.51 Metaphorical conceptualisation may provide a partial answer. To be sure, the bartender who arrays glasses of different shapes on the bar in front of him does so to ease the burden of memorising a diverse set of drink orders (Beach 1988). But the spatial arrangement he uses will probably be determined by the particular metaphorical construals his culture provides vis-à-vis space (such as ‘important is central’) and time (e.g. ‘the past is behind’, ‘the future is in front’).52 Likewise, Even if an author like Seneca is translating a Greek idiom (proficiens = ὁ προκόπτων) that figures philosophical development as spatial advancement (i.e. ‘cutting one’s way forward’: cf. e.g. Epict. Diss. 1.4.1.1), the Roman Stoic elaborates this metaphor as a (sea or land) journey tout court: in his letters, as Gunderson (2015: 41) writes, the metaphor is ‘dramatically re-staged for the eyes of the reader’. 51 Heersmink (2013) attempts to formulate a ‘taxonomy’ of artefacts used in situated cognitive systems. 52 For these metaphors, see Lakoff and Johnson 1980: 92; Boroditsky 2000; Alverson 1994; for their effects on non-linguistic spatial tasks, see Gentner et al. 2002. 50

116

willia m mich ael s h o rt

when we imagine Otto choosing to rely on a notebook for recording his memories, this is again because his (our) culture conceives of memory above all in written terms.53 Relatedly, Roman technical practice can remind us not to over-emphasise the importance of mechanical and digital technologies in the history of distributed cognition – technologies such as iPads, navigational charts and plotting tools or networked computers and web search engines – or at any rate to be more aware that the use of technologies like these in extended and distributed processes of cognition is historically and culturally conditioned. No doubt innovative technologies like these have been important in permitting human beings to extend and distribute our cognition out into the environment and thus enhance it in significant ways. Yet, as we have seen, in historical societies where such technologies could not even be contemplated, still human beings naturally and effortlessly extended their thinking beyond skin and skull – not only to facilitate thought, but to realise new forms of thought. This reminds us that human beings may be ‘natural born cyborgs’ or, in Clark’s terms (2003: 7), ‘creatures whose minds are special precisely because they are tailor-made to mix and match neural, bodily and technological ploys’. More fundamentally, however, we are – to recall Edwin Hutchins’s formulation (1995: 172) – ‘cognitive bricoleurs’, assembling thought out of whatever is at hand: other people, an arch or aqueduct, a floor.

53

Perhaps like Leonard Shelby, the protagonist of Christopher Nolan’s Memento (2000), he could also have tattooed significant information on his skin in place of memory. But this choice would probably also be motivated by culturally defined notions of the body as ‘text’.



r oman -per iod t h eat res 117

7 R o m a n - pe r i o d Thea tres a s D i s t ri but e d C o g n i t i v e Micro-ecolog ies Diana Y. Ng

While Roman drama – the comedies, tragedies, pantomimes and so forth – enjoyed a very short floruit,1 the venue for these performances – the stone theatre – is one of the most widespread and enduring architectural types across the entire Roman Empire. The Roman-period theatre accommodated not only what are now considered typical theatrical – namely, dramatic – performances, but also festivals established by elites as public benefactions, the gathering of the civic assembly for public business and other occasions. As such, the theatre simultaneously was deeply embedded within the cultural, social and historical contexts of the cities in which they were built and played host to the activities that nurtured those contexts. Due to the variety of functions that it served, the theatre was also open to citizens and foreigners alike, with each group bringing its own knowledge – or lack thereof – of the organisation and structure, social relationships, history and public personalities specific to a city. How, in buildings designed for communal congregation and interaction, was the theatre-goer’s cognition of multiple aspects of urban society supported? I argue that in this building type, cognition was distributed across the individuals in attendance, the seating arrangements and the decoration of the building. To be in the theatre was to see, occupy and know one’s place relative to others and – in some cases – to the larger entities of region and empire. In other words, it was a cognitive micro-ecology.

Distributing Cognition in the Roman-period Theatre Evelyn Tribble’s model of the Shakespearean theatre as a cognitive ecology is profoundly influential on the present study. The interdependence of multiple tools and of internal and external representations in creating a functional ‘computational environment’, and the impact of evolving circumstances (for example, the creation  1

Roman comedies and tragedies produced in the third and second centuries bce borrowed heavily from Greek predecessors, and it is not clear that any specifically culturally Roman plays were produced. The full production of new Roman plays had ceased for the most part by the end of the first century ce. See Bieber 1961; Beacham 1992; Potter and Mattingly 1999.

118

dian a y. ng

of new kinds of charts, projections, plotting tools and so on) on different elements of this environment, or cognitive ecology, was laid out in Hutchins’s Cognition in the Wild. In that work, the ‘historically contingent’ nature of humans’ understanding and usage of external tools was made clear (Hutchins 1995: 114). Using Hutchins’s work as a guide, Tribble and John Sutton called for a historical cognitive science that likewise stressed a ‘system-level mode of analysis’ that not only acknowledged but centralised the notion that human cognition and its engagement with things beyond the skin was historically, culturally and socially determined and bound together (Tribble and Sutton 2011: 95–7). The skills necessary for completing cognitive tasks – be they memory, sorting or categorising – in such a cognitive ecology were not merely biological capacities, but also shaped through exposure to the social (including educational or occupational) milieux and technologies in existence within a given historical period (Tribble and Sutton 2011: 97–8; see also Anderson 2015). Tribble systemically analysed the early modern Globe theatre as a cognitive ecology and highlighted how its operation depended on the confluence of many related external cognitive tools (Tribble 2005, 2011). The Globe was a specially designed place with certain physical and technological features such as three stage doors that facilitated the actors’ recognition of performance cues (Tribble 2011: 40–4, 2005: 143–5). The effectiveness of metrical or rhyming verse was a scaffold for the actors’ remembering and confident delivery of lines in the light of the absence of expectation of exact textual fidelity – itself related to the more limited role of print in the early modern world (Tribble 2011: 71–6, 2005: 149–52). The basic spatial and narrative information provided by plots allowed actors to cope with a large cognitive burden of many roles in many plays performed within a short span of time (Tribble 2011: 47, 2005: 146–7). The skills of boy actors in using all of these cognitive artefacts were made possible by the then socially normal apprenticeship model that trained these young performers to play female parts with a high degree of professionalism (Tribble 2011: 126–50). There was, Tribble demonstrated, no such thing as a discrete place of Shakespearean performance. Everything that was part of the Globe was the product of the technologies, economics and cognitive habits of early modern England (Tribble and Sutton 2011: 99). Extending the system-level analysis of architectural space to the churches of Reformation-era England, Tribble and Keene likewise demonstrate that the building, its interior decoration and the way that both of these shaped the congregants’ bodily experience and consequently their attention and memory were linked to profound changes in theology and culture (Tribble and Keene 2011: 47–70). The interiors of previously Catholic churches were radically altered, with the removal of idolatrous visual imagery and the phasing out of stained glass, the stripping away of reliquaries, the doing away with the stone altars and the obscured yet visible rituals of Mass, and the cessation of aromatic and complex musical and aural elements (Tribble and Keene 2011: 48–54). The symbolic system of Catholic religious practice was destroyed. Instead of a sort of theme park that pulled the worshippers’ ­­ attention in multiple directions at the same time and which – in the eyes of



r oman -per iod t h eat res 119

Reformers – prevented thought and true spiritual engagement, the Protestant transformation of these buildings into austere places accommodating long periods of sitting, like the introduction of the large overhanging pulpit as the visual and aural focus, was (as Tribble and Keene convincingly demonstrate) part of a new cognitive ecology designed to prioritise the Scriptures and to enhance memory of the preached sermon (Tribble and Keene 2011: 54–60). The congregation’s encounters with the printed words of the Bible and the heard words of the preachers’ sermons at the heart of Protestantism were no longer subject to distracting bells, pictures and movement. Beyond these cognitive scaffolds for new modes of worship, external representations of social class still found their way into the newly configured church, as the front rows of pews placed to the front of the nave closest to the pulpit were more highly decorated and occupied by members of the gentry (Tribble and Keene 2011: 58). The fruitfulness of the historical cognitive sciences for buildings that were not specifically suited to technical adepts such as the trained actors of the Globe, but rather designed for the masses under the influence of – and to influence people’s responses to – larger historical and cultural currents, provides an ideal model for the analysis of the Roman-period theatre as a civic centre. In calling for further developments in an interdisciplinary historical cognitive science and its relationship to the extended mind hypothesis, Sutton has cautioned against what he calls ‘first-wave Extended Mind’, an approach that prioritises the principle of parity that Sutton claims requires the external devices to perform the same cognitive processes as internal neural structures (Sutton 2010: 194–201, 2007a: 40–1).2 Rather, he defines a ‘second-wave’ extended mind that is based on ‘complementarity’: that exograms or external cognitive tools are not equivalent to but supplement biological processes that are inherently unstable (Sutton 2010: 204–8, 2007a: 41–4). While Sutton’s model is not in fact a radical break from the ‘first wave’, his discussion of complementarity is nonetheless significant as it centres on information encoding and retention in historical practice, such as the use of loci that trace back at least to the Roman Republican period and that were important in the medieval era.3 The emphasis on complementarity draws the second-wave extended mind closer to Hutchins’s distributed cognition, Sutton notes. Distributed cognition, and the idea of external cognitive artefacts complementing internal representations and processes, is crucial to the analysis of the Roman-period theatre  2

I am very grateful to Miranda Anderson for pointing out to me in the following citations Andy Clark’s own earlier explications that exograms are not necessarily one-to-one equivalents to internal cognitive mechanisms and processes. Clark (1997: 220) cautions us against ‘mistaking the problem-solving profile of the embodied, social and environmentally embedded mind for that of the basic brain’ as the exograms ‘are best seen as alien but complementary to the brain’s style of storage and computation’. Clark (2008b: 97–9) emphasises that ‘functional similarity’ of engrams and exograms is demonstrated in the process of information retrieval and the acts of interpretation and reasoning that then occur so as to shape subsequent behaviour.  3 See Short, this volume. Sutton considers the mnemonics of the loci and the memory to function as though they were external devices in having a static quality, and in being easy to manipulate and reorder as necessary; Sutton 2007b: 26–8, 2010: 208–13.

120

dian a y. ng

below. Not only is Sutton’s promotion of interdisciplinarity the basis of this chapter, but the complementarity model is, I believe, more in line with ancient evidence pertaining to the aspects of theatrical experience under consideration. The theatre building, spatial arrangement and sculpture are characterised as scaffolds that support the transmission and cognition of new information or interact with internal representations to produce a more complete knowledge of important information. As will be demonstrated below, they possess many of the properties of exograms or external storage devices – or cognitive artefacts – as defined by Donald: they come in a range of formats that are by nature physically permanent; they are expandable and enable different kinds of organisation and cataloguing; they can be ‘reformatted’ or cleared and reused; their interpretation requires ‘visual interpretive skills’ that, according to the historical cognitive sciences, are culturally and socially acquired (Donald 1991: 314–19). In addition, as Kirsh has pointed out, the interaction with external resources offers the advantage of greater cognitive efficiency. The cognitive effort associated with internal retrieval and articulation of abstract concepts and specific facts is reduced through architectural organisation and visual representation. In addition to remembering, the cognitive tasks of sorting, comparison and incorporation of new things within existing systems are facilitated by the cooperation of internal and external resources (Kirsh 2013b: 181–5). The analysis of the Roman-period theatre in such a system-level and complementarity-focused framework demonstrates that the building, its decoration and the way people occupied the space allowed for the quick explanation and comprehension of prevailing social organisation and relationships in Rome and in cities of Asia Minor, and allowed the theatre attendees to make sense of historical and political developments.

Seating Assignments and Costume as External Symbols of Class and Prestige In considering how cognition and memory are distributed in the total physical environment of the Roman-period theatre, it is important to bear in mind that the theatre is an architectural form specifically designed not just for crowds but also for creating attentional direction. The seats are constructed so as to face both downward and forward towards the performance/activity areas of the orchestra and the stage building and towards the sections of seats below one’s own vantage point. This same arrangement occurs also in bouleuteria and ōdeia, which are in essence smaller, roofed theatres for a more limited audience. Such architecturally controlled viewing is not compatible with other urban public spaces such as markets, baths or gymnasia, stoas or basilicas or streets. In an amphitheatre, the lower rows of seats could be seen by all, but the arena was the visual focus; as multiple events took place simultaneously in the arena, the division of audience attention was intended. There was no permanent stage building that consistently confronted the viewers’ collective gaze as in the theatre. Objects such as public portraits that crowded the other public spaces of Roman cities would – by their placement in the stage building or closely adjacent to it – be rendered more conspicuous.



r oman -per iod t h eat res 121

Thus, the Roman urban dweller or visitor coming into the theatre was physically oriented into a position of purposeful viewing. Unlike in modern theatres or cinemas, with their darkened auditoria and bright stages or screens, in a Roman-period theatre the cavea, or seating area, was also an important visual target that was regulated beyond practical needs for ingress and egress. In Roman-style theatres, the cavea was divided according to social rank. Senators, equestrians and plebs had their own places to sit as assigned by law, first by the Lex Roscia Theatralis of 67 bce, then by the Lex Julia Theatralis under Augustus (Rawson 1987), which mandated that the seats in the orchestra were to be reserved for the senators, and that equestrians would have their own fourteen rows of reserved seats behind them. In the Large Theatre at Pompeii, built prior to the stone theatres in the city of Rome, the seats right at the orchestra, called bisellia, are broader than other seats, and are set apart both in appearance and spatially from other parts of the cavea. These seats were accessed through entrances located at the sides of the orchestra, which themselves carried the boxes that were set aside for magistrates and other guests of honour, such as the Vestal Virgins. The equestrians were separated from the senators by a low wall, and instead of entering the cavea through the orchestra, they had their own entranceways which opened into the first praecinctio, or horizontal passageway separating levels of seating (Sear 2006: 130–2; Parker 1999: 163–4). In Asia Minor, one sees the same kind of architecture at the extremely well-preserved Roman-style theatre at Aspendus in Pamphylia. There are the similar boxes, the entranceways to the lowest course of seats, and the high wall dividing the ima cavea from the rest of the auditorium (Sear 2006: 366–7). The social division between classes is thus externalised into the spatial environment of the theatre, and an abstract yet fundamental concept that regulated Roman society is made tangible. Not only is the separateness of senators, equestrians and plebeians made clear, but the hierarchy between these groups is inversely manifested in the seating positions – lowest seats closest to the orchestra for the highest class, then steps backward and upward for each class below. The conspicuousness of the Vestal Virgins and other honoured guests in separate boxes added further visual detail that fleshed out the picture of Roman public prestige (Parker 1999: 166–7). The visibility not only of the Roman social ranks but of their members facilitated the sharing of information that would otherwise be too unwieldly to manage; there is no need for lists of names or confirmation of rank-qualifying wealth or offices when one can simply see who is seated where in the theatre. In the event that a senator should be demoted, or an equestrian elevated, the information about that change would also be represented and thus transmitted throughout the assembled audience in the theatre. The savings in both cognitive effort and time needed to grapple with the small and technical pieces of information that result from a direct spatial and visual arrangement of bodies in the theatre are therefore significant. These seating arrangements and the architectural features of the theatre that supported them are clearly a product of legislation and therefore of the social and historical contexts of the late Republican and Augustan eras that then carried

122

dian a y. ng

forward into later periods. The primary users of these cognitive artefacts had to be deeply embedded within those same contexts in order for the seating and spatial separation to make sense; these exograms then complemented internal representations of the Roman social orders. Yet, these external representations created a transmissibility of both abstract concept and concrete facts that also benefited those who had very little familiarity with Roman society. One passage from Tacitus’ Annales is especially informative about how the spatial divisions of the cavea and the seating arrangements were crucial to the initial encoding of social hierarchy on the part of newcomers, and were used by Romans themselves to reinforce knowledge of the institutions at the core of Roman social life. During the Frisians’ tour of Rome inter ea quae barbaris ostentantur intravere Pompei theatrum, quo magnitudinem populi viserent. illic per otium (neque enim ludicris ignari oblectabantur) dum consessum caveae, discrimina ordinum, quis eques, ubi senatus percontantur, advertere quosdam cultu externo in sedibus senatorum; et quinam forent rogitantes, postquam audiverant earum gentium legatis id honoris datum quae virtute et amicitia Romana praecellerent, nullos mortalium armis aut fide ante Germanos esse exclamant degrediunturque et inter patres considunt. they visited the usual places shown to barbarians, and among them the theatre of Pompey, where they were to contemplate the size of the population. There, to kill time (they had not sufficient knowledge to be amused by the play), they were putting questions as to the crowd seated in the auditorium – the distinctions between the orders – which were the knights? – where was the senate? – when they noticed a few men in foreign dress on the senatorial seats. They inquired who they were, and, on hearing that this was a compliment paid to the envoys of nations distinguished for their courage and for friendship to Rome, exclaimed that no people in the world ranked before Germans in arms or loyalty, went down, and took their seats among the Fathers.4 This passage confirms that the seating by rank in the theatre made it possible for the Frisians to perform a variety of cognitive tasks. They were able to perceive the separation between the Roman orders, and to pose questions as to which order was located in different sections of the cavea. Having understood that the lowest set of seats were reserved for the senators, the cream of Roman society, the Frisians were also able to tell that there were non-Romans in positions of prestige and to inquire after the reason. Crucially, when informed of the virtues that allowed the other foreigners to sit in the lower ranks, the Frisians were quickly able to project themselves into the Roman social order, and, as they ‘went down’, to physically insert themselves into what they considered to be their rightful place, not only in the theatre and Roman society, but in the multicultural empire as well. From the  4

Tac. Ann. 13.54, translated by Grant 1996.



r oman -per iod t h eat res 123

perspective of the Roman tour guides, the socially segregated seating was a cognitive artefact that served as a scaffold for knowledge of Roman society encoded over a lifetime within the prevailing system, allowing for the easy distillation of that system and for a basis for communication to others of imperial diplomatic friendships. Furthermore, the Romans’ control over this exogram allowed for the easy integration of new elements – in this case both the Frisians and the honoured foreigners that they had noticed – into their correct social and physical positions within the cognitive ecology as historical or political circumstances demanded. The seating arrangement in the Greek-style theatres of Asia Minor was different from that mandated in Rome. Preserved in the cavea at some theatres, including that of Hierapolis (Ritti 1985: 118–22) and Pisidian Antioch (Chaniotis 2007: 61; SEG 50.1290), are inscriptions of the names of the tribes to which different segments of a culturally Greek city belonged; this is an indication that these cities seated their audiences by their historical intra-polis affiliations. This practice is part of a long Greek tradition, evidence of which we find at the Theatre of Dionysus in Athens (Bieber 1961: 71). In the great theatre of Ephesus in Asia, the markers of tribal seating were, in the early second century ce, very conspicuous and impressive, as we can tell from the information given in the inscriptions of the Salutaris foundation and recovered statue bases verifying the terms of this text (IK 111 27; Wankel 1979: 167–204).5 As part of a civic ritual endowed in 104/5 ce by an Ephesian citizen named C. Vibius Salutaris, on the occasion of nearly every theatrical gathering, thirty-one portable silver and gold images, representing Artemis, selected Roman emperors, Roman and Ephesian civic institutions, legendary and historical city founders, and personifications of the tribes of Ephesus, would be present. In the theatre, these shining effigies would be set up, not on stage, but in the cavea on nine bases, grouped according to the seating of the boulē, the gerousia, the ephēboi and each tribe of the city (IK 111 27.202–7). Some bases for several groupings of statuettes have been found in the Ephesian theatre, confirming the seating by tribe as in other Asian or classical Greek theatres (IK 111 36A–D). Related to the seating arrangements, a concession was made for those who would actually carry these sacred statues back and forth from the temple: these priests and sacerdotal servants asked for seats in the first section of the theatre, and the city decreed that they should be granted their request (IK 111 27.440–2). Thus, information about the organisation of the Ephesian population, tracing its origins back to its legendary early history, about the political structure of the city and about the prestige of the religious cult of Artemis Ephesia was embedded in the physical environment of the theatre and guided the way in which people occupied the space. Just as in Rome, a stranger to early second-century Ephesus visiting the theatre on a day of official government business or for the enjoyment of festivals would come quickly to understand the society of Ephesus by being situated in a building specially designed to embody  5

See also Roger 1991, which provides his English translation of the text based on Merkelbach’s edition in IK 111. Ng (2018) analyses the interrelationship between procession, effigies and the inscribed text as cognitive supports for memory, based on periodic rehearsal of information.

124

dian a y. ng

the history and organisation of the entire city in a way similar to that experienced by the Frisian visitors to first-century ce Rome. The seating of the public by tribe and – in Rome – by social order in specific zones of the theatre not only aided in the encoding of information for newcomers to the community, but also reinforced citizens’ identification with the subgroup to which they belonged. Though on the streets and in the marketplaces their bodies would mingle and jostle, within the architectural confines of the theatre each person’s awareness of his or her place in relation to others was impossible to ignore. Another long-standing feature of Greek theatrical seating – and one that finds conceptual but not formal parallels with Roman theatres – was the allocation of marble proedria or seats of honour to especially prestigious attendees. Proedria of the type preserved at the theatre at Priene were located right at the edge of the orchestra as large marble chairs contrasting with the rest of the bench seating (Sear 2006: 5). Other theatres, such as those at Pergamon and Hierapolis, included small, formally distinctive sections of seats in the lowest levels of seating above the orchestra that were obviously meant for special guests or citizens of high standing.6 The sections resembled balconies and were located several rows up, in order to be able to look directly at the very elevated stages of the stage building. The honour of sitting in these places is mentioned several times by Dio Chrysostom in his oration to the Rhodians, delivered to its popular assembly most likely in the city’s theatre. The right to this special seating, even if granted only once by a great city like Rhodes, he claims with perhaps some rhetorical exaggeration, surpassed the prestige of a permanent statue voted by some lesser community (Orationes 31.110). He also links this privilege, as well as crowns and statuary (discussed below), with being kept in remembrance (Or. 31.16), which suggests that he and his audience perceived these physical and spatial dispensations as aides which preserved the citizens’ memory of special individuals. Adding to the effect of the physical separation between the classes in Rome was the clothing worn by the crowd, with the very poorest wearing dark clothing as well as sitting in the summa cavea, while citizens in higher social ranks had white togas with various levels of coloured trim according to rank: white, scarlet and purple trim for equestrians, broad purple trim for senators (Parker 1999: 166–7). These clothing exograms, not necessarily a part of daily life, complemented individuals’ engrams of social class membership and allowed for easy comparison and sorting of other citizens in a large group setting. The Roman social system was supported by its citizens’ procedural memory for putting on and reading the costumes of rank. Just as with seating, dress also facilitated the incorporation of novel elements and was available to others from outside Roman society to use in their growing knowledge of Roman culture. In the episode described by Tacitus, the Frisians used the visual cue of costume to detect non-Romans, and then to find a place for their delegation that matched their own estimation of Frisian bravery and loyalty  6

For Pergamon, see de Bernardi Ferrero 1970: 27; for Hierapolis, see de Bernardi Ferrero 1966: 57–8.



r oman -per iod t h eat res 125

to Rome. The combination of layout, seating and costume helped the Frisians to identify a larger framework, the Romans to communicate the framework and the Frisians to create a space for themselves. The culturally and historically specific cognitive ecology of the theatre in Rome is bound to and continuously shaped by lived circumstances beyond its walls. The conspicuity of dress for certain members of a theatrical audience also occurs, in a different way, in the Greek East. A long dossier of inscriptions from Oenoanda in Lycia detailing the foundation of a traditionally Greek musical festival by a local benefactor named Demosthenes provides useful information.7 Though the money for all the festivities and their accoutrements came from Demosthenes, the person who would administer this festival and legacy was the agōnothetēs. This agōnothetēs, the first of which was the nephew of Demosthenes, who was then succeeded by elite men elected from the boulē (lines 17–20, 30), had the privilege of wearing a purple robe and taking a seat at the front of the theatre not only during the festival performances, but also during all meetings of the boulē and the popular assembly, which obviously met in the theatre also (lines 51–3, 56–8). This great honour was extended not just to the current office holder, but to every man who would later serve, suggesting that the first row of the theatre at Oenoanda would be marked by the presence of a slowly but consistently expanding group (line 88). The agōnothetēs was also to wear a gold crown bearing the busts of Apollo and the emperor Hadrian, who personally approved the endowment of the festival (line 56). In Ephesus, the governing bodies of the city similarly granted Salutaris the right to wear a gold crown in the theatre, marking him out in such a way – via the shimmering gold on his head – that visually connected him to the very icons that he had paid for and that were set up on bases in the cavea. Nor was the honour of a gold crown unique to these instances. As Angelos Chaniotis has noted, elite citizens all over the Greek East were distinguished in the theatre by their costume (often a combination of crowns and purple robes and, in one case, a silver robe that dazzled the eyes), as well as by their entrance into the theatre, and by verbal acclamations for their contributions to the city (Chaniotis 2007: 62–4; Parker 1999: 168–70). This was part of a tradition that was already widespread in the Hellenistic period, and which, like the reliance on private benefaction itself, continued unabated in the Roman Empire. As mentioned above, Dio Chrysostom, in the first century ce, identified the right to wear a gold crown in the theatre as still one of the highest honours that a city could bestow. Chaniotis imagines that these elites would perhaps ‘have their small share in glory when they stand up in the theatre, dressed in their best clothes, to be for a few minutes the centre of attention’ (Chaniotis 2007: 58). The visual devices for gaining the crowd’s attention engaged internal representations of the elevated social position held by those who had the right to the special costumes and oral expressions of honour and support. That these practices were both widespread and durable must be an attestation of their effectiveness. This focus on individual excellence and the  7

Published in Wörrle 1988. Mitchell 1990 provides an English translation of this dossier.

126

dian a y. ng

social pressure exerted on wealthy citizens by the community is integral to Greek urban culture and, as has been argued, especially important from the Hellenistic period on into the Roman period as an outlet for elite expressions of social prestige when true political power resided in dynastic or imperial capitals and their administrative representatives. The purple robes and gold crowns functioned as parts of an external symbolic system that was effective because, first, they were tied to prevailing notions about public honour and, second, people were skilled in interpreting these visual cues. The high cost of purple and gold and the iconographic ­­association between crowns and victory and virtue made these objects meaningful. Rather than just being ‘best clothes’, these garments and adornments communicated social approval and excellence in service to the community. Though the details of the particular offices or benefactions of each notable citizen meriting such recognition were not stored in these artefacts, they nevertheless made immediately accessible the fundamental information about the superiority or prestige of certain citizens.

Portrait Statuary as Cognitive Scaffold for Individual Identity At the same time, individual elite identity had long been situated in this architectural form in the Roman world. The stone theatres of Rome are all named for the famous and powerful: Pompey, Balbus, Marcellus. In Asia Minor, the theatres were not named for great men, but they were often the product of elite monetary contribution. The theatre at Aphrodisias was first constructed in the first century ce by an Aphrodisian freedman of Augustus named Zoilos, whose connections to the imperial house – reflected in the iconography of the sculptural decoration – were vital to the prosperity of the town (Reynolds 1991: 15–16; Smith and Erim 1991). In many theatres, such as at Perge and Nysa, the names of those responsible for the construction, decoration or renovation of different parts of the theatre were carved into the stone, to stand as a permanent testament of their generosity (for Perge, see Şahin 2000: 201–8; for Nysa, see Bean 1980: 182). Salutaris in Ephesus had the record of his gift carved into the parodos, or entrance passageway, of the theatre, so that the task of recalling his specific act of generosity from those of others of his station could, on the one hand, be offloaded and, on the other hand, serve as a backdrop to the civic procession that he had endowed. While the inscriptions, when present, and the attendance of notable citizens in eye-catching dress were scaffolding for the internal knowledge of their generosity and careers, the disconnection of this information from their bodily presence that would allow for its retrieval and retention after their deaths was made possible by the external representation of the public honorific portrait. These monuments of personal identity and public virtue were set up all over the Roman Empire, jostling for position in the public areas with the highest traffic (Stewart 2003: 91, 136–40; Fejfer 2008: 51–63; Tanner 2000: 25–6). The theatre, being the place of popular assembly and the location of an imposing stage building to which vision was directed and in or



r oman -per iod t h eat res 127

near which the portrait statues were installed, afforded a viewing context for these artefacts that was unmatched by other sites. The publication of Aphrodisian portrait statuary has provided valuable insight into elite portrait statues in a theatrical setting (Smith et al. 2006: 19–28). Found in association with the stage building at the theatre were a number of statue bases, inscribed with the name of the citizen who was honoured with the statue, and in many cases with additional information, such as the offices held by this person, other personal or public virtues, specific reasons for the honour of a statue, by whom the statue was granted, and who would be responsible for its maintenance (Smith et al. 2006: 19–28, 75–97). For example, from the actual stage of the theatre came two second-century ce statue bases that were associated with portraits for one Attalos Adrastos, the gymnasiarchos and stephanēphoros of the city, who was honoured by the boulē, the dēmos, the gerousia and the neoi for leaving almost all his money for sacrifices, feasts and to his offices (Smith et al. 2006: 82 nos. 71, 72). Another second- or third-century base from the stage of the theatre supported the statue of Ti. Cl. Diogenes, the high priest of Asia, sebastophantēs, agōnothetēs, euergetēs, nomothetēs, gymnasiarchos, a man who was philanthrōpos – loving of people – and philopolitēs – loving of (fellow) citizens, honoured by the boulē and the dēmos (Smith et al. 2006: 83 no. 89). The list of offices and virtues, displayed under a single portrait, reads almost like the multiple roles that an actor in Shakespeare’s Globe would assume over the course of a play. These inscribed statue bases, with their lists of different offices along with a variety of abstract virtues, presented a highly condensed and selective biography of an elite’s public life (Stewart 2003: 168–9). In the theatre, both the statue and its accompanying inscription facilitated cognitive tasks of recall and of sorting and comparison between members of the entire class of wealthy benefactors and civic leaders active within the city in any one period and over the history of a city. As elite life, especially in the eastern provinces, was to a great degree defined by competition for public recognition through certain defined avenues of pursuit,8 it was vital to provide the broad citizenry with the means to tell one notable public servant and patron from another. The acts of comparison were important in the physical context of the theatre, for it was during the assemblies of the citizens that they would vote to bestow the statues and other honours on deserving elites. The backdrop of honorific portraits was not only external storage of information, but also drove discussion of the behaviours and activities that would merit a public statue. These very questions – how the honour of a public portrait motivated elite benefactions and what a city owes to someone deemed worthy of this monument – are central to Dio Chrysostom’s oration to the Rhodians (Or. 31). Not only does Dio frame the granting of a public portrait in exchange for elite service in contractual terms, but he also alludes to the cognitive importance of the statue and its inscribed base, condemning the re-inscription of a  8

For euergetism in the Greek East, see, for example, Mitchell 1993: 206–11; Zuiderhoek 2009. Ng 2015 considers the evidence for whether elite benefactions of architecture versus spectacles were understood as having a commemorative purpose.

128

dian a y. ng

base so that a statue could be recycled to honour another individual both because the contract between elite and city would have been breached and because the person originally honoured with the statue could fade from public awareness with the erasure of the identifying information from the statue base (Or. 31.81, 90–2). The inscribed statue base thus served as an important scaffold for the memory of individuals and their generosity, of the community’s approbation and therefore of the rightfulness of the elite benefactor’s elevated social status and prestige.9

Mythological Friezes as External Representations of Historical Narrative Not only was the cavea of the theatre the place where the cognition of social structure and civic organisation was situated, but the relief sculpture decoration of some stage buildings of Imperial-period theatres arguably also partly carried the cognitive burden necessary for the reception and memory of key elements supporting the civic identity of Imperial-period communities. Since we lack specifically ­­relevant textual references to these friezes of the kind that exist for the types of cognitive artefacts previously discussed, the analysis of these sculptures is an attempt to understand the visual depictions of highly locally relevant myths in the gathering place of the community as facilitating the comprehension and memory of evolving historical narratives and political relationships. In addition to generic imagery of theatrical masks and ornamental carving, some theatres in Asia Minor featured extensive reliefs that highlighted narratives of legendary history or cultic traditions that distinguished a city from its regional peers. The stage building of the theatre at Nysa on the border of Lydia and Caria, dated to 180–200 ce, featured a series of sculpted panels on the lowest podium level (Kadioĝlu 2006: 162; Lindner 1994: 192; Newby 2003: 202). These reliefs include scenes of Persephone’s marriage to Hades, from her abduction on the plains of Nysa to the pursuit by her distraught mother Demeter, as well as scenes related to the rearing of the infant Dionysus by the nymphs of Nysa as he was hidden away from the wrathful Hera. As central events of these myths occurred in the area of the city itself, they were considered part of local history.10 Furthermore, the abduction and subsequent marriage of Persephone and Hades had a religious significance to Nysa, as a cult to Kore and Pluto had been active around a Plutonium about three kilometres from Nysa since at least the third century bce (SEG 4.412; Welles 1934: 54–60 no. 9; Bagnall and Derow 1981: 31 no. 51). The depictions of these mythical and religious events at the theatre were also matched with references to contemporary religious rituals and performance, in a scene that has been interpreted as depicting a revelry  9

Ng (forthcoming) analyses Roman public portraits as external representations, and considers how reiterative, periodic interactions with these statues served as memory retrievals that strengthened the retention of information about those honoured with a public portrait. 10 See Lindner 1994 for the publication of these reliefs. Newby 2003: 202–12 interprets certain iconographic elements differently from Lindner 1994.



r oman -per iod t h eat res 129

associated with Dionysiac worship – in which a Roman soldier appears to participate (Newby 2003: 209) – as well as in one that seems to match Strabo’s description of the offering of sacrifices at the Plutonium (Lindner 1994: 195–6; Newby 2003: 209; Strabo 14.1.44). Though the podium reliefs of the stage building are consequently chronologically illogical, nevertheless they provide a selective account of Nysa’s civic identity as a formative location in the biographies of important deities. Though Persephone’s abduction on the plains of Nysa is related in poetry, the need for Nysans or foreign visitors to recall the text or the details of the story is negated by the visual representation before their eyes. The combination of scenes from mythic and heroic time with those representing contemporary cultic practice also acts as an abbreviated explanation of why Pluto was locally worshipped in a particular way – remarkable enough to be mentioned by Strabo in his Geography. The addition of these relief panels to the theatre during the Severan period, during which local cults became politically important, suggests that certain narratives were gaining renewed significance in Nysa and warranted historically and religiously specific cognitive artefacts in the venue of public assembly. A similar way of uniting religious practice with local history is found in reliefs from the theatre at Hierapolis in Phrygia, installed around the same time – early third century ce – as those at Nysa. Though the urban theatre at Hierapolis was first built in the first century ce under the Flavians, it was during the reign of Septimius Severus that the stage building received a substantial renovation, paid for by the corporation of purple-dyers in Hierapolis. This renovation added to the stage building the relief decoration on the podium and across the top of the porta regia, or central doorway of the stage building (D’Andria 2003: 150–4; Ritti 1985: 108–13). The podium reliefs at Hierapolis are similar to those at Nysa in their depiction of two mythological cycles, and in the combination of mythology and cultic activity. Radiating from the centre of the stage outwards, the podium reliefs depict the biographical cycles of Apollo, the patron deity of Hierapolis, and his twin, Artemis, beginning with their births and continuing through famous solo and joint exploits related to Asia Minor, such as the musical contest between Apollo and Marsyas and the slaughter of the Niobids. These mythological scenes transition into scenes of worship and ritual, as Apollo is shown presiding over a ritual purification dance by maidens and Artemis is transformed from the huntress into the cult image of the Ephesian Artemis, to whom offerings are being made by cult personnel.11 The excavators at Hierapolis suggest also that one of the cultic rituals – that of the purification dance for Apollo – depicted on the podium may be a specific reference to the plague brought back from the Near East by the armies of Marcus Aurelius in the second century ce (D’Andria and Ritti 1985: 76–8). That Hierapolis was afflicted by this plague, and sought the help of Apollo Kareios and of the oracle of Klaros, is documented in an inscription displayed at the Temple of Apollo located just to the south of the theatre. 11

The Apollo and Artemis reliefs from the podium of the theatre at Hierapolis are fully published in D’Andria and Ritti 1985. See also Newby 2003: 196; Price 2005: 121.

130

dian a y. ng

Further representations of recent events are found above the porta regia in a frieze commemorating the celebration of the Pythian Games granted to Hierapolis by Septimius Severus, the main panel of which shows the imperial family in the process of making an offering.12 Although the Severan family is surrounded by allegorical figures of places, virtues and even of Aion, eternity, the scene is very much tied to the specific time during which the stage building was reconstructed and when Hierapolis was given the right to host prestigious international games. In addition, portraits of Hadrian and perhaps other emperors (Bejor 1991: 1–22; Gualandi 1977; Ritti 1983), as well as the clipeate portraits of the Pergamene kings Attalos II and Eumenes, were also originally situated somewhere in the scaenae frons (Ritti 1985: 120; see also Verzone 1978). The theatre at Hierapolis is, in a sense, a visual archive of the city’s history, from its origin – represented by the tribal names carved in the cavea alluding to the Hellenistic-period Seleucid founders (Price 2005: 124) – to the present, and in this archive, the events associated with real people and events are joined to the worship of Apollo. The succession of Hellenistic dynasties, the passage of the city from Greek into Roman authority, and the highs and lows of recent Hierapolitan history are given in condensed and easily retrievable and digestible form. The citizens’ internal representations of all or part of these events, the centuries of local history and its fruitful relationship with the reigning sovereign would have been complemented by these exograms. The decoration of the stage building, in conjunction with the division of the seating sections, might perhaps be analogous in function to the early modern theatrical plot in providing only the necessary information, spatially organised so as to make clear one’s place in the larger group of participants and overall narrative. This fundamental grounding in civic history and contemporary politics having been established, Hierapolitans were well prepared to absorb further developments in later years. In contrast, the theatre at Aphrodisias features an actual epigraphic archive. In the third century ce, a large body of correspondence between Aphrodisias and Roman emperors was inscribed on the outside walls of the theatre (Reynolds 1982: 33–7). The archive includes letters dating back to Octavian, whom the theatre’s original patron Zoilos had served, and records the granting of certain privileges to Aphrodisias, such as the free status of the city and its exemption from levies, billeting and other obligations by Octavian and the senate of Rome, as well as the subsequent confirmation of these rights by successive emperors through to the reign of Gordian III in the mid-third century ce. In addition, subscripts from letters of emperors to other cities concerning the rights of Aphrodisias are also inscribed in the so-called Archive Wall (Reynolds 1982: 37).13 The creation of the Archive Wall centuries into the existence of the city is, on the one hand, an obviously symbolic gesture meant to highlight the imperially favoured history of a 12

These reliefs are published in Ritti 1985: 59–77 and Newby 2005: 249–51. See Ng 2016: 237–40 for the increasing importance and depictions of local cults and iconography related to imperially granted international games in the Severan period in Asia Minor. 13 The documents of the Archive Wall are nos. 6–21 in Reynolds 1982: 38–135.



r oman -per iod t h eat res 131

smaller provincial town. On the other hand, the carefully curated collection of diplomatic correspondence renders a large amount of information spanning centuries of time quickly retrievable by people whose regular cognitive burden surely would not have tolerated such a load. Because of the function of the theatre as a gathering place for the popular assembly, the viewers of the Archive Wall would have been socially conditioned to think of the building as a place for politics and diplomacy in Aphrodisias’ interest. The documents of the Archive Wall, just as with the sculptural archive at Hierapolis and Nysa, allowed the people in the theatre ‘to get more direct cognitive access to, and better retention of’, the ongoing historical developments within the context of a preferred version of civic history (Sutton 2007b: 23). Thus, it creates a baseline from which future debate on public business could be conducted and diplomatic relationships could be advanced or amended.

Conclusion This chapter has demonstrated that the Roman-period theatre presented a unique cognitive ecology, in which history and society both shaped and were given shape by an array of cognitive artefacts that – by virtue of its purpose as a place of frequent meetings and spectatorship – were coupled with ancient citizens in a consistent, if not constant, fashion. The attendees of the theatre, whether they be locals or visitors to a community, possessed or came to possess a procedural memory for the correct way of dressing, of reading the costumes of others, of finding their rightful seat and of interpreting the sculptures before them. This working memory of how to live in Roman or culturally Greek cities was not confined to the theatre, of course; it reached into all aspects of daily life and was variously reinforced. Yet, being in the physical realm of the theatre allowed huge and abstract concepts like class membership, elite obligations to the community and civic history to come to the fore in an intersecting way that was unlikely to occur elsewhere because the theatre was also the place for the deliberations and votes that perpetuated social systems and history. The cognitive ecology of the theatre thus accomplished much more than supporting retention of information and meaning, but created the ideal conditions by which to process any new people, events or circumstances that might be introduced. In sum, it contributed to a flexibility of mind. Though this chapter has focused on the architectural and decorative aspects of the theatre so as to demonstrate the usefulness of distributed cognition in the study of Roman-period architectural spaces and visual culture, the cognitive ecology of this space encompassed civic rituals and spectacles that were themselves cognitive artefacts. Future research on the Roman-period theatre in historical cognitive science would be rewarded by investigation of ephemeral, yet reiterative, cognitive scaffolding and thereby move closer to a consideration of the non-dramatic theatrical experience in its totality.

132

george ka z ant z i di s

8 Co g n i t i o n , E mo t i o n s a n d the Feeling Bod y  in the H i ppo cra t i c C orp u s George Kazantzidis

A long-standing question in affective sciences concerns the exact relationship between ‘appraisal’1 and what are usually identified as the bodily aspects of emotion. In contrast with earlier approaches, especially in the field of psychology, which proposed that appraisal should be treated as a disembodied cognitive phenomenon,2 there has subsequently been a tendency to implicate the body more actively: emphasis on bodily differentiation has increasingly helped to associate emotions with distinctive physiological profiles,3 while it is now acknowledged that changes in the state of one’s non-neural body can influence the way we think as well as emotion experience more generally.4 Even so, as Giovanna Colombetti observes, today’s appraisal theory still maintains that the cognitive component of emotion ‘remains’ largely ‘in the head, separate from the body’; consequently, the bodily aspects of emotion, though regarded as ‘instrumentally related’ to it, continue, in essence, to be seen as ‘extrinsic to the appraising process’: the body is at best ‘a means to the appraising process, not a part of it’.5 The aim of this chapter is to explore the place assigned to emotions in the Hippocratic Corpus (fifth to fourth century bce) by drawing attention to the remarkable fact that, despite the many detailed theories regarding the seat and function of

I would like to thank Prof. Douglas Cairns for inviting me to participate in the ‘History of Distributed Cognition’ project at the University of Edinburgh and for his invaluable comments during the editing of this chapter.  1 For recent discussions of the role of appraisal in emotion, see Ellsworth and Scherer 2003 and Scherer and Ellsworth 2009; see also Frijda 1993 and Thagard and Nerb 2002.  2 As Lazarus (1982: 1019) puts it, ‘cognitive activity is a necessary as well as sufficient condition for emotion’. See especially Arnold 1960 and Lazarus 1966, 1991. Solomon (1976) equates emotions with ‘judgements’ and Greenspan (1988) says they are ‘thoughts’; cf. Frijda 1986 and Scherer 1993.  3 Kreibig 2010. On the ‘specificity-debate’, concerning the question of whether various emotions can be distinguished by unique patterns of somatovisceral arousal, see Ekman et al. 1983; Cacioppo et al. 1993; Panksepp 1998.  4 See Gibbs 2005: 239–74 and, more recently, Colombetti 2014 on the notion of the ‘feeling body’.  5 Colombetti 2014: 96. See Hurley 1998.



cogn ition, emot ions a nd t h e f eel i ng b o dy 133

the mind in classical medicine,6 we find no systematic account of what we would identify today, in an affective context, as an ‘appraising process’. When discussed by medical writers, emotions such as fear, sadness, surprise and happiness tend to be almost exclusively identified with their associated bodily symptoms;7 once we look beyond these symptoms, there is no ‘mind-stuff’ left behind out of which the emotion can be constituted (or, at least, doctors express no interest in exploring it in detail). This model of thought, as I propose to show, comes in contrast with Aristotle’s cognitivist approach in the Rhetoric. According to Aristotle, while the capacity to experience a range of emotions is innate in our species, the actual experience of any given emotion is informed by the judgements and beliefs that engage it, giving it its ‘aboutness’;8 emotions, πάθη, in this interpretation should be understood to mean ‘all those things on account of which people change and differ in their judgements (μεταβάλλοντες διαφέρουσι πρὸς τὰς κρίσεις), and upon which pain and pleasure follow’ (Rh. 1378a19–23). This is not to say of course that Aristotle disregards the role of the body. The Rhetoric’s reference to pleasure and pain (1378a20–1) is sometimes reformulated in terms of ‘disturbance’ (tarachē, 1382a21, 1383b12–13, 1386b23) and the pleasures or pains in question can also be qualified as ‘sensory’, i.e. depending on bodily perception (Ethica Eudemia 1220b14); pleasure and pain thus involve the phenomenology of emotion as well as the element of valence.9 In De anima 403a17–20, too, it is explicitly stated that emotions cannot take place unless ‘through the body’,10 μετὰ σώματος, and they do not belong to the soul alone (this serves as a confirmation that the soul does not maintain an independent existence, since what is affecting it affects also the body). But even when seen as somatic phenomena, emotions are always elicited by phantasia (a capacity that itself also depends on the body and its powers of perception: De an. 427b14–429a9; cf. 403a8–10)11 of states of affairs in the world, and they involve complicated mental processing: as Aristotle famously puts it in De anima 403a25, emotions are ‘reasonings set in matter’/‘enmattered principles’, logoi enuloi, that is, they necessarily consist of both the body and cognition. By contrast, Hippocratic emotions, whether seen as specifically localised in a certain part of the body or diffused throughout it, appear to be generated automatically, without any evaluative judgement occurring between the perceived stimulus and the actual physical response: the focus remains always on how  6

See Gundert 2000. It is worth noting, in this respect, that a number of words which in contemporary, non-medical literature are employed metaphorically to describe emotional states (for example, tromos and phrikē) are primarily conceived in the Hippocratic Corpus as referring to conditions of the body; cf. Cairns 2013, 2015.  8 See Konstan 2006; cf. Nussbaum 2001. For a recent general discussion, see Pearson 2014.  9 Cf. below, at n. 70. 10 Contrast the fact that noein in Aristotle sometimes is the entelecheia of nothing bodily; see Van der Eijk 2000: 69–70; cf. Kahn 1992: 375–9; Johansen 2012: 221ff.; Miller 2012: 310ff. See also Barnes 1971–2. This does not, however, affect the element of cognition in emotion, since emotions are always ‘with body’. 11 On phantasia as a defining feature of the Aristotelian theory of emotions, see Striker 1996: 291; cf. Pearson 2014: 166–7.  7

134

george ka z ant z i di s

a certain emotion is determined by the body’s chemistry (its humours, temperature and so on) rather than by any kind of cognitive intervention. On the whole, this chapter aims to highlight the fact that the Hippocratics are not interested in discussing how emotions are cognitively shaped; this question emerges in a different epistemological environment as a ‘philosophical’ rather than strictly ‘medical’ issue. At the same time, their pronounced emphasis on the body as essentially a sense-making mechanism which absorbs lived experience and filters an individual’s interaction with his or her environment shares something of enactivism’s tendency to ‘decentralise’ consciousness and to tie it intimately with embodied activity, not only within the brain, but also in the full body in action.12

Emotions and the Body in the Hippocratic Corpus Hippocratic writers address the question of the mind’s location in the body in remarkably different ways.13 Some assume that intelligence (phronēsis, sunesis) is contained in the blood14 while others propose the heart as responsible for cognition (gnomē).15 In other, more complicated versions, the power to think (dianoia) is attributed not to fixed organs in the body but to elemental qualities, defined as ‘fire’ and ‘water’, which act in combination and, depending on how they are mixed with each other, determine a person’s skills, such as acuteness, a good memory, precision of the senses or even proneness to certain emotions.16 In one of the oldest treatises in the Corpus, composed in the last quarter of the fifth century bce, cognitive faculties are directly linked to the brain. As we read in De morbo sacro 17 (6.386 L.), it is with the brain that we humans perceive, think and experience an emotion: Εἰδέναι δὲ χρὴ τοὺς ἀνθρώπους, ὅτι ἐξ οὐδενὸς ἡμῖν αἱ ἡδοναὶ γίνονται καὶ εὐφροσύναι καὶ γέλωτες καὶ παιδιαὶ ἢ ἐντεῦθεν [ἐκ τοῡ ἐγκεφάλου], καὶ λῦπαι καὶ ἀνίαι καὶ δυσφροσύναι καὶ κλαυθμοί. καὶ τούτῳ φρονέομεν μάλιστα καὶ βλέπομεν καὶ ἀκούομεν καὶ διαγινώσκομεν τά τε αἰσχρὰ καὶ καλὰ καὶ κακὰ καὶ ἀγαθὰ καὶ ἡδέα καὶ ἀηδέα, τὰ μὲν νόμῳ διακρίνοντες, τὰ δὲ τῷ συμφέροντι αἰσθανόμενοι. Men ought to know that from the brain, and from the brain only, arise our pleasure, joys, laughter and amusement, as well as our sorrows, pains, griefs, and tears. It is especially through this organ that we think, see, hear, and distinguish the ugly from the beautiful, the bad from the good, the pleasant from the unpleasant. Sometimes we judge according to convention; at other times according to what we sense to be the best for our interests.17 12

Clark 2008b. For some general discussions, see Gundert 2000 and Van der Eijk 2005: 124–31; cf. Singer 1992. 14 See Flat. 14 (6.110 L.) and Morb. 1.30 (6.200 L.). 15 Cord. 10 (9.88 L.). See Langholf 1990: 40–6, 50–1. 16 Vict. 1.35 (6.512–14 L.). See Jouanna 2007; Bartos 2015: 248–9; Enache 2015. 17 Translation in Jones 1923b: 175 (modified at points). 13



cogn ition, emot ions a nd t h e f eel i ng b o dy 135

On the face of it, the text establishes an intimate relationship between cognition and the emotions. It is worth noting, for instance, that before linking the brain to thought and judgement, the author mentions it first in association with a number of basic affects18 which clearly present it as lying at the centre of a person’s emotional life. What we seem to be missing, though, is a clear explanation of how exactly thoughts and ideas influence emotion (appraisal, in this sense, is mostly highlighted in the text in connection with moral and aesthetic judgements). Furthermore, the fact that emotion and cognition share a common material substrate should not necessarily indicate that in order to experience an emotion one has also to ‘think’, that is, to form an evaluative judgement, in the process. In fact, the author himself seems to be drawing a fine line between the two when it comes to distinguishing between different aetiologies for cognitive and emotional dysfunction.19 Thus, we are told that intelligence and perception are impaired when the brain shifts its position (kineisthai) because of excessive moisture. As a result the patient starts ‘seeing and hearing’ strange things and becomes delirious for so long as the brain does not remain still (Morb. sacr. 17 = 6.388 L.). On the other hand, emotional disturbance is said to follow the brain’s corruption (metastasis) by bile or phlegm: when, for instance, too much phlegm enters the cavity of the head and affects it with excessive coldness, the patient is seized by sudden fits of ‘causeless distress and anxiety’ (ἀνιᾶται δὲ καὶ ἀσᾶται παρὰ καιρὸν) because the brain ‘has been contracted contrary to custom’ (Morb. sacr. 18, 6.388 L.). This neat distinction20 allows us, in essence, to posit that an emotion can, on occasion, be generated automatically when a certain part of the body has been affected in a specific way – and while the rest of that part (responsible for intelligence) remains largely intact.21 Emotions, according to Antonio Damasio, can be triggered and executed biologically at a non-conscious level; being part of a largely innate and automated 18

Modern lists of ‘basic affects’ (Tomkins 1962: 243–4) or ‘basic emotions’ (Ekman and Friesen 1969; Damasio 2003: 44–6; Panksepp 2005) typically include: anger, disgust, fear, happiness, sadness and surprise; for an expansion of this list in Ekman’s more recent writings, see Ekman 1992, 1999; Ekman and Cordaro 2011. In the Greek passage cited above, we can easily recognise the equivalents of happiness (ἡδοναὶ, εὐφροσύναι, γέλωτες) and sadness (λῦπαι, ἀνίαι, δυσφροσύναι, κλαυθμοί); cf. Thumiger 2016. Furthermore, while disgust is not mentioned explicitly in the text, we can detect it behind the author’s statement that the brain helps us distinguish (διαγινώσκομεν) between pleasant (ἡδέα) and unpleasant (ἀηδέα) things; on ἀηδία and disgust in the Hippocratic Corpus, see Kazantzidis 2017. If this is the case, disgust would then appear to be more openly connected – than, say, happiness or sadness – with what we would identify as an ‘appraising process’. On disgust as a higher cognitive emotion, see Miller 1997: 10, with reference to Smith and Ellsworth 1985. 19 See Gundert 2000: 22, 26. 20 Cf. Vict. 1.35–6: both mood and intelligence are dependent on the state of the soul, conceived as a mixture of fire and water. While the soul’s intelligence (and sanity), however, is a matter of proportion between dry and moist elements in the body, its disposition – which determines, for instance, whether a person will be ‘irascible’ (ὀξύθυμος), ‘quarrelsome’ (δυσμενής) or ‘easy going’ (εὔνους) – is linked to its ‘movement’ (Vict. 1.36, 6.522–24 L.); see Gundert 2000: 23–4, with Jouanna 2007. 21 As Aristotle would agree: see De an. 403a19–24 (cited below, n. 30), Rh. 1379a10–22.

136

george ka z ant z i di s

life-regulation system, they can bypass conscious thought.22 One of Damasio’s main arguments is based on observations that electrical stimulation of certain areas in the brain can cause laughter accompanied by a sensation of merriment and mirth; what is more, though this mirthful feeling is induced artificially, the brain consequently generates cognitive-perceptual incongruities to try to account for these emotional experiences: each time a subject laughs, due to electrical stimulation, she then attempts to cognitively attribute her laughter with reference to various stimuli in her environment.23 This hypothesis can be extended to pathological instances as well – and it may be appropriate to include them in our discussion, considering that most of our evidence from the Hippocratic Corpus concerns emotions in a pathological setting, that is, as side effects of things happening inside the body and disrupting its balance. As Damasio notes, in this context, neurological damage in the brain forces a patient to behave emotionally in strange ways: Patient C. would burst into the most arresting crying or the most spectacular laughter for no perceptible cause. Not only would the motive of the outburst be inapparent, but its emotional value might be diametrically opposed to the affective tenor of the moment . . . The outbursts might follow each other in quick succession, leaving C. barely time enough to take a breath and say that he was not in control, that neither laughter nor crying were really meant as such, that no thoughts in his mind justified such strange behaviour.24 The same kind of reasoning is present in the Hippocratic text I have been discussing: significantly, the ‘distress’ and ‘anxiety’ which result from the brain’s corruption with phlegm are said to occur παρὰ καιρόν, literally ‘at the wrong time’, which means that they lack correspondence with the external environment and, rather than being shaped by evaluative judgements, they are exclusively defined by the body’s chemistry at that precise moment. Similar observations can be found in De victu 1.35 (6.518–20 L.): when the balance between ‘fire’ and ‘water’ in the soul (conceived in that case as a material entity)25 is disturbed and the element of coldness prevails, patients tend ‘to cry for no reason, they fear what is not frightening and they grieve for things that are not appropriate’ (κλαίουσί τε οὐδενὸς ἕνεκα δεδίασί τε τὰ μὴ φοβερὰ λυπέονταί τε ἐπὶ τοῖσι μὴ προσήκουσιν).26 These and other passages27 illustrate how a missing link between a person’s emotions 22

Damasio 2003: 27–80. See especially Fried et al. 1998; cf. Arroyo et al. 1993; Shammi and Stuss 1999; Wild et al. 2003. 24 Damasio 2003: 77–8. See also Parvizi et al. 2001; cf. Marshall and Zimbardo 1979 and Bejjani et al. 1999. 25 See Van der Eijk 2005: 128. 26 See Jouanna 2013: 101. 27 See, for example, Coac. 4 (5.588 L.): ἐκ καταψύξιος φόβος καὶ ἀθυμίη ἄλογος, ‘excessive coldness causes unreasonable fear and sadness’. Cf. Ps.-Aristotle, Pr. 30.1, 954b35. On alternating fits of laughter and crying as symptoms of emotional imbalance with an organic cause, see Epid. 3.3.17 (3.142 L.). 23



cogn ition, emot ions a nd t h e f eel i ng b o dy 137

and what Damasio calls ‘the affective tenor of the moment’ can be evaluated as a pathological sign, indicating that something is wrong with the body.28 At the same time, the fact that an emotion can occur ‘inappropriately’ without an identifiable stimulus (and while the patient remains ‘sane’ in other respects)29 quite simply shows that its cause can be purely organic. In other words, it is possible for someone to feel an emotion because some mechanism in the body has been triggered which is producing a certain affective reaction. This reasoning does not apply to ‘pathological’ emotions only, that is, to emotions that make their appearance as side effects in the course of disease. As Giovanna Colombetti notes, in everyday life too, ‘There seem to be many occasions in which one experiences one’s body in specific affective ways, irrespective of the fact that one appraises one’s context as irritating, conducive to one’s goals, funny, and so on.’ Thus, sometimes, we may ‘wake up feeling edgy and irritable before [we] can even wonder why; often in the early afternoon, irrespective of the situation, [we] feel distracted and unmotivated, including a sense of heaviness and slowness’.30 These cases demonstrate that bodily arousal does not have to be cognitively attributed or explained to be experienced as an emotion; occasionally, the way we feel can simply be a matter of our bodily state, without the intervention of conscious thoughts and judgements pointing to a specific (mental) cause. While the brain is mentioned in De morbo sacro as the ‘source’ of emotion, the part of the body which is best endowed with ‘feeling’ (αἰσθάνεσθαι), that is, where an emotion manifests itself physically, is the heart (καρδίη) and the diaphragm (φρένες).31 Every part of the body, according to the text, participates in intelligence in proportion to the air it receives from the brain; however, since the best quality of air stays in the head, what is left for the rest of the body is by definition inferior. What is more, the heart and the diaphragm, because of their thin and solid structure, are said to receive no air at all, and they are literally devoid of the material which distributes ‘intelligence’ across the body (τῆς μέντοι φρονήσιος οὐδετέρῳ μέτεστιν).32 As a result, their special link to emotions is explained in purely mechanical terms: due to the fact that veins (φλέβες) extend from all over See Aph. 6.23 (4.568 L.): ‘if fear or sadness persist for a long period of time (πολὺν χρόνον), this indicates melancholia’, literally, an excess of black bile in the body. Reference is made here to unnaturally prolonged, negative emotions which are either disproportionate or unrelated to external circumstances; see Horwitz and Wakefield 2007: 58 and Radden 2009: 183–4. 29 See, for example, Epid. 7.1.11 (5.384 L.), where a female patient is sane and can recognise everyone but displays unbalanced (παρὰ καιρόν) emotional behaviour manifesting itself in outbursts of anger. 30 Colombetti 2014: 92. Cf. Aristotle, De an. 403a19–21: the fact that emotions have a physiological basis ‘is shown by the fact that sometimes we are not at all irritated or frightened, even though severe and obvious troubles befall us, while at other times we are stirred by small and insignificant things, when the body is swollen and is in the same state as when we are angry’. See also Ps.Aristotle, Pr. 30.1, 954b16–18, with Schütrumpf 2015: 359. 31 Morb. sacr. 19–20 (6.390–4 L.). 32 See Gundert 2000: 21–2, 29. Cf. Pl. Ti. 77b5–6, with Price 2010: 128, 134. 28

138

george ka z ant z i di s

the body to the area of the chest, the latter becomes extremely sensitive33 and it ‘convulses’ (σπᾱται) when someone ‘is physically hurt, feels tension (τάσις) or becomes excessively joyful’. It is extremely important to notice at this point how, in designating what is essentially a reflex-like type of reaction, the medical author equates the way in which an emotion registers in the body with a response to a purely somatic event – the pain one feels because of disease or a wound that has been inflicted on the body. What allows him to do this seems to be the fact that there is no expressed commitment on his part to the idea that emotions have to be shaped by judgements and be cognitively inflected. Just as the body is instinctively strained (συντείνεσθαι) when reacting to pain (ἀνιώμενον),34 in the same way the heart is automatically receptive to emotion because it happens to be located in a part of the body that is more sensitive than the others. This part has the special power to move and ‘feel’ (αἰσθάνεταί τε μάλιστα) but retains no capacity whatsoever to cognise. αἰσθάνεσθαι has been used earlier in the text (ch. 13, 6.384–6 L.) to describe the sensations and reactions of inanimate substances: everything, including ceramic pots and the tissues of the human body, senses and reacts to the south wind.35 Similarly, in our text the heart’s ‘sensing’ of distress or happiness is seen to be executed at the level of movement; while the heart has no intelligence, it can still enjoy a privileged relationship with emotion precisely because of its aptitude to ‘convulse’ (σπᾱται). Damasio’s observations are once again pertinent at this point. In stressing that simple emotional responses are present in all living organisms and reflect a system’s adaptivity and concern for itself, he identifies ‘movement’ as a basic requirement for affectivity.36 When a simple, brainless creature instinctively moves away from something that it senses as posing a danger, it is orchestrating a kind of movement that mimics, and essentially amounts to, emotions of fear and distress. Accordingly, when a living organism is automatically attracted to a source that maintains its well-being, this act of approximation is no different from the process of emotion that we humans identify as happiness. On a larger scale, as Raymond Gibbs observes, ‘emotions arise as we become displaced and dislocated to another position in adaptive response to some situation. One way of characterising the felt dimension of emotional experience is in terms of “affective space”, or the space we move through as we experience distinct emotions.’37 Movement is thus seen as an essential requirement for emotion (especially by those adopting an enactivist stance),38 and this seems to apply also to the heart in our medical text: 33

See also Morb. sacr. 6 (6.370 L.); cf. Oss. 19 (9.194–6 L.) and VM 22 (1.632–4 L.). Cf. the use of the verb in ch. 18, where it indicates mental distress (ἀνιᾶται . . . παρὰ καιρὸν). For the blending of physical with mental pain, see also Epid. 6.6.5 (6.316 L.). 35 See Gundert 2000: 29 n. 97; cf. VM 15 (1.606 L.) with Jouanna 1990: 38 n. 5. 36 See especially Damasio 1999: Part II (‘Feeling and Knowing’); cf. Damasio 2003, 2010: ch. 2. Cf. Gibbs 2005: 243–5 on emotions as ‘felt movements’; on the notion of ‘primordial affectivity’, see Colombetti 2014: 20–4. 37 Gibbs 2005: 244. For the notion of affective space, see Cataldi 1996. 38 For the emphasis placed by enactivism on sensorimotor contingencies, and the way in which movement affords a more nuanced perception of the surrounding environment, see O’Regan and 34



cogn ition, emot ions a nd t h e f eel i ng b o dy 139

what establishes the special link between emotion and this particular organ is not its high degree of intelligence (as Aristotle would be inclined to suggest)39 but the latter’s ability to ‘convulse’ and to react to (physical) pain instinctively. In this light, it should be seen as no coincidence that, while introducing us to the area of the diaphragm as a location that lacks in ‘thought’ and ‘intelligence’ but is especially endowed with feeling, the medical author mentions examples of both ‘sadness’ and ‘happiness’ as occurring specifically ‘in excess’ but also ‘unexpectedly’ (ἐξ ἀδοκήτου).40 What provides the basis for discussion, in other words, seem to be intense emotional reactions to sudden events, presumably in line with the author’s ensuing attempt to suggest that one can experience an emotion physically without having the time to evaluate its (mental) cause properly. As William James notes in his monumental article ‘What is an emotion?’ (1884: 193–4), If we fancy some strong emotion, and then try to abstract from our consciousness of it all the feelings of its characteristic bodily symptoms, we find we have nothing left behind, no ‘mind-stuff’ out of which the emotion can be constituted . . . What kind of emotion of fear would be left, if the feelings neither of quickened heart-beats nor of shallow breathing, neither of trembling lips nor weakened limbs, neither of goose-flesh nor of visceral stirrings, were present, is quite impossible to think. The reasoning outlined in this passage fits nicely with the way in which the Hippocratics usually speak of emotions, that is, by leaving aside the appraising ­­process – which (in theory, at least) shapes them – and by focusing directly on how an affective reaction manifests itself physically. Thus, for example, one medical writer defines ‘anger’ (ὀξυθυμίη) simply by referring to the fact that it ‘contracts the heart and the lungs and draws the hot and the moist substances into the head’; likewise, a state of contentment corresponding with ‘happiness’ (εὐθυμίη) is exclusively associated with the heart’s ‘relaxation’ and its lack of (physical) tension.41 One of the reasons that allow medical writers to ignore, as it were, the cognitive shaping of emotion can be traced in their tendency to systematically highlight cases in which the external cause of emotion is said to be taking place suddenly, leaving no time for conscious thought and evaluation.42 David Konstan is rather cautious about accepting that reactions of this kind actually fall into the category of emotion: The response to a loud noise, sometimes labelled the startle reflex, or to a looming presence may be generated in an automatic way; the latter reaction Noë 2001a; cf. Thompson 2007b: 221, connecting locomotion and perception with emotion and feeling. 39 See Knuuttila 2004: 34. 40 Morb. sacr. 20 (6.392 L.). 41 Epid. 6.6.5 (6.316 L.). On this text, see Pigeaud 1981: 446 and 1991: 19. 42 For example, Morb. sacr. 13 (6.378–80) attributes an excessive flux of phlegm and the chilling of the body to fear which is induced ἐξ ἀδήλου, ‘suddenly’, ‘unexpectedly’, ‘with no prior warning’.

140

george ka z ant z i di s

r­esembles fear, but perhaps it is better described as analogous to fear, like an animal’s or even a human being’s instinctive response to a predator. We must be wary of classifying such automatisms as emotions.43 To this line of thought one may of course juxtapose the idea of emotion as ‘embodied appraisal’. As Jesse Prinz points out, ‘in developing a theory of emotion, we should not feel compelled to supplement embodied states with meaningful thoughts: we should instead put meaning into our bodies and let perceptions of the heart reveal our situation in the world.’44 According to this interpretation, it would be inaccurate to try and discriminate between appraisal (as a purely intellectual act) and the bodily feelings that occur in emotion experience;45 what is more, rather than assume that bodily sensations are experienced as mere responses to the appraisal, and lack evaluative character, it may be preferable to treat them as the means though which we experience something as dangerous, scary, offensive, enjoyable and so on – in other words, as sense-making mechanisms which enact appropriate reactions aimed at maintaining our well-being. Thus, even in situations where an emotion is elicited rapidly and there is no time to consciously evaluate, for example, an object of threat, it would not be correct to characterise our experience of fear as merely a reflex; the bodily feelings that may occur in this experience do not feel like mere reactions ‘lacking world orientation; rather, they are felt as part of the experience of appraising the situation in a certain way’.46 Hippocratic writers display the tendency to approach emotions at the level of what Konstan dismisses as ‘elementary automatisms’. In doing this, they leave aside any explanation of how cognitive intervention works and move directly to physical symptoms, almost as if an emotion is something that can be sufficiently explained and accounted for when one focuses on its visible impact on the body. As we read in De humoribus 9 (5.490 L.), in a section of the text that is devoted to ‘things happening to the body’ (οἷα τὰ σώματα) when we experience an emotion: οἷα τὰ σώματα, μύλης μὲν τριφθείσης πρὸς ἑωυτὴν, ὀδόντες ᾑμώδησαν, παρά τε κοῖλον παριόντι σκέλεα τρέμει, ὅταν τε τῇσι χερσί τις, ὧν μὴ δεῖται, αἴρῃ, αὗται τρέμουσιν, ὄφις ἐξαίφνης ὀφθεὶς χλωρότητα ἐποίησεν. Οἱ φόβοι, αἰσχύνη, λύπη, ἡδονὴ, ὀργὴ, τἄλλα τοιαῦτα, οὕτως ὑπακούει ἑκάστῳ τὸ προσῆκον τοῦ σώματος τῇ πρήξει, ἐν τούτοισιν ἱδρῶτες, καρδίης παλμὸς, καὶ τὰ τοιαῦτα τῶν δυναμίων. How the body behaves: when a mill grinds the teeth are set on edge; the legs shake when one walks beside a precipice; the hands shake when one lifts a heavy load; 43

Konstan 2006: 25–6. See, however, the extensive bibliography on the affective modulation of the startle reflex as a basic example of ‘emotional priming’, especially Lang 1995; Lang et al. 1998; Bradley and Lang 2000; cf. Hamm et al. 2003: 198–201. 44 Prinz 2004a: 58. 45 See Frijda 1993. 46 Colombetti 2014: 109; cf. Robinson 1995.



cogn ition, emot ions a nd t h e f eel i ng b o dy 141 the sudden sight of a snake causes pallor. Fear, shame, pain, pleasure, passion and so forth; to each of these the appropriate member of the body responds by its action. Instances are sweat, palpitations of the heart and so forth.47

What is perhaps the most intriguing aspect of this passage (but also the most difficult one to interpret) is the author’s close association of bodily reactions which emerge as a result of specific emotions (for example, trembling with fear while walking by a precipice or turning pale at the sudden sight of a snake) with the image of one’s ‘hands shaking under the burden of a heavy load’.48 I would suggest that the insertion at this point of a purely somatic response as analogous to bodily feelings associated with emotion is designed precisely to highlight the fact that even in cases of an affective experience the body can act, more or less, autonomously, without being necessarily shaped by specific thoughts and judgements – in the same way that when our hands tremble with pain, this happens not because we have evaluated the situation as physically exhausting, but simply because our body gets tired and reacts instinctively. The emphasis placed on the autonomy of the ‘acting’ body (τοῦ σώματος τῇ πρήξει),49 and the implicit attempt to downplay the mind’s ruling function, would also seem to explain the author’s choice to discuss fear by invoking the example of a snake which is ‘seen suddenly’, ἐξαίφνης ὀφθείς: the external stimulus in this case leaves little time for conscious thought and evaluation, causing the body to ‘turn pale’ (χλωρότητα ἐποίησεν) before one is able to assess fully what one has seen as well as appraise the specific threat posed by it.50 This particular example is cited as a favourite among neuroscientists who deny that emotions have to be cognitively formed and maintain that some of our emotional responses are mediated by direct pathways from areas in the brain associated with perception to other areas which are linked primarily with affect.51 To quote Prinz again: Perceptual centers are in the business of recognizing perceivable features such as shapes. The amygdala is in the business of instructing other brain areas to initiate bodily responses when certain perceptual features are perceived. Often there is no need for an intervening judgement in this process. Neither perceptual centers nor the amygdala harbor the resources to reflect on something as lofty as a demeaning offense. It would even be generous to say that the amygdala 47

Translation in Jones 1931: 81. See Overwien 2014: 215: ‘die Bedeutung dieser Passage ist unklar’; cf. Pigeaud 1981: 43. 49 Cf. Vict. 4.86 (6.640 L.), where similarly the notion of an acting body (πρήξεσι παντὸς τοῡ σώματος) presupposes that all psychic activity has been previously subordinated to it. 50 Before mentioning what happens to the body when we experience an emotion (οἷα τὰ σώματα), the author speaks of ‘accidents grieving the mind, gnōmē, either through vision or hearing’, τὰ ἀπὸ συγκυρίης λυπήματα γνώμης, ἢ διὰ τῶν ὀμμάτων, ἢ διὰ τῆς ἀκοῆς. The use of ἀπὸ συγκυρίης is by no means coincidental. The phrase can be translated as ‘accidentally’, ‘by chance’ or even ‘suddenly’, and it is clearly designed to downplay the role of the cognitive faculty: what we deal with here are instances of automatic perception rather than thorough mental processing. 51 See Öhman et al. 2001; cf. LeDoux 1996. 48

142

george ka z ant z i di s

­ arbors the concept danger. No concepts are necessary at all when the amygdala h sets one’s heart racing (via the hypothalamus) after one sees a snake.52 Priority here, as also in the Hippocratic text, is given to bodily processes enacted at an automatic, non-conscious level.53 Instead of a thought or judgement, it is the body’s chemistry (conceived as neural and chemical commands in the brain with minimal computational requirements or an inner state of coldness that makes someone turn pale) which defines how we appraise (physically) an external situation. In this model, the body is not a mere interacting object, nor is it extrinsic to the process of appraisal, but rather it becomes itself a vehicle of significance, participating actively in the construction of affect-laden stimuli. Affective responses, in other words, ‘assist in seeing an object as what it is from the very moment that visual stimulation begins’,54 and they do not merely occur after the object has been identified. Closer study of the way that emotions are recorded as pathological symptoms in the Hippocratic Corpus confirms that they are conceived as being intrinsically linked to the human body. In general, medical writers do not distinguish between (what we would call) psychological symptoms and purely somatic conditions.55 This is owing to the fact that what is normally considered the cause of a somatic event accounts simultaneously for the emotion. This should explain, for instance, why in most of the cases where fear is mentioned as a symptom the patient is said to be affected with excessive coldness,56 almost as if it were impossible for someone to experience an emotion without having previously entered a specific bodily state. Accordingly, a pathological emotion ceases to be present only after its somatic cause has subsided:57 when bile flows in excessive quantities into the head, as we read in De morbo sacro 18 (6.388 L.), the patient is seized by fear which disappears only when the humour has moved down to the body and into the veins. This set of ideas, as we have seen above, is often combined with the observation that some of these emotions seem ‘unreasonable’ (ἄλογος), in the sense that, though having an identifiable internal cause, they cannot be cognitively attributed or explained.58 Rather than evaluate them as indications of how a patient perceives, understands and reacts to the world around him, doctors are thus inclined to use them more as information about things taking place inside the human body, such as changes in one’s humoural constitution or temperature or a vital organ’s failure to function ­­ 52

Prinz 2002: 140. Cf. Prinz 2004b: 74–5. See also Robinson 2010: 652–3. On unconscious elicitation of emotion, see the recent studies of Öhman and Flykt 2000; Berridge 2003; Berridge and Winkielman 2003; Winkielman et al. 2011. 54 Barrett and Bar 2009: 1326–7; cf. Colombetti 2014: 101–6. 55 See Van der Eijk 2005: 131, with reference to Epid. 1.18 (2.652 L.); for a thorough analysis, see Singer 1992. I discuss these issues in detail in Kazantzidis (forthcoming). 56 See, for example, Coac. 473 (5.690 L.): φόβος μετὰ καταψύξιος. 57 φόβος παρέστηκε μέχρις ἀπέλθῃ πάλιν ἐπὶ τὰς φλέβας καὶ τὸ σῶμα: ἔπειτα πέπαυται, ‘fear persists up to the point that the phlegm returns to the veins and the body; then it stops’. 58 See, for example, Coac. 4 (5.588 L.): ἐκ καταψύξιος φόβος καὶ ἀθυμίη ἄλογος. 53



cogn ition, emot ions a nd t h e f eel i ng b o dy 143

properly. Similarly, as regards therapy, when a doctor appeals to the need to regulate a patient’s emotions, this is not meant to have a psychological effect. In Epidemics 2.4.4 (5.126 L.), we read that it is mainly for the sake of ‘restoring humoural balance’ (ἐγχύμωσις) and colour (ἀνάληψις χρώματος) in the patient that one should be willing occasionally to induce in him feelings such as anger, happiness or fear. The reasoning behind this is that each of these emotions has its own, specific impact on the body and can thus contribute to physical treatment: for example, inducing fear, a ‘cold’ emotion, would have helped counteract and reduce a patient’s high temperature; conversely, if a patient is freezing, making him angry would have raised his temperature. A clarification is due at this point: the passage does not seem to imply that emotions can be endogenously produced; on the contrary, we have to assume that in producing, for example, fear, the doctor will need to create a scenario which typically functions as an exogenous elicitor of the emotion.59 Consequently, while doctors focus on the organic, somatic aspects of emotion, their background assumption is still that emotions are responses to external phenomena. However, the fact remains that in the context of therapy the handling of a patient’s emotional state is not meant to have a curative effect on a psychological level: manipulating the subject’s affective condition is not a way of penetrating into his mind and ensuring his psychological well-being, but it is rather conceived as the means to reach into the interior of his material body. Hippocratic medicine views the human body primarily as a collection of fluids (χυμοί),60 not organs, which are dynamically linked with each other and determine, at any given moment, a person’s character and disposition.61 Even in cases where a Hippocratic text appears to deal with a patient’s psychology independently of the body, there is always the assumption that emotional incidents which warrant medical intervention should eventually be explained in terms of a somatic cause.62 One such exceptional instance is found in Epidemics 3.17.11 (3.134 L.), where a woman, who is said to be ‘grumpy’ by nature (δυσάνιος γυνὴ), falls physically ill with acute fever ‘following an incident of grief’ (ἐκ λύπης μετὰ προφάσιος).63 While the text clearly gives priority to a psychological event, which subsequently takes a physical form, Galen insists on reading it as yet another clinical case whose origin should be attributed to humoural imbalance. Building presumably on parallels from other gynaecological texts which connect fear and sadness to the presence of black bile in the uterus,64 he explains δυσάνιος not as the patient’s psychological trait 59

Although what scenario precisely that would be remains subject to speculation. In light of the frequent references to ‘sudden’ stimuli of emotion in the Hippocratic Corpus (e.g. Morb. sacr. 20: ἐξ ἀδοκήτου; Hum. 9: ἐξαίφνης), one would have imagined the doctor producing fear by ‘startling’ the patient. 60 See Nat. hom. 4 (6.38–40 L.), with King 2013b. 61 See Gundert 2000: 18–19. 62 See, for example, the case of Nicanor in Epid. 7.86 (5.444 L.), with King 2013a: 267–8; cf. Ps.Aristotle, Pr. 30.1, 955a5–8. 63 See Thumiger 2015: 124. 64 See, for example, Mul. 2.182 (8.364 L.).

144

george ka z ant z i di s

(which is how the Hippocratic text intends it) but as a state of mind that is induced because too much ‘melancholic menstrual blood’ (μελαγχολικὰ καταμήνια) has been trapped in the woman’s body. In doing so, he even expresses his amazement (θαυμάζω γε μήν) at the fact that the detail of the original physical cause has been omitted in the Hippocratic passage.65 Galen’s reading at this point is only consistent with what we find in the majority of Hippocratic discussions of emotions: it falls in line with the tendency to downplay the idea that emotions form themselves as mental reactions to external events and stresses the need to understand them as direct consequences of things happening inside the body.

How Enactive Is the Hippocratic Body? The emphasis placed by the Hippocratics on the biological machinery of emotions could be seen as forming the background for Aristotle’s observations in De anima 403a29–b1, which highlights the differences between a ‘physicist’66 and a ‘dialectician’ in their treatment of the ‘affections of the soul’ (πάθη): Διαφερόντως δ’ ἂν ὁρίσαιντο ὁ φυσικὸς [τε] καὶ ὁ διαλεκτικὸς ἕκαστον αὐτῶν, οἷον ὀργὴ τί ἐστιν· ὁ μὲν γὰρ ὄρεξιν ἀντιλυπήσεως ἤ τι τοιοῦτον, ὁ δὲ ζέσιν τοῦ περὶ καρδίαν αἵματος καὶ θερμοῦ. Τούτων δὲ ὁ μὲν τὴν ὕλην ἀποδίδωσιν, ὁ δὲ τὸ εἶδος καὶ τὸν λόγον. Hence a physicist would define an affection of the soul differently from a dialectician; the latter would define, for example, anger as the appetite for returning pain for pain, or something like that, while the former would define it as a boiling of the blood or warm substance surrounding the heart. The one assigns the material conditions, the other the form or account.67 Aristotle stresses here a simple fact, namely, that whether one places emphasis on how an emotion is cognitively processed or, alternatively, on the emotion’s bodily impact and physical manifestation is simply a matter of perspective and different epistemological priorities. Accordingly, it would be fair to say that although doctors tend to concentrate on how the body reacts in emotion experience, this by no means entails that they are ignorant of the fact that such an emotion has been previously elicited on a mental level. As we have seen at the beginning of the text, De morbo sacro 17 (6.386) explicitly states that all emotions derive from the brain (ἐκ τοῡ ἐγκεφάλου) – although it should be stressed again that we are still Galen, In Hipp. lib. III Epidem. Comment. 17a 778 K.: Διότι καταμηνίων ἐπίσχεσις ἦν μελαγχολικωτέρων, ἐφ’ οἷς εἰκότως ἡ γυνὴ δυσάνιος ἐγένετο, ‘for there was clearly a retention of the menses, because of which the woman, naturally, became grumpy’. 66 On the close association between ἰατρός and φυσικός, see Arist. Resp. 480b22–4. 67 Translation in Barnes 1984: i. 643. For some discussions of this passage, see Frede 1992: 103–4; Van der Eijk 2000: 66–9; Knuuttila 2004: 32–4; Leunissen 2010: 51–3. For a general discussion of Aristotle’s theory of emotion, see Cooper 1996.

65



cogn ition, emot ions a nd t h e f eel i ng b o dy 145

missing an ­­explanation of how this works precisely, in terms of the value systems, judgements and beliefs which are implicated in the whole process. In a unique and exceptional passage (by Hippocratic standards at least), we are even told that ‘the mind, by itself, γνώμης ξύννοια, distinct from the bodily organs or events, χωρὶς τῶν ὀργάνων καὶ τῶν πρηγμάτων, can feel misery and joy, fear and optimism, hope and despair’ (Epid. 6.8.10, 5.348 L.), for example through recollection of past events or through anticipation of the future.68 In sum, although the Hippocratics choose to deal with what Aristotle calls the ‘material conditions’ (ὕλη) of emotions, this does not mean that they lose sight of the fact that mental processing is also needed, one way or another, as a contributing factor and formal element of affective experience. That said, it would be a mistake to assume that the ‘tension’ outlined in Aristotle’s passage above can simply be resolved by saying that doctors prefer to speak about the body, whereas philosophers are more interested in exploring the soul. The two models are construed on different sets of priorities which effectively lead us to look at the body differently – as regards its relation to the ‘mind’ but also its potential to function as a sense-making mechanism in its own right. If we look back at Aristotle’s passage on anger, the ‘cause and effect’ relationship is not made explicit. As Philip van der Eijk points out, Aristotle does not give a clear answer to the question whether the boiling of hot material around the heart is causative of the emotion anger, or the result of it, or a physical manifestation of it . . . It is of course possible to interpret a statement such as ‘fear causes shivering’ in the sense that ‘fear’ and ‘shivering’ are mental and physical aspects of two different states, of which the one leads to the other. The fact that such a statement contains two incomplete descriptions of these two states might then be accounted for by considerations of relevance. But there are no explicit remarks by Aristotle to this effect.69 Still, whereas in De anima 403a29–b1 things remain obscure (the boiling of blood can be as primary in the experience of anger as the desire to retaliate is), there are plenty of other occasions, especially in his works on ethics and rhetoric, where Aristotle inclines towards a cognitivist approach. As we read, for example, in Rhetoric 1378a19–23, ‘under the category of emotion, we should understand all those things through which people change and differ on account of their judgements (μεταβάλλοντες διαφέρουσι πρὸς τὰς κρίσεις), and upon which attend pain and pleasure (οἷς ἕπεται λύπη καὶ ἡδονή), for example anger, pity, fear, and all Smith (1994: 267) translates the difficult phrase γνώμης ξύννοια as ‘the mind’s consciousness’. The passage does not seem to suggest that the mind is not enmattered; rather, it is concerned with instances in which intellectual activity, irrespective of the external circumstances, can generate, by itself, scenarios which have an affective impact. 69 Van der Eijk 2000: 66, 67 n. 30. On the difficult question of how Aristotle thinks that bodily ­processes are related to ‘mental’ processes, see Burnyeat 1992 and Sorabji 1992. 68

146

george ka z ant z i di s

other such things and their opposites’.70 What Aristotle actually says in this passage is that emotions are those conditions of the soul that affect a person’s overall outlook, hence the emphasis placed on the distinctive ‘mental verdicts’ (κρίσεις) involved in them; furthermore, assuming that one could conceive of emotions as ‘representational states of the soul’, it then follows, according to Jamie Dow, that the pain or pleasure which ‘attend upon’ them can be the ‘pain or pleasure at the circumstances thus represented’. Be that as it may, emotions in Aristotle are individuated by reference to their characteristic beliefs. One cannot describe the pain that is peculiar to anger, for instance, without saying that it is pain at ‘a perceived injustice’;71 neither can we be specific about the pleasure associated with this emotion unless we explain it as deriving from the expectation, on which we dwell ‘in our minds’ (dianoia), to avenge the slight that has been suffered.72 In Hippocratic medicine, as I have tried to show, with the exception of a few cases, the cognitive constituent of emotions remains of relatively little significance: emotions are either identified with their physical manifestations (Epid. 6.6.5, 6.316 L.), or they are shown to enjoy a privileged relationship with parts of the body devoid of intelligence (Morb. sacr. 19–20, 6.390–4 L.), or, finally, when an external stimulus is mentioned, it is systematically placed in contexts which leave little time for conscious thought and deliberation (Hum. 9, 5.490 L.; Morb. sacr. 13, 6.378–80 L.). Rather than explore the way in which cognition intervenes and helps in shaping an emotion, the Hippocratics tend to illustrate how ‘bodily changes follow directly the perception of the exciting fact’ and how, essentially, ‘our feeling of the same changes as they occur is the emotion’.73 Accordingly, it is no coincidence that while patients are very often observed in the Hippocratic Corpus as exhibiting strange emotional behaviour as a side effect of their disease, it is extremely rare to find a doctor advising that one should attempt to address these emotions directly during treatment.74 The main reason ‘psychotherapy’ is almost entirely absent from the Hippocratic writings has precisely to do with the doctors’ conviction that a pathological emotional state cannot be dealt with unless one finds out and treats its physical cause.75 At this point, we need to ask whether such a conception of the emotions, as predominantly physiological events which are not necessarily attributed to a central executive system of higher cognition, helps to establish a line of connection with current notions of the ‘enactive’ body. Decentralising consciousness and distributing it across the body requires two basic steps: on the one hand, as Giovanna Colombetti (2014) argues, we need to start by rejecting the idea that appraisal 70

As Konstan (2006: 33) points out, the definition is not tailored here to a forensic context but reflects Aristotle’s ideas on how emotions work in general. For a detailed discussion of this passage, see Dow 2015: 133–44. 71 See Eth. Nic. 1135b25–9. 72 See Rh. 1378b1–9. See Nussbaum 1994: 88; cf. Nussbaum 1996. 73 James 1981: 1065. 74 See Holmes 2010: 199–200. 75 See Gundert 2000: 25 with Entralgo 1970: 341–2.



cogn ition, emot ions a nd t h e f eel i ng b o dy 147

‘remains largely in the head’; on the other hand, it is essential that we see the body as actively participating in the formation of emotion experience: not only that emotions cannot be abstracted from bodily symptoms, but that these bodily feelings provide the lens through which we assess the world around us and react emotionally to external events: Acknowledging [that] all our affective life is bound up with . . . evaluations . . . need not lead to the further claim that appraisal exists as a separate process or mechanism that triggers emotional responses. Indeed . . . neuroscientific considerations as well as phenomenological ones suggest otherwise and indicate rather that evaluating the world and responding emotionally to it are not distinct processes. From an enactive perspective, the process of appraising is best characterized as an organismic activity, not separate but overlapping with what are usually seen as noncognitive, bodily components of emotion.76 The idea of a brain (or a ‘ruling part’ of the body) orchestrating emotion experience is similarly absent from the Hippocratic texts. At the same time, and perhaps more importantly, the doctors’ systematic identification of emotions with bodily feelings, in contexts which seem to suggest that conscious thought is being bypassed, promotes the concept of an affective body that is given a certain autonomy and can act prior to (thus defining) the way we think about things or incidents which we consider to have an emotional impact on us. In this context, it is worth stressing that in some medical texts bodily parts are even said to ‘feel’ an emotion themselves: thus, blood can contribute to a patient’s fear when it is excessively chilled77 but, as we read at De flatibus 8 (6.100 L.), it can also feel fear for itself (τὸ αἷμα φοβεόμενον), when too much ‘cold air’ has entered the body and causes the blood to ‘leap up’ (καθαλλομένου) and move to warmer places. This striking image (which is reminiscent of patients behaving in the same way under mental duress)78 rests on the assumption that human bodily parts can act independently, and that, essentially, emotions can be fully formed on a micro-level and across the body without being necessarily defined by an appraising process.

Conclusion In her seminal 2000 article, entitled ‘Soma and psyche in Hippocratic medicine’, Beate Gundert observes that:

76

Colombetti 2014: 111–12. Coac. 4 (5.588 L.). 78 Compare, for example, Virg. 3 (8.468 L.), where corruption of the blood causes ‘frightening hallucinations’ (φοβερὰ ὀνομάζει) and forces female patients to commit suicide by ‘jumping’ (ἅλλεσθαι) into wells (κελεύουσιν καὶ καταπίπτειν ἐς τὰ φρέατα). See Lami 2007: 46–9. Cf. Pl. Symp. 186c5– e3, where Eryximachus attributes emotional agency to individual parts of the body. 77

148

george ka z ant z i di s

There are several accounts of sense perception in the Hippocratic Collection. Hearing, for example, is explained by the resounding quality of the empty and hollow space of the inner ear, and the dry, hard bone and membrane surrounding it. Seeing is attributed to the reflecting quality of the thin, clear membranes of the eye, which are supplied with moisture from the brain. Olfaction occurs on account of the brain’s moistness, as it perceives smells that it draws up together with air. Although most of these accounts mention a connection between the particular sense-organ and the brain, the role of the brain in perception seems undecided . . . [With only few exceptions] these accounts, then, . . . give a physical account [my emphasis] of perception.79 By contrast with some philosophical accounts of the body, which present the latter as remaining relatively inactive under the control exerted by one ruling part (be it the ‘soul’ or the ‘mind’),80 the ‘Hippocratic body’ is by definition a more complex and dynamic mechanism, where different parts play significant roles in contributing to functions such as consciousness and perception.81 As Gundert succinctly illustrates, rather than look for a central executive system, medical writers prefer to think in terms of a ‘perceiving body’ whose connection to the brain (or any other organ that happens to be assigned with the power of intelligence and cognition) is not always clearly established: while hearing, for instance, can be fully performed and executed as a sense in the area of the ear, this is predominantly attributed to the organ’s physical aptitude to receive sounds and it is not explained with direct reference to how it connects to the brain. I would suggest that the Hippocratic concept of what we call ‘emotion’ falls in the same picture. So far I have been discussing emotions as being connected with parts or functions of the body that are not directly linked to intelligence; at the same time, however, the image which is emerging is that of ‘conscious’ organs on a different level: the heart, for instance, is described in De morbo sacro 19–20 (6.390–4 L.) as devoid of φρόνησις, but it can still ‘sense’ (αἰσθάνεσθαι) – and it can sense better than any other organ (αἰσθάνεταί τε μάλιστα) the physical vibrations through which emotions materialise. Similarly, the body in De humoribus 9 (5.490 L.) is presented as ‘responding’ with its appropriate part to each of the emotions, but again it is said to do so through ‘action’ (τῇ πρήξει) – which means that it assumes a role of its own. In this context, the Hippocratic tendency to speak of emotions as being elicited by ‘sudden’ external events (ἐξ ἀδοκήτου, Morb. sacr. 20, 6.392 L.; ἐξαίφνης, Hum. 9, 5.490 L.) facilitates the emphasis placed on bodily arousal as well as on the subsequent suggestion that much of our emotional self turns out to be in some sense built on a substrate 79

Gundert 2000: 20–1. Cf. Lo Presti 2007. See, for example, Epicharm. DK B12: ‘it is the mind that sees and hears things; everything else is deaf and blind.’ Similarly, Cic. Tusc. 1.20.46 attributes to ‘the philosophers’, in general, the idea that the body cannot sense anything (‘neque est enim ullus sensus in corpore’) unless through its connection to the mind. Cf. Sext. Emp. Math. 7.129–30, with reference to Heraclitus. 81 See Lo Presti 2016, with Gundert 1992. 80



cogn ition, emot ions a nd t h e f eel i ng b o dy 149

of embodied capacities. A ‘dynamical perspective on emotional expression’, as Raymond Gibbs (2005: 273) points out, dissolves the traditional divide between behaviors that are automatically generated and those that are intentionally produced. Both kinds of expressions arise as emergent products of self-organizing processes that constrain the degrees to which various emotions are felt, expressed, and acted on. As such, emotions cut across brain, body, and world, and are thus neither purely mental nor purely physiological phenomena. In sum, ideas about emotions in the Hippocratic Corpus help to highlight what Andy Clark (2008b), in his ‘extended mind’ hypothesis, describes as ‘meshing’: rather than see the brain as that which realises all that matters in human existence, we should rather be thinking in terms of an ongoing intermingling of cognitive activity with the perceptual and motor matrixes within which memory, emotion, language and all other aspects of life are meshed. ‘Human cognizing’, in this light, ‘includes inextricable tangles of feedback, feed-forward, and feed-around loops: loops that promiscuously criss-cross the boundaries of brain, body and world.’82 Emotions in Hippocratic medicine can be seen as performing this role: they contribute to the construction of an affective, sensing body that can act independently of a ‘ruling part’, and often functions as the primary seat out of which meaning, in the way we feel and interact emotionally with the world around us, is materialised.

82

Clark 2008b: xxviii.

150

c hr ist opher g i l l

9 E n a ct i v i s m a n d E mbod ied C og nition in S t o i ci s m a n d P la to’s Tim aeus Christopher Gill

The aim of this chapter is to highlight points of resemblance between certain ancient theories (Stoicism and that of Plato’s Timaeus) and a modern theory, enactivism, that falls within the group of contemporary theories of cognition sometimes characterised as ‘distributed’. Before turning to this comparison, certain questions about terminology need to be addressed. These modern theories focus on ‘cognition’, a widely used term in the contemporary philosophy of mind and neuroscience. A key question for these theories is whether cognition is a function restricted to the brain or more broadly embodied (distributed over brain and bodily p ­­ rocesses) or extended still further (over brain, body and environment). ‘Cognition’, in this body of theory, seems to be understood quite broadly, as including functions relating to understanding of all kinds (including perception and belief-formation), and also executive functions such as decision-making. ‘Cognition’ does not standardly include emotion or affect (though some theorists argue that the two functions are interlinked); motivation might be classed either as cognition or separately.1 How does this terminology match up with that of ancient Greek theory? The main relevant ancient category is ‘psychology’, taken to embrace the functions of the psyche (a term meaning ‘mind’, ‘soul’ or ‘life’, but with no single modern equivalent). These functions were very broadly conceived, as including belief, reasoning, knowing, emotion, desire, motivation and sometimes more basic functions such as perception and movement or even growth and nutrition (all the functions of living creatures, in fact). The standard ancient contrast was between psyche and body; it was a matter of debate which psychological functions were embodied. Psychological functions were sometimes subdivided between rational ones (typically, belief, reason and knowledge) and non-rational ones (typically, emotion and desire).2 ‘Cognition’, in the modern theoretical sense, does not obviously map directly on to this framework; the closest equivalent might seem to be  1

See e.g. Colombetti 2014: 18–19, who argues for a closer link between cognition and emotion than is normal in cognitive science.  2 See e.g. Everson 1991: 1–4; also Annas 1992: 3–10, on physicalism and mind-body relations in ancient philosophy of mind.



ena ctiv ism an d emb od i ed co g ni t i o n 151

‘rational’ functions (belief, reason and knowledge, also deliberation and choice), but this equivalent is only approximate. Any attempt to chart points of resemblance and contrast between ancient and modern theories needs to take account of this difference. To address this problem, I proceed here in the following way. I outline the relevant aspects of the ancient and modern theories (what I see as potentially similar or analogous features) in the terminology in which they are framed, that is, as relating to ‘cognition’, on the one hand, or to different ‘psychological’ functions, on the other. I also describe the relevant functions and how they are interrelated. On this basis, I aim to identify salient points of resemblance and difference between certain ancient theories and modern theories of distributed cognition, especially enactivism. Obviously, the conceptual and historical background of the terminological difference I have just noted could be explored in its own right.3 Also, the points of resemblance highlighted could be examined much more exhaustively. But these are both tasks for other occasions. As already indicated, this chapter aims to explore the scope for fruitful comparison between certain ideas in Stoicism and Plato’s Timaeus and the modern theories of enactivism and embodied cognition more broadly. I begin by exploring Stoic analogues for a view found in some versions of enactivism, namely that the psychological activities of animals, including human beings, form an integral part of their self-sustaining response to their environment. The principal Stoic theories considered are those concerning natural teleology, animal (including human) development, and the roles of ‘impression’ and ‘motive’ in their psychological thinking. Modern enactivism holds that cognitive processes and emotions (seen as closely linked) are fundamentally embodied. I consider how far we can find an analogous view in Stoic thinking. Gauging whether these points of resemblance also apply to Greek philosophical thinking as a whole is not feasible in a discussion of this scale. However, I provide a context for the relevant features of Stoic thought by considering certain partly analogous ideas in Plato’s Timaeus, a work often seen as influential on Stoicism. In the Timaeus, these similarities with modern ideas are qualified by a dualistic approach to psychology, and this fact helps to highlight the more holistic character of Stoic thinking in this area. The second theme of this chapter also figures in both Stoicism and the Timaeus. This is the idea that the universe as a whole offers an expression of rational thought which human beings should take as paradigmatic for their own psychological life and aspirations. The idea that the universe is a locus of cognition does not figure within the set of modern theories which are central for this volume and the project on which it is based, although there are some parallels in contemporary thought noted later. However, this ancient idea is worth considering here, partly because of the linkage with the other Stoic and Platonic themes discussed and partly because, like the theories of extended and distributed cognition, it poses a  3

For some brief comments on this larger question, see the start of the section ‘Stoicism and Embodiment’.

152

c hr ist opher g i l l

challenge to the standard modern assumption that cognition is a purely human psychological function.

Stoicism and Enactivism The starting point for this discussion is constituted by certain distinctive ideas in Stoic philosophy that can be seen as comparable with key themes in modern enactivism. In his outline of enactivism, Dave Ward identifies two main strands. One is ‘a biologically-inspired approach to understanding cognition, according to which cognition is a dynamic relationship between organism and environment that aims at sustaining the organism’s biological viability’. Ward highlights the notions of autopoiesis (self-construction), adaptivity and sense-making as central to this relationship. The second strand ‘understands cognition – in particular visual perception – in terms of a skilful grasp of the patterns linking sensory stimulation and movement’. The first strand is more important for the comparison with Stoic ideas, but the second is suggestive for interpreting Stoic ideas about animal development.4 The question of embodiment is discussed more fully shortly. However, it is useful to note that enactivism embraces an embodied approach to cognition.5 So, in so far as Stoic thinking also involves an embodied approach, this is relevant for its relationship to enactivism. It is also worth highlighting two themes linked by Shaun Gallagher with the ‘biological’ approach to embodiment. One is that, for this approach, anatomy and bodily movement are salient contributors to brain processes. The second is that cognition is not seen as a separate and independent function but viewed holistically as shaped by, among other factors, proprioceptive and affective processes. A relevant part of the background to these ideas is the reaction, in both these theories (and others), against what is sometimes described as the ‘classic’ modern view that cognition should be understood purely as a function of the human brain, taken largely in isolation, a view sometimes presented as an updated version of Descartes’s mind-body dualism.6 Which Stoic ideas are relevant for the comparison with enactivism? Three seem most obviously relevant: Stoic thinking on natural teleology, on development (understood as ‘appropriation’) and on impression and motive. The first idea is rather widely shared in ancient philosophy (Plato’s Timaeus is another notable example);7 the second two ideas are more specifically Stoic. The Stoics believed that purpose and providential care were in-built in the natural universe. One aspect of this belief was the idea that the bodies of human beings are shaped so as to enable them to fulfil the capacities and activities cognate to them. As explained  4

Available at (last accessed 7 February 2018). Ward cites Di Paolo 2005 on the first point and O’Regan and Noë 2001a on the second.  5 See e.g. Colombetti 2014: chs. 4–6.  6 Available at (last accessed 6 February 2018). See also Varela et al. 1991 and Gallagher 2005.  7 See Sedley 2007, esp. chs. 4, 7.



ena ctiv ism an d emb od i ed co g ni t i o n 153

in texts such as Plato’s Timaeus and Cicero’s On the Nature of the Gods book 2, on Stoic natural teleology, this view embraces virtually all aspects of bodily shape and structure, position of organs and anatomical components (bone, flesh and so on). Also, psychological functions are allocated to specific bodily seats or locations and connections are traced between these locations and other bodily parts.8 (Platonic and Stoic ideas on this topic are examined more fully shortly, in connection with the theme of embodiment.) Why is this idea relevant here? Natural teleology does not play a significant role in current thought and is often seen as being in conflict with modern, post-Darwinian thinking on animal (including human) evolution.9 However, ancient natural teleology promoted two ideas relevant to enactivism. One is the view that psychological processes, including the equivalents of modern ‘cognition’,10 are fundamentally embodied, at least in the sense that they form part of the functions of biological entities which are fundamentally embodied. The other is that ancient teleology, particularly in its Stoic form, formed the basis for the idea that human beings and other animals are naturally disposed to sustain their own identity, one that embraces bodily and psychological functions conceived as fundamentally interconnected. This view forms a background for the second Stoic theory. A highly distinctive (though also influential) feature of Stoicism is the idea that human beings, like other animals, are naturally disposed to carry out two interrelated kinds of development in the course of their lives. On the one hand, they are naturally inclined, from birth onwards, to promote and sustain their own identity or ‘constitution’ (sustasis). They are naturally drawn towards features of the environment that can support this objective, by obtaining ‘the things according to nature’, such as bodily health, property and family and social support. In (adult) human beings, this process is seen as crucially informed by rationality and as leading to advanced rational and ethical skills or virtues, as well as to recognition of the overriding value of the virtues for human life. However, these more advanced features are still seen as an extension of the primary motive to promote one’s own natural identity as a rational animal. The second strand in development starts from the desire to benefit and care for others of one’s kind, a desire seen as in-built in all animals, including human beings. Here, too, the basic desire is seen as leading to more complex and advanced functions in adult humans, including family and communal involvement and the recognition of all other humans as relatives or fellow-citizens, in so far as they share these rational and social capacities. However, these more advanced social functions are still seen as compatible with, and derived from, the more basic animal motive to benefit others of one’s kind.11  8

See Pl. Ti. 44d–47e, 69c–76e; Cic. Nat. D. 2.134–50. See also Burgess 2000; Steel 2001; Johansen 2004: 142–52; Sedley 2007: 120–1, 124–5.  9 However, for the claim that a form of ‘immanent teleology’ is not alien to modern biology, see Weber and Varela 2002; on the related question of value and normativity in (biological) enactivism, see Barrett 2017. 10 See the first three paragraphs of this chapter. 11 On the two strands in ‘appropriation’, see (for the first strand) Long and Sedley 1987 (=  LS),

154

c hr ist opher g i l l

In what respects does the Stoic theory of development prefigure the enactivist approach? We can recognise here analogues for three enactivist themes that express what Ward terms ‘a dynamic relationship between organism and environment that aims at sustaining the organism’s biological viability’, namely, ­­autopoiesis, adaptivity and sense-making.12 The Stoics identified as the core motive in the first strand of animal development the desire to preserve and sustain one’s own identity (or ‘constitution’, sustasis), signifying, in the first instance, the biological identity as human being or some other type of animal. In support of this claim, Stoics pointed to the in-built capacity of animals to deploy their own organs and bodily form to defend themselves against others and to maintain themselves. This process is sometimes presented as a form of adaptivity, at least in the sense that animals need to learn how to use their own bodily form in order to adapt successfully to the environment in which they find themselves. Stoic descriptions of developing animals behaving in this way (for instance, human babies struggling to stand up or tortoises turning themselves the right way up) also evoke what Ward describes as the second main strand in enactivism, namely (the development of) ‘a skilful grasp of patterns linking sensory stimulation and movement’.13 Also implicit in the Stoic account of the first strand of animal development is the idea that this constitutes a progressive process of making sense of their environment. For human beings, for instance, this consists in discerning things that will enable them to maintain and sustain their identity, in the first instance by obtaining what the Stoics describe as ‘the things according to nature’ (such as health and property).14 Although the process of sense-making is more explicit in the first than the second (social) strand of development, the second strand can also be understood as a progressive process of sense-making. The desire to benefit others of one’s kind (paradigmatically, by reproduction and parental love) is presented as a second type of primary desire, which in human beings can be developed further in ways informed by rationality. The two strands of development, taken together, constitute a means by which human beings can progressively make sense of their environment. Also worth noting is the fact that both strands of development are characterised as forms of oikeiōsis, ‘appropriation’ or ‘familiarisation’. Both strands are presented as ways in which animals make sense of their environment by recognising how far it provides things that are ‘their own’ (oikeios), that is, cognate with, or conducive to, their identity. Also, both forms of ‘ownership’ or sense-making are presented as aspects of the broader natural process by which nature as a whole ‘owns’ or ‘appropriates’ its 57A–C, 59D; for the second 57D–H; also Cic. Off. 1.11–14 (both strands), 50–3 (second strand). On the theory (first strand), see Inwood 1985: ch. 6; Gill 2006: 129–66; also 166–77 on some Stoic-influenced versions of the theory; (both strands) Reydams-Schils 2005: ch. 2. On Aristotelian versions of the theory, see Sharples 2010: ch. 17. There has been some scholarly debate about how far the two Stoic strands form part of an integrated theory; I see them as closely linked. 12 See n. 4. 13 See LS 57A–C. For a different view of animal adaptability (linked with a more evolutionary view of natural history), see LS 13I (Epicureanism). 14 See LS 59D (= Cic. Fin. 3.17–22); also refs in n. 11.



ena ctiv ism an d emb od i ed co g ni t i o n 155

component parts, including human beings, by making the inclination to appropriate themselves an in-built part of their identity.15 How do Stoics analyse the psychological dimension of this process? Although Stoics sometimes make use of the conventional Greek philosophical terms for psychological processes, such as perception, belief and desire, they also evolve a distinctive terminology and framework. They use this terminology to characterise the process by which animals make sense of the features of their environment in a way which promotes the maintenance of their identity. One key term is phantasia (literally, what ‘appears’), typically translated as ‘appearance’, ‘impression’ or ‘(re) presentation’. Another is hormē, usually translated as ‘impulse’ but perhaps better rendered as ‘motive’. A striking feature of both terms is their breadth of reference. Phantasia includes functions subdivided by Aristotle into perception, imagination (that is, phantasia in a more restricted sense), thought and belief. Hormē embraces what Aristotle would subdivide into desire (orexis), appetite (epithumia) or wish (boulēsis).16 Both Stoic terms apply equally to non-human and human animals, though not plants. Stoics differentiate human from non-human functions by the claim that (adult) human responses involve rationality, understood as capacities such as those of language use and inferential reasoning.17 The Stoics have sometimes been criticised, in antiquity and in modern scholarship, for restricting rationality to adult humans in this way.18 However, this restriction needs to be set against the inclusiveness of the scope of reference of the notions of phantasia and hormē. In the case of adult human beings, the process of making sense of one’s environment and responding adaptively to this environment is analysed in these terms. Action is presented as depending on ‘assent’ to a ‘rational impression’. More precisely, the assent is to predicates included in the propositions which are the objects of assent. The relevant predicates are terms such as ‘beneficial’, ‘appropriate’ and ‘good’. Motives (hormai) are not formed without this process, but, if assent is given to such predicates, this leads directly to action.19 In the case of non-human animals, motives to action are triggered by non-rational impressions without the need for assent. Although this point is not explicated in our sources, the relevant impressions of non-human animals and children must also involve the pre-rational or sub-rational recognition of things as being beneficial, appropriate or good (or some such notion) in order to produce motivation.20 For instance, we are told that ‘an animal’s first motive is to preserve itself, because nature made it congenial to itself 15

See LS 57A (= Diog. Laert. 7.85); Cic. Fin. 3.16; also Inwood 1985: 184–94; Gill 2006: 37–8. See LS 39 and 53A, 57A, P. On Stoic psychological categories considered in relation to those of Aristotle, see also Inwood 1985: chs. 1–3; Long 1996: 266–75. 17 See LS 53A(5), T–U. Only adult human beings are regarded as rational because the development of rational capacities such as language learning and reasoning is a gradual one; see LS i. 322–3; Gill 2006: 140–1. 18 See e.g. Sorabji 1993: part 1. 19 See LS 33C and I; Epictetus, Discourses 1.18.1. See also Inwood 1985: chs. 2–3; Brennan 2003: 260–9; Gill 2006: 138–43. 20 See Inwood 1985: 75–80. 16

156

c hr ist opher g i l l

from the beginning’ (Diogenes Laertius 7.85), and so it makes sense to suppose that non-human animals, as well as human ones, respond to impressions relevant to this objective. Motives, as a class, are characterised as being ‘towards’ or ‘away from’ something. Typically, they are directed towards things ‘according to nature’, such as health or property, and away from their opposites.21 How far is this type of analysis compatible or comparable with the enactivist approach? Some features of the analysis might seem to point, rather, to a cognitive approach, notably the requirement that, in adult humans, motives depend on assent to rational impressions and thus on capacities such as language use and inferential reasoning. The Stoic approach is, certainly, cognitive, rather than, for instance, behaviourist, in attributing a key causal role in action to cognition22 – at least, cognition of a certain kind, namely recognition of what is beneficial. On the other hand, the Stoic view on adult humans forms part of a much broader explanatory framework, covering adults, children and non-human animals. In the latter cases at least, the kind of cognitive processes involved (in Stoic terms, phantasiai) consist, rather, in what Magda Arnold characterises as ‘sense judgements’.23 Even in the case of adult humans, we need not suppose that assent to impressions is necessarily a conscious or fully articulated process.24 Also, and more importantly, the Stoic theory of motivation as a whole has a pronounced ‘activist-pragmatist’ character, to use the terminology of Giovanna Colombetti.25 As already stressed, ­­motivation to action in general is conceived in enactivism as part of the sense-­­ making response of an animal to its environment, in a way that contributes to maintaining the animal’s identity by recognising what is beneficial for that identity. Enactivism, in other words, aims to provide an ‘action-focused’ framework for understanding cognition; to this extent we can say that the Stoic theory anticipates the enactive approach.

Stoicism and Embodiment The enactivist approach incorporates an embodied view of cognition, as noted earlier. What is the Stoic position on this question and how does this relate to the features of Stoic thinking considered so far? In modern debates on embodied cognition, Cartesian mind-body dualism typically functions as an extreme paradigm. For the Cartesian view, human life consists in an interaction between a non-material mind, which is the sole locus of mental activities, and a material, extended body.26 The Cartesian view has, I think, few modern defenders; however, 21

See Arius Didymus, Epitome of Stoic Ethics, 7c, 7e, 9, 9b; translation in Inwood and Gerson 1997: 214, 216–17; text and translation in Pomeroy 1999: 46–7, 48–9, 53–5. 22 On contemporary debate (in psychotherapy) about the relationship between cognitive and behavioural approaches, see Dozois and Beck 2011. 23 Arnold 1960: 175; see also Colombetti 2014: 84–5. 24 See Brennan 2003: 262. 25 See Colombetti 2014: 104–5. 26 On the Cartesian model, contrasted with the functionalist one, see Smith and Jones 1986.



ena ctiv ism an d emb od i ed co g ni t i o n 157

a position ­­ often adopted (for instance, by Alvin Goldman) is minimal embodiment. According to this view, all the important features of mental or psychological life are located in the brain. The body is significant only in so far as the brain houses cognitive processes about the body (‘B-formatted representations’). This view is sometimes illustrated by reference to the philosophical thought experiment of the brain in a vat of chemical fluid, maintaining psychological life without direct connection to a living body.27 Shaun Gallagher charts several stronger versions of embodiment,28 of which the most relevant for the present question are biological and enactive embodiment. For both these positions, it is crucial that the body, as distinct from the brain, is seen as making a significant and direct contribution to cognition, of a kind that goes beyond ‘B-formatted representations’. The human being (or animal) is viewed as a unified whole, rather than a combination of a distinct and largely independent body and mind/brain. In ancient philosophy, the extreme paradigm (paralleling the Cartesian view) is Platonic mind-body dualism, especially as presented in the Phaedo. Although this was influential on later Platonic thought, particularly Neoplatonism, psychological physicalism in one or other version was also common in ancient thought, especially Hellenistic philosophy.29 Plato himself offers a partly embodied account in the Timaeus, considered later (‘Plato’s Timaeus and Embodiment’). Stoicism is normally seen as adopting a strong version of the embodied approach. The heart is presented as the main seat of psychological functions (the locus of the ‘controlling centre’, hēgemonikon), though with extensions reaching to other organs. The vehicle of psychological life is pneuma (‘breath’, or a combination of the elements of fire and air), which is itself conceived as being a physical entity.30 However, these points are not enough, by themselves, to show that Stoicism takes an approach that would count as embodied, in a strong sense, in terms of current debate. If closer inspection showed that the Stoics thought that everything psychologically significant took place in the heart-based ‘controlling centre’, their view would amount only to the kind of minimal embodiment adopted by modern thinkers such as Goldman. Relevant too is the cognitivist dimension of Stoic thinking on motivation, noted earlier, which again points to a highly centralised psychological model.31 We need to form a clearer picture of the role of the body as a whole, not just the heart, and of the relationship between the controlling centre and other parts of the body, before trying to determine whether or not the Stoic approach is embodied, in terms relevant for the modern theory of distributed cognition. In pursuing this question, I refer especially to two distinctive features of Stoic thought where the question of mind-body relations arises: first, accounts of 27

See Goldman and Vignement 2009; Goldman 2012. See (last accessed 6 February 2018). 29 See Annas 1992: 2–9. ‘Physicalism’, in this connection, regards the psyche as falling within the scope of ‘physics’, i.e. study of nature, rather than as embodied; however, a number of ancient theories, including Stoicism, combine physicalism and embodiment. 30 See LS 45C–D, 53G–H; also Annas 1992: 37–50, 61–70. 31 See text to n. 22. 28

158

c hr ist opher g i l l

­­ development (as ‘appropriation’), taken in conjunction with Stoic ideas about psychological functions, and, second, accounts of the emotions or passions (pathē). Both topics clarify further how far Stoic thinking is properly compared with modern enactivism. A notable feature of Stoic thinking on the first strand in development is the claim that animals, including humans, have an in-built self-awareness or self-perception. One manifestation of this, noted earlier, is the fact that animals have an instinctive recognition of the functions of the parts of their body, in particular, for defending themselves against other animals, and so maintaining their ‘constitution’.32 Another manifestation is presented in a striking passage from Hierocles’ account of development, which is worth citing at some length (Hierocles 4.38–53 = LS 53B(5–9), LS text and translation abbreviated and slightly modified:33 Ἐπεὶ τοίνυν ἐξ ἀμφοτέρων ἐστὶ τὸ ζῷον … σύνθετον, ἐκ σώματος καὶ ψυχῆς, ἄμφω δ’ ἐστὶ θικτὰ καὶ πρόσβλητα …, καὶ θάτερον μέν ἐστιν αὐτῶν δύναμις αἰσθητική, τὸ δ’ αὐτὸ τοῦτο καὶ τρόπον, ὃν ὑπεδείξαμεν, κεινεῖται, δῆλον ὅτι διανεκῶς αἰσθάνοιτ’ ἂν τὸ ζῷον ἑαυτοῦ. τεινομένη γὰρ ἔξω ἡ ψυχὴ μετ’ ἀφέσεως προσβάλλει πᾶσι τοῦ σώματος τοῖς μέρεσιν, ἐπειδὴ καὶ κέκραται πᾶσι, προσβάλλουσα δὲ ἀντιπροσβάλλεται· ἀντιβατικὸν γὰρ καὶ τὸ σῶμα, καθάπερ καὶ ἡ ψυχή· καὶ τὸ πάθος συνερειστικὸν ὁμοῦ καὶ ἀντερειστικὸν ἀποτελεῖται. καὶ ἀπὸ τῶν ἀκροτάτων μερῶν εἴσω νεῦον ἐπὶ τὴν ἡγεμονίαν …, ὡς ἀντίληψιν γίνεσθαι μερῶν ἁπάντων τῶν τε τοῦ σώματος καὶ τῶν τῆς ψυχῆς· τοῦτο δέ ἐστιν ἴσον τῷ τὸ ζῷον αἰσθάνεσθαι ἑαυτοῦ. Since an animal is a composite (suntheton) of body and psyche, and both of these are tangible and impressible . . . and one of these is a sensory faculty (dunamis aisthētikē) which undergoes movement . . . it is evident that an animal perceives itself continuously. For by stretching and relaxing, the psyche makes an impression on all the body’s parts, since it is blended with them all, and in making an impression it receives an impression in response. The body, like the psyche, reacts to pressure, and the outcome is a state of their joint pressure upon, and resistance to, each other . . . it [this outcome] travels to the controlling centre, with the result that there is an awareness (antilēpsis) both of all the body’s parts and of those of the psyche. This is equivalent to the animal’s perceiving itself. What, exactly, is Hierocles aiming to bring out in this passage? Partly, he seems to be stressing that, just as animals have an instinctive recognition of the capacities of their bodies for external relations (such as self-defence or standing up), they have an instinctive awareness of the capacities of their bodies for psychological functions. The functions cited here are those of the ‘sensory faculty’ and the ‘controlling centre’. Hierocles insists, in fact, that self-perception, in this sense, is prior to 32 33

See text to n. 13. Hierocles was a late but orthodox Stoic writing around 200 ce.



ena ctiv ism an d emb od i ed co g ni t i o n 159

external perception. In addition, it is stressed that this internal recognition is not a purely mental or psychological process but involves the body in a very central way (and indeed ‘all the body’s parts’). Hence, self-perception, in the sense of instinctive awareness of one’s psychological capacities, amounts to perception or awareness of oneself as a psychophysical, embodied whole. This view is sometimes compared to the modern idea of ‘proprioception’, as distinct from external perception or ‘exteroception’.34 How does Hierocles’ account, which is focused on instinctive, primary processes, present in animals from birth, relate to standard Stoic accounts of psychological functioning? Here is one representative statement (by the main Stoic theorist, Chrysippus):35 ἡ ψυχὴ πνεῦμά ἐστι σύμφυτον ἡμῖν συνεχὲς παντὶ τῷ σώματι διῆκον ἔστ’ ἂν ἡ τῆς ζωῆς εὔπνοια παρῇ ἐν τῷ σώματι. ταύτης οὖν τῶν μερῶν ἑκάστῳ διατεταγμένων μορίῳ τὸ διῆκον αὐτῶν εἰς τὴν τραχεῖαν ἀρτηρίαν φωνὴν εἶναι, τὸ δὲ εἰς ὀφθαλμοὺς ὄψιν, τὸ δὲ εἰς ὦτα ἀκοήν, τὸ δ’ εἰς ῥῖνας ὄσφρησιν, τὸ δ’ εἰς γλῶτταν γεῦσιν, τὸ δ’ εἰς ὅλην τὴν σάρκα ἁφὴν καὶ τὸ εἰς ὄρχεις ἕτερόν τιν’ ἔχον τοιοῦτον λόγον, σπερματικόν, εἰς ὃ δὲ συμβαίνει πάντα ταῦτα ἐν τῇ καρδίᾳ εἶναι, μέρος ὂν αὐτῆς τὸ ἡγεμονικόν. The psyche is pneuma (‘breath’) that is integral to our nature, extending continuously throughout the body as long as the regular breath of life is present in the body. Of the parts of the psyche which are assigned to each segment of the body, the one that extends to the throat is voice; that to the eyes, sight; that to the ears, hearing; that to the nostrils, smell; that to the tongue, taste; that to the entire flesh, touch; and that which extends to the genital organs, since it has a different principle (logos), is seminal. The heart is the location of the part where these all meet, which is the governing part (or ‘controlling centre’, hēgemonikon) of the psyche. The main points emphasised here recur in other sources. On the one hand, the Stoics insist, by contrast with Plato (in the Timaeus), that there is a single controlling centre or seat of psychological processes and that the distinctively psychological functions of impression, motive and (in adult humans) assent and rationality are centred there. On the other hand, they also present this organ as the centre of a network running throughout the body (described in various images such as tree and octopus), a network that is psychological as well as physical in nature. The controlling centre, as its name suggests, is presented as authoritative in forming impressions and motives. However, its judgements depend on the reports (which are not merely physical reactions) of the other parts of the body. A further feature 34

See Long 1996: ch. 11, esp. pp. 257–9; also Gill 2006: 38–43. Quotation taken from Galen, On the Doctrines of Hippocrates and Plato 3.1.10–11 (De Lacy), translated by Long 1999: 567, slightly modified.

35

160

c hr ist opher g i l l

underlined here, as in the Hierocles passage, is that the whole network is embodied through and through. This is not only in the sense that the vehicle of psychological process (pneuma) is material, but also that psychological processes are firmly located in the body. This applies also to those of the controlling centre, whose location (the heart) seems to have been chosen partly because of its being central to the body and also because of the communication system (veins and arteries) linking this part to the other parts of the body.36 The second distinctive Stoic idea considered here relates to the emotions or passions. This theory has been widely debated and often criticised, in antiquity and modern scholarship, but most recent treatments have offered a positive view of its credibility and coherence. On the face of it, the theory expresses a strongly cognitive approach in that emotions, which are understood as functions only of adult humans, are taken to depend crucially on beliefs of a certain kind. The Stoic theory, certainly, is cognitive in this respect; it thus differs from accounts of emotions, ancient or modern, which present emotions as functions of a non-rational part of the personality, distinct from the part which forms beliefs. But Stoic emotions are not just beliefs; they also contain other dimensions (affective, subjective and psychophysical), which are widely associated with emotions. This feature of the Stoic conception comes out clearly in one ancient account of the four generic emotions (Andronicus, On Passions 1 (= LS 65B), LS translation slightly modified): Λύπη μὲν οὖν ἐστιν ἄλογος συστολή· ἢ δόξα πρόσφατος κακοῦ παρουσίας, ἐφ’ ᾧ οἴονται δεῖν συστέλλεσθαι. Φόβος δὲ ἄλογος ἔκκλισις· ἢ φυγὴ ἀπὸ προσδοκωμένου δεινοῦ. Ἐπιθυμία δὲ ἄλογος ὄρεξις· ἢ δίωξις προσδοκωμένου ἀγαθοῦ. Ἡδονὴ δὲ ἄλογος ἔπαρσις· ἢ δόξα πρόσφατος ἀγαθοῦ παρουσίας, ἐφ’ ᾧ οἴονται δεῖν ἐπαίρεσθαι. Distress (lupē) is an irrational contraction, or a fresh opinion that something bad is present, at which people think it right to be contracted. Fear is an irrational shrinking, or avoidance of an expected danger. Appetite (epithumia) is an irrational stretching, or pursuit of an expected good. Pleasure is an irrational swelling, or a fresh opinion that something good is present, at which people think it right to be swollen.37 As indicated here, emotions involve beliefs, more precisely two kinds of belief. One belief is that the present or future situation involves something good or bad; the second is that, because this is so, it is appropriate to react in a certain way. The reaction is presented in psychophysical terms, as a ‘contraction’, ‘shrinking’, ‘stretching’ or ‘swelling’. These terms are not meant as metaphors but as literal descriptions of the response of the controlling centre, located in the heart, to the 36 37

See LS 53G–H; also Annas 1992: 61–4; Long 1999: 560–72; Gill 2010: 95–100. For similar ancient accounts, see LS 65C–D, G. A complication in the Stoic theory is that most emotions are seen as misguided and defective, although there are three generic ‘good emotions’; I do not explore this complication here.



ena ctiv ism an d emb od i ed co g ni t i o n 161

formation of the relevant belief. Couched in more technical Stoic terms, an emotion is a motive (hormē) which results from ‘assent’ to a ‘rational impression’ that something is good or bad. We might be inclined to think that the psychophysical response constitutes the emotion, while the belief is merely the trigger for the emotion. But, for the Stoics, it is, quite clearly, the overall experience, consisting in the two kinds of belief and the response, which constitutes the emotion.38 How far do these features of Stoic thought enable us to say whether their approach is embodied, in modern terms? I think it is clear that the Stoic approach goes significantly further in this direction than the minimal embodiment view of Goldman. Both Hierocles’ account of self-perception and Chrysippus’ model of psychic workings attribute psychological functions to the body as an interconnected system, while also allocating a special, unifying role to the controlling centre, which is itself treated as a part of the body and not a radically distinct entity. The emotions are seen as responses of the person (the adult human being) regarded as a psychophysical whole. While beliefs form a crucial part of this response, they are integrally linked with the correlated psychophysical reactions (‘contractions’ and so on). Although these reactions, like the beliefs, belong to the controlling centre, rather than other parts of the body, they are conceived as reactions of the controlling centre qua bodily part and as reactions which are experienced subjectively by the person as a psychophysical whole.39 Also worth noting here is the Stoic belief in ‘pre-emotions’ (propatheiai), instinctive reactions which are not quite (or not yet) belief-based emotions in the full sense. Pre-emotions are presented as reactions involving various parts of the body as a whole and not just the (embodied) controlling centre.40 Taking all this into account, it seems reasonable to see the Stoic theory as constituting a relatively strong version of the embodied approach. How far do these ideas lend further support to the claim that Stoicism prefigures the modern enactivist approach? One relevant factor is that embodiment, as such, is a feature of enactivism, so, to the extent that Stoicism holds an embodied approach, it is in principle closer to enactivism. Put more broadly, enactivism views psychological life (including cognitions and emotions) from the standpoint of a whole person (human, animal or organism) actively engaged in realising his or her objectives as a whole. In exploring this question, particularly as regards emotions, it is useful to refer to Giovanna Colombetti’s synthesis of affective science and enactivism. Colombetti advocates an approach which erodes the boundary between cognition and emotion, rather than treating these functions as sharply different in kind; she also advocates a strongly embodied character of emotion.41 Also, as 38

See Inwood 1985: ch. 5; Brennan 2003: 269–70; Graver 2007: chs. 1–2; Gill 2010: 200–5. See Gill 2006: 246–9; also refs in n. 38. 40 See Seneca, On Anger 2.3.1–2, 24 (= LS 65X), including this list of possible pre-emotional reactions: ‘pallor, floods of tears, sexual arousal, heavy breathing or a sudden brightening of the eyes.’ See also LS 65Y; Graver 2007: ch. 4; Konstan 2017. 41 See Colombetti 2014: 94–8 (arguing for a more fully embodied account of emotions than is often offered in most contemporary psychology), 98–101 (aiming to erode the boundaries between ­cognition and emotion). 39

162

c hr ist opher g i l l

noted earlier, she recommends taking an ‘activist-pragmatist’ view of psychological responses, which sees them as responding to what is relevant in the situation of the agent. Stoicism fits well into this framework in that the Stoics’ account of emotions integrates closely beliefs and correlated psychophysical reactions, rather than treating them as distinct. Stoic emotions, like other types of motives (hormai), can also be seen as expressing an ‘activist-pragmatist’ approach in that they express reactions towards or away from relevant features of the environment regarded as good or bad.42 So these dimensions of Stoic thinking confirm the impression that their theory prefigures an enactivist approach.43

Plato’s Timaeus and Embodiment The Stoic view of embodied psychology, and the analogy with the enactivist approach, is partly anticipated by a striking passage in Plato’s Timaeus (69–72). This passage forms part of a section in Plato’s dialogue (69–76) which anticipates, and probably influenced, the Stoic view of the human body as, in large measure, providentially designed. As suggested earlier, although ideas of providential design are alien to most modern thinking, the ancient idea of the body as a psychological whole, functionally adapted to respond to its environment and maintain its identity, is close to modern enactivism.44 In the Timaeus, 69–72 is the passage of most interest for the present inquiry since it presents the psychological functions as located in various parts of the body, although the head, seat of the brain, is regarded as the chief location.45 How far does this passage present what we can describe as embodiment in a strong sense, viewed from the standpoint of modern debate on this topic? The account, in fact, exhibits competing tendencies in this respect. On the one hand, there are indications of the mind-body dualism that is prominent elsewhere in Plato, particularly in the Phaedo. Plato’s spokesman Timaeus distinguishes between an immortal (and rational) psyche located in the head and a mortal psyche, which experiences pleasure, pain and emotions, such as boldness, anger and hope, located in the rest of the body (69c–e). However, the mind-body dualism is mitigated by the fact that the immortal part of the psyche is seen as expressing itself in a certain type of movement, namely a circular one. The body’s characteristic movements (up, down, right, left, forward, back) are explained as a product of the disruption 42

See text to n. 25. Kazantzidis (Chapter 8 in this volume) also highlights links between Hippocratic thought and embodied (especially enactivist) approaches to emotion. It seems likely that Plato’s treatment of embodied psychology in the Timaeus (see section below), as well as the Stoic account, was influenced by Hippocratic thought; on possible Hippocratic influence on Stoicism, see Tieleman 2003: 180–6. However, as Kazantzidis points out, there is a striking absence of attention to the cognitive dimension of emotion and other psychological states in the Hippocratic writings, by contrast with both Plato and Stoicism. 44 See text to n. 8. 45 On the physiology of the head and brain, see also Ti. 73c–d, 75c–76d. 43



ena ctiv ism an d emb od i ed co g ni t i o n 163

of the circular motions by the process of locating the psyche in the human body. Sensations are explained as a further disruption or ‘collision’, between the body’s motions (both circular and straight) and external objects.46 So, in these respects, Plato’s account represents a qualified kind of mind-body dualism. In other respects, the account comes closer to an embodied approach to psychology. In broad terms, what is presented is an embodied version of the tripartite psyche of the Republic, subdivided into rational, spirited and appetitive parts. As in the Republic, the spirited part is the natural ally of reason, whereas the appetites are less readily subject to reason’s control.47 Hence, in the Timaeus, when the spirited part, located in the heart, receives a message from reason that some injustice or wrongdoing is occurring (from an outside source or from unruly appetites), this gives rise to psychophysical responses which are communicated throughout the body. The spirit/heart ‘boils’ and acts as a message-system so that ‘every bodily part that is sensitive may be keenly sensitised, through all the narrow [blood] vessels’ (70b). A more indirect and complex form of communication is attributed to ‘the part of the psyche which has appetites for food and drink’ (that is, the stomach) which is more ‘beast-like’ and less subject to direct control. Since it is not equipped to respond to reason in the same way as the heart-based spirited part, communication is effected by means of the liver. The physical reactions of the liver to excessive or misguided eating or drinking (that is, becoming bitter, bilious or nauseous) are presented as responses to the intervention of the mind, which aims to control in this way excessive appetites in the stomach (71b–d).48 If we compare this picture with the Stoic one, there is a good deal in common. The human being is treated as, to a large extent at least, a psychophysical whole, coordinated by a single communication system. Although the mind or intellect, located in the head, plays the leading or directing role (like the Stoic controlling centre), other parts of the body are presented as locations for psychological functions. Plato’s account, in fact, gives a more vivid and specific picture than we find in Stoic sources of the different roles, both psychological and physiological, of the various parts. To this degree, the Platonic picture, like the Stoic one, expresses a relatively strong form of psychological embodiment. In so far as Plato views this embodied psychology from the standpoint of natural teleology, presenting the human being as a self-sustaining psychophysical agent, the account also expresses an enactivist approach.49 On the other hand, the indications of mind-body dualism qualify the extent to which the human being can be understood as being, in a 46

See Ti. 43a–44c, esp. 43a–c: the term aisthēseis (‘sensations’) is linked etymologically with aïssein (‘shake’). On rational, divine psyche and circular motions, see 44b, 90c–d. On the combination of circular and other motions in bodily activity (43b–c) and in sensation (43c–44a), see Johansen 2004: 142–52; on circular motions, see Sedley 1999. 47 See Resp. 436a–443e, esp. 439d, 440a–441a, 442a. On significant differences between the psychological models of the Republic and Timaeus, see Johansen 2004: 153–5; Gill 2006: 297–8. 48 For interpretation of this passage, see refs in n. 49. 49 Support for this view (though without reference to enactivism) is offered in Johansen 2004: 154–9; Gill 2006: 299–304, 2010: 120–1. Gill presents a reconstructed ‘Stoic’ reading of the passage.

164

c hr ist opher g i l l

full sense, a psychophysical whole. The account draws a sharp distinction between the divine (rational) psyche and the mortal psyche, and the (spirited and appetitive) functions of the latter psyche are more closely linked to bodily workings and context.50 The description also underlines the contrast between the active and directive reason and the other parts, which do, or should, respond to the orders of reason.51 In addition, Timaeus twice stresses the importance of organising the body’s workings in such a way as to minimise the disruption of the body to the intellectual activity of the rational, divine part of the psyche (69d–e, 70e–71a). Sensations, as noted earlier, are presented as a consequence of the disturbance to the mind’s circular motions by contact with the more diverse movements of the body, rather than as a means by which the human being can sustain itself by responding attentively to its environment.52 Hence, although it is striking that the mind’s activities are characterised in what are in some sense physical terms (circular motions),53 in addition to the other manifestations of embodied psychology, there remain limitations in the extent to which we can characterise this approach as one of strong embodiment. For the same reason, there are limits to which we can characterise the standpoint as prefiguring enactivism, despite the fact that Platonic natural teleology, like Stoic, moves some way in this direction.

The Universe as Rational The final topic in this chapter is another idea which also plays a prominent and distinctive role in Stoicism and Plato’s Timaeus, namely, that the universe as a whole is in some sense rational and provides a paradigm of rationality for human beings. I outline this idea and consider how far it can be helpfully discussed in terms of the theories of cognition that form part of the project on which this volume is based, including enactivism and embodied cognition more broadly, or in terms of modern theory in general. The core idea is that the universe is not only animate or living (it is an animal, zōon), but also contains mind (nous) and rationality (logos). The rationality, as well as the vitality, is, in one sense, concentrated in one part, the world-psyche in Plato, and immanent god, identified with fiery-air or breath (aithēr or pneuma), in Stoicism. But these parts are also integral to the whole and to the cohesion or unity of the whole. In the Timaeus, the world-psyche is inseparably fused with the world-body, and in Stoicism, god-pneuma interpenetrates all other elements in the universe by ‘total blending’.54 So mind and rationality are characteristics of the universe, conceived as an animate being, as a whole. A related shared idea is that human beings 50

Ti. 69c–e, 70a–b, d, 70e–71a. Ti. 70a–b, 71a–b. To this extent, the Platonic picture of these parts is ‘passivist-cognitivist’, rather than ‘activist-pragmatist’, to use terms discussed by Colombetti 2014: 104. 52 See n. 46. 53 See n. 46. 54 See Ti. 30b–c, 36e; LS 46A–B, 48, 54B, F–G. 51



ena ctiv ism an d emb od i ed co g ni t i o n 165

achieve their own perfection by modelling themselves on these features of the universe (mind and rationality), which they are naturally equipped to do.55 One question raised by this set of ideas is this. What is the basis for characterising the universe as a whole as living and, more importantly, as having mind and rationality? In both theories, the answers are relatively clear. The universe is living or animate because, in Plato’s case, the world-psyche is characterised by movement (elsewhere in Plato, psyche is defined as self-moving), and the movements of the world-psyche give rise, through a rather complex intermediate process, to those in the universe as a whole.56 In Stoicism, god-pneuma is also characterised as the animating, shaping, active dimension in the universe as a whole.57 Why, in addition, should we ascribe ‘mind’ and ‘rationality’ to the universe? In both theories, the movements or animating processes are seen as marked by features, notably order, structure, harmony and completeness or wholeness, which are taken to be indications of mind (nous/ratio) or rationality (logos). In both cases, the clearest indications are taken to be the regular movements of the planets (what we call the solar system, but conceived in geocentric form).58 These are analysed, in Plato’s case, in mathematical terms, based on contemporary astronomical work done in the Academy, and described by Plato as ‘the circle of the same’.59 Human beings pattern themselves on the universe in a rather literal way, by bringing the circles in their minds (inside their heads) into line with those in the planetary system.60 In Plato to some extent, and in Stoicism to a greater extent, these features of the planetary system are reflected in natural processes on earth (night and day, the cycle of the seasons, etc.) and in forms of regularity and order that pervade all forms of life on earth.61 The natural capacity of humans as rational animals to pattern themselves on the universe is thus just one aspect of this pervasive world-order. In reflecting on the relationship of these ancient ideas to modern ones, an initial question is whether the features seen by Plato and the Stoics as indicative of mind and rationality are ones that we can appropriately describe as ‘cognitive processes’. It is true that they sometimes use language which may strike us as cognitive: thus, in the Timaeus, at one point (37b–c), the world-psyche is said to have opinions and knowledge. However, as noted earlier, the main ground for ascribing mind and rationality to the universe, in both ancient theories, is the fact that the planetary system, in particular, exhibits – what we (human beings) can recognise as – order, structure and completeness in the physical movements of the planets in space. Ancient critics of Stoicism, in particular, challenged this extension of the language of rationality, arguing, in effect, that it made an illegitimate move from the credible claim that the universe is intelligible to the further, problematic claim that it is 55

See Ti. 47b–c, 90d; LS 54H(1), 63C(3–4). See Ti. 34c, 36a–e, 38b–40b; for psyche as self-moving, see e.g. Phdr. 245c, Leg. 895b–897b. 57 See LS 46A–B. 58 See Ti. 30a–c, 34a–b, 35a–37a, 38b–40b. See Cic. Nat. D. 2.15, 43, 54–6, 97. 59 See Burnyeat 2000: 57–67; Zedda 2000. 60 See Ti. 47b–c, 90d; Sedley 1999. 61 See Ti. 39b–d, 47a–b; Cic. Nat. D. 2.49–50, 83, 102–3, 120, 131–2. 56

166

c hr ist opher g i l l

intelligent.62 This is also a possible modern response; a modern counter-response would depend on considering whether the notion of cognition can span human psychological functions and processes built into the universe as a whole, a large question whose implications are noted shortly. The Platonic and Stoic ideas about cosmic rationality have certain links with those already discussed here; for instance, in both theories all the points discussed depend on strong claims about natural teleology. To what extent can these ancient ideas about cosmic rationality also be usefully linked with contemporary debate about cognition? From the standpoint of current debate, the ancient ideas can be seen as pointing in divergent directions. On the one hand, the Platonic and Stoic ideas can be taken as an extension of the features of their thinking which prefigure enactivist and embodied psychology. As suggested here, in Stoicism (and to some extent Plato’s Timaeus), human beings and other animals are viewed from an enactivist-type perspective, as organic, psychophysical wholes, with cognition forming an integral part of their organic life and response to their environment. In these further Stoic and Platonic ideas, the universe as a whole is viewed as a psychophysical organism or animal, and cognitions and other psychological processes are presented as integral elements in its organic life – though in this case there is no reference to an external environment. For related reasons, these ideas can be seen as reflecting an embodied approach to cognition, in a certain sense. Physical processes in the universe, above all the regular, circular movement of the planets, are taken as expressions of rationality and, as just noted, sometimes characterised in terms of the language of cognition. In the Timaeus, in particular, the identification of rational thought with actual circular motions is striking, especially against the background of the mind-body dualism which is a more prominent feature of Platonic thought, and which is also indicated in this work. The idea that, in both Plato and Stoicism, human beings should develop their own rationality by imitating the patterns of rationality in the universe as a whole confirms the linkage between the human and cosmic aspects of this theme and thus supports the link with the embodied approach to psychology. From a different standpoint, these ancient ideas can also be seen as anticipating another approach in modern cognitive science and philosophy. This approach underlines the idea that cognition can be ‘extended’ or ‘distributed’ from the human mind/brain (or from the individual human mind/brain) to objects, tools or artefacts, or other individuals, or social and organisational groups and structures, that is, the ‘environment’, broadly conceived. This kind of approach is sometimes characterised as a ‘functional’ one, in that the cognition is defined in terms of the function that it fulfils rather than by reference to its role in its biological or organic context.63 A functional approach, in this sense, differs from the more biologically 62

See Cic. Nat. D. 2.22 (Stoic position), 3.22–34 (criticisms). See also on this debate Sedley 2007: 225–30. 63 See Michael Wheeler’s outline of distributed cognition, available at (last accessed 7 February



ena ctiv ism an d emb od i ed co g ni t i o n 167

grounded embodied enactivism on which I have mostly focused here. However, some enactivists argue that their theory can also accommodate versions of the extended mind theory.64 It might seem tempting to associate these Platonic or Stoic ideas with the functional version of extended cognition. What the ancient theories accentuate, one might suggest, is the functional character of the thinking involved (the fact that this expresses order, structure, harmony and completeness), rather than the links with the biological or cosmic context. However, there are difficulties in pressing this second linkage between these ancient ideas and the modern theory of extended cognition. The modern theory takes human, brain-based, cognition as central or paradigmatic, even if it also recognises that this can be ‘extended’ or ‘distributed’ more widely. For Plato’s Timaeus and Stoicism, on the other hand, the rationality in the universe is taken as paradigmatic and human beings are presented as becoming more rational by taking the universal rationality as normative.65Also, the idea that the universe embodies rationality in the sense claimed by Plato and the Stoics is a challenging and unfamiliar one in modern thought, and it does not normally form part of contemporary debate about cognition. However, I close by noting two current approaches which argue for ideas comparable with the Platonic and Stoic ones, though not generally in the context of cognitive theory. One is the idea that ‘information’ is not just a unit of analysis or a component of artefacts (especially computers) but is in some sense the building-block of the natural universe.66 The other is the idea that the universe has, or is, a mathematical structure, in its essential nature, and is not just an entity which can be analysed in terms of mathematical structure. (An implication of the second idea, at least in some theories, is that there are a number of actual or possible universes with the same structure, sometimes called ‘multiverses’.)67 These ideas seem, on the face of it, relevant for the comparison with Platonic and Stoic ideas because they present cognition not just as a function of human beings or as a projection by human beings on to other things but as a core element in the universe. Although the modern theories, as far as I am aware, do not also present the universe-based versions of cognition or rationality as paradigmatic for human thinking, this would be a possible extension of these lines of thought. A further, and perhaps more contingent, linkage between these approaches and, at least, the Platonic equivalent is this. The emergence of these ideas in modern thought seems to have been stimulated by drawing out the wider implications of certain theoretical or technological innovations for understanding the universe or nature (or human nature) more generally. In the modern sphere, the innovations have been in computing and artificial intelligence 2018), referring, among other works, to Clark 2003 and Noë 2009. See also Andy Clark’s outline of the extended mind, available at (last accessed 6 February 2018). 64 See Di Paolo 2009; Gallagher 2013; Colombetti 2015. 65 See n. 60. 66 See e.g. Wheeler 1990a, 1990b. 67 See e.g. Tegmark 2014.

168

c hr ist opher g i l l

(­regarding information) or quantum mechanics (regarding mathematical structure or ‘multiverses’). In Plato’s Academy, the innovations were centred on mathematics and the analysis of astronomical observation in mathematical terms.68 Pursuing further the nature of these modern theories and the analogy with Platonic or Stoic ideas would be a rather complex undertaking and would lead me beyond the scope of the subject of this chapter. However, simply citing these points of connections may serve to highlight the contemporary interest of ancient ideas about human and cosmic rationality that might otherwise seem to have little resonance today.69

68 69

See references in n. 59. I am most grateful to the volume editors and their readers for advice and suggestions, especially to Mark Sprevak for bibliographical guidance on the final topic, and to Giovanna Colombetti for very helpful comments and references. I would also like to acknowledge the support of a Leverhulme Emeritus Fellowship for the research and composition of this chapter.

e narge i a, en ac t iv ism an d t he an ci ent readerl y i m ag i nat i o n 169

10 Ena r g eia, E n a ct i vis m a nd the A n ci e n t R e a de rly Ima g ina tion 1 Luuk Huitink

Introduction In recent years reader-response theorists have begun to explore the potential of the ‘second-generation’ enactivist theory of cognition for the analysis of readers’ quasi-visual responses to literature.2 While these studies bring to bear a range of perspectives on the phenomenon by which readers may at times feel that they can vividly and directly perceive the storyworld, they converge on a view of vision and imagination, not as driven by mental images ‘seen’ with the mind’s eye (the positing of internal representations being a central tenet of ‘first-generation’ cognitive studies), but as enactive, that is, a way of acting. As I shall explain in greater detail below, enactivism posits that (quasi-)visual experience is not all ‘in the head’, but emerges through interactions with the world around us. What I wish to contend in this chapter is that there are salient and non-trivial points of resemblance between these enactivist views of the readerly imagination and the ancient rhetorical concept of enargeia (usually translated as ‘vividness’). This term, which on a minimal ancient definition is said to be ‘speech which brings the subject-matter before the eyes’, figures prominently in ancient criticism and rhetoric and is used in particular to characterise the vivid, quasi-visual effects which epic poetry, historiography, oratory and occasionally other kinds of literature may have on readers.3  1

This chapter is part of the work of the Heidelberg ERC group Experience and Teleology in Ancient Narrative (ERC Grant Agreement no. 312321 (AncNar)).  2 For general introductions to ‘second-generation’ cognitive criticism, see Kukkonen 2014; Kukkonen and Caracciolo 2014; Cave 2016. Other pioneering studies on which I draw are Grünbaum 2007; Jajdelska et al. 2010; Bolens 2012; Kuzmičová 2012a, 2012b, 2013; Troscianko 2013, 2014; Caracciolo 2014. An application of enactivist insights to the style of Homer is offered by Grethlein and Huitink 2017.  3 Anon. Segu. 96: ἔστι δὲ ἐνάργεια λόγος ὑπ’ ὄψιν ἄγων τὸ δηλούμενον. For lists of ancient definitions along these lines, see Manieri 1998: 123 n. 404; Lausberg 1998: 359–61. Related critical terms are φαντασία, πρὸ ὀμμάτων τίθεσθαι/ποιεῖν, διατύπωσις, inlustratio, evidentia, demonstratio, sub oculos subiectio. Throughout, I use ‘readers’ to include live audiences, putting to one side the question of whether the mode of reception influences the experience of mental imagery, for the reason that more research is required; cf. Kuzmičová 2013: 123. Besides, my chief concern is with ancient

170

lu UK hui t i nk

Now, at first sight this contention must appear implausible. For, whilst acknowledging that enargeia is a multifaceted concept, recent classical scholarship holds that the term primarily refers to ‘pictorial vividness’, as brought about through the transference of picture-like internal representations from author to reader by means of detailed descriptions (ekphrasis), and this account ostensibly matches the references to concepts like ‘mental images’ and ‘the mind’s eye’ found in ancient comments.4 Thus, a scholiast (Σ bT Il. 23.362–72) praises the first lines of the description of the horse race for Patroclus’ funeral games in the Iliad by stating that Homer ‘has projected the entire mental image (phantasia) so vividly (enargōs) that the readers are no less captivated than the spectators (theatōn)’ (πᾶσαν φαντασίαν ἐναργῶς προβέβληται ὡς μηδὲν ἧττον τῶν θεατῶν ἐσχηκέναι τοὺς ἀκροατάς). Plutarch (Artaxerxes 8.1) expresses admiration for Xenophon’s account of the Battle of Cunaxa in the Anabasis, on the ground that ‘Xenophon brings it all but before the eyes (opsei) and through his vividness (enargeian) all the time places the reader, much affected and sharing in the dangers, near to the action, as if it had not been concluded, but is going on’ (Ξενοφῶντος δὲ μονονουχὶ δεικνύοντος ὄψει καὶ τοῖς πράγμασιν ὡς οὐ γεγενημένοις, ἀλλὰ γινομένοις ἐφιστάντος ἀεὶ τὸν ἀκροατὴν ἐμπαθῆ καὶ συγκινδυνεύοντα διὰ τὴν ἐνάργειαν). Quintilian (Institutio oratoria 8.3.62), finally, calls enargeia a ‘great virtue’ (magna virtus) of style, claiming that ‘a speech does not adequately fulfil its purpose or attain the total domination (plene dominatur) it should have, if it goes no further than the ears, and the judge feels that he is merely being told the story of the matters he has to decide, without their being brought out and shown to his mind’s eye (oculis mentis)’ (‘magna virtus est res de quibus loquimur clare atque, ut cerni videantur, enuntiare. Non enim satis efficit neque, ut debet, plene dominatur oratio, si usque ad aures valet atque ea sibi iudex, de quibus cognoscit, narrari credit, non exprimi et oculis mentis ostendi’). Given this emphasis in our ancient sources, the case for a non-pictorialist, enactivist interpretation of ancient enargeia will have to be made carefully and gradually. I shall first summarise the prevailing ‘pictorialist’ account and show that supplementing it with an alternative approach is both desirable and methodologically feasible. I will then introduce the main tenets of the enactivist account of vision and imagination, before presenting two case studies which show that this account is often better able to deal with the way ancient critics speak about enargeia than the pictorialist one. The goal of this chapter is twofold. On the one hand, by critics’ responses, and they at least encountered the texts stripped of their performative context; cf. Nünlist 2009: 12.  4 Important analyses of the sources within their wider intellectual context include Zanker 1981 (see p. 297 for the rendering ‘pictorial vividness’); Meijering 1987; Vasaly 1993; Innocenti 1994; Manieri 1998; Otto 2009; Webb 2009, 2016; Bussels 2012; Sheppard 2014; Germany 2016. As the overviews of key sources of especially Meijering (1987: 39–44), Manieri (1998: 79–192) and Otto (2009: 67–134) show, one way in which enargeia is a multifaceted concept is that it comprises more than just visual imagery; for instance, references to phonetic effects and onomatopoeic devices in various sources indicate that auditory imagery also comes under its purview. However, the present chapter only focuses on aspects of enargeia which are broadly visual, but not ‘pictorial’.

enargeia , ena ctiv ism an d t he readerl y i m ag i nat i o n 171



­­illuminating aspects of enargeia which the pictorialist account plays down or leaves unexplained, I aim to contribute to a deeper understanding of ancient reading habits. On the other hand, I also hope that, conversely, critics’ ongoing efforts to shape a second-generation cognitive criticism may benefit from taking note of the ancient rhetorical tradition, which they have so far largely ignored.

Enargeia and phantasia: The Pictorialist Account Ancient remarks on enargeia like those quoted at the outset evince a confidence in the feasibility and efficacy of quasi-visual reader responses which contrasts sharply with modern critical attitudes to the phenomenon. To be sure, folk-psychological assumptions about how the imagination works still make it seem plausible that we might in theory be able to ‘picture’, say, Mme Bovary or a room she is in when we come across sufficiently detailed descriptions whilst reading Flaubert’s novel (cf. Jajdelska et al. 2010: 437, 440–1). However, most readers actually experience mental images as feeble and fleeting and as rarely, if ever, attaining anything like the vivacity and vitality of actual visual perception.5 Partly for this reason, modern critics for a long time did not set much store by quasi-visual responses and tended to distinguish them sharply from (supposedly more rewarding) intellectual and reflective types of response (see Esrock 1994: 1–6). Iser, for instance, arguably the most influential reader-response theorist of the past half-century or so, concedes that mental imagery ‘accompanies’ reading, but insists that it remains distinguished by its ‘optical poverty’, as becomes evident when ‘one sees the film version of a novel one has read’; verbal descriptions are not so much processed mimetically, that is, by imaginatively perceiving their referents, as semiotically, that is, on the basis of the words’ conceptual meanings and connotations: ‘Even if we are given a detailed description of a character’s appearance, we tend not to regard it as pure description, but try and conceive what is actually to be communicated through it’ (Iser 1978: 137–8). The terminology used by ancient critics is often thought to be revealing of certain preconceptions about vision, imagination and language which help explain the difference between ancient and modern attitudes. In particular, scholars have pointed out that in Hellenistic (especially Stoic) philosophy phantasiai were, at least in part, conceived of as pictorial images that were permanently stored in memory in the form of physical imprints on the mind; they were thought to form as a result of visual perception and in turn to give rise to, and be originary of, thought and language. The term enargeia itself was used in this connection to refer to situations  5

See Scarry 1999: 3–4; Cave 2016: 78–9. There is, to be sure, individual variability in people’s capacity to ‘picture’ things (Cui et al. 2007), but no literary theory can wholly avoid positing a ‘model’ reader and it is not absurd to look for common denominators when talking about the cognitive abilities of human adults; nor do we have to rely on introspection alone: empirical studies (e.g. Allington 2011) show that many readers find it difficult to construct clear and complete mental images from verbal descriptions.

172

lu UK hui t i nk

in which phantasiai are grasped with particular ‘clarity’ and we can accept them as true and accurate representations of the world. These ideas, so the argument goes, informed ancient criticism and meant that mental images were regarded as potentially very vivid and as the indispensable medium connecting the visual and the verbal on the one hand, and perception, imagination (‘seeing’ in the mind’s eye) and conceptual thought on the other. Specifically, ancient critics will have supposed that the individual items in a detailed description (ekphrasis) could elicit lifelike mental images from readers’ memories which they pieced together until they saw a detailed ‘picture’ of the description’s referent in the mind’s eye.6 In order to see how this is supposed to have worked out in practice, we turn to a much-discussed passage from Quintilian, one of our richest sources. Quintilian’s remark, quoted in the previous section, that a speech, provided it possesses the quality of enargeia, may come to ‘dominate’ the judge, introduces his treatment of enargeia as an element of the ornate style.7 He next declares that an elaborate analysis would be cumbersome (Inst. 8.3.63) and then mostly contents himself with citing a number of passages which apparently achieve the coveted effect. Among these is a vignette of the aftermath of a luxurious banquet from an otherwise lost speech of Cicero’s, which was presumably designed to stir the audience’s indignation: ‘videbar videre alios intrantis, alios autem exeuntis, quosdam ex vino vacillantis, quosdam hesterna ex potatione oscitantis. humus erat inmunda, lutulenta vino, coronis languidulis et spinis cooperta piscium’: quid plus videret qui intrasset? (Quint. Inst. 8.3.66) ‘I seemed to see some coming in, some going out, some reeling with drink, some dozing after yesterday’s potations. The floor was filthy, swimming with wine, littered with wilting garlands and fishbones.’ What more could anyone have seen who had entered the room? (trans. Russell 2001, slightly adapted) Now from a modern perspective, this passage does not live up to the strong claim which Quintilian makes for it. While it may be ‘graphic’ in a loose sense of that word, it offers such sparse information about the outward appearances of the banquet hall and the banqueters that an actual witness would surely have seen more than it records (cf. Vasaly 1993: 97; Webb 2009: 108). Its effect rather depends on the careful selection of ‘telling’ details, but if we are precise about that, it must be  6

The clearest (and most explicitly cognitive) accounts along these lines are Vasaly 1993: 91–9; Webb 2009: 93–7, 110–4, 2016: 205–11. See also Zanker 1981: 308–10; Bussels 2012: 61–71. Elaborate overviews of the Hellenistic philosophical ideas about phantasia and enargeia are offered by Manieri 1998: 27–51, 113–23; Otto 2009: 31–66. But mutatis mutandis, thinking about words as vehicles of images stretches back to classical times (cf. Manieri 1998: 27–75; Sheppard 2014: 1–13) and is an implicit cornerstone of Homeric poetics (Squire 2013).  7 Inst. 8.3.61–71. He provides other brief discussions of enargeia in sections on the virtutes narrationis (Inst. 4.2.63), the arousal of emotions (6.2.25–36) and the figures of thought (9.2.40–4).



enargeia , ena ctiv ism an d t he readerl y i m ag i nat i o n 173

acknowledged that the reference to ‘seeing’ obscures the fact that the impact of these details is not so much sensory as semantic, that is, based on what Vasaly calls the words’ ‘symbolical meanings’ – of, say, the ‘tottering’ (vacillantis), from which readers can infer the guests’ drunkenness, or of the ‘garlands’ (coronis), which should have no place at a Roman dinner table (Vasaly 1993: 20, 98–9; similarly Webb 2009: 109–10). However, because of the phantasia-theory outlined above, things may have looked different from an ancient point of view. If for Quintilian and others the generous provision of visual details does not seem to be at stake, then this is perhaps because of the perceived clarity of phantasiai and their close connection with words and prior visual perception: as long as authors verbalised the picture to be painted in a series of stock motifs of which they could assume readers to have prior visual experience, then the latter could also be assumed to be able to ‘work their way back up the chain’ and ‘reconstruct . . . the original mental image which gave rise to the words that prompt the reader’s own mental image’.8 A strong belief on the part of ancient critics that a few concrete words effortlessly evoked mental images may even have blinded them to other possibly relevant linguistic aspects of enargeia: in accounting for the vividness of a description like Cicero’s ‘we might prefer to focus on the use of figures, on the choice of vocabulary and the arrangement of words. But this is not the case for ancient critics’ (Webb 2009: 93; original italics). Furthermore, since phantasiai (as elicited by words, but not words qua words) were regarded as the basis of thought, ancient critics may have almost automatically rolled visual and more reflective responses into one; when Quintilian and others talk about ‘seeing’ the storyworld, this involves, in the words of Webb’s slogan, both ‘sight and insight’, as ‘seeing’ phantasiai in the mind’s eye was but the starting point for a more elaborate response, which ‘brings with it knowledge, both intellectual and sensory’.9 Finally, in line with a current trend in Classics to cast even basic sensory and cognitive processes as socio-intellectual constructs,10 it has been suggested that these considerations may have more than purely theoretical significance. Vasaly speculates that ancient readers ‘may well have possessed powers of pictorial visualisation much greater and more intense than our own’ (1993: 99). Germany thinks that they may have been ‘peculiarly susceptible to vivid descriptions’, as that ‘would certainly explain a great deal about why our sources speak about enargeia the way they do and with such perfect consistency’ (2016: 19). Webb, finally, sketches  8

Webb 2009: 96; cf. Calboli Montefusco 2005: 51–2. The imagination was regarded as reproductive rather than creative, and several procedures were developed to cope with the formation of mental images of things previously unseen; Lucretius (4.739–43), for instance, says one can imagine a centaur by combining memory-images of a man and a horse; cf. Vasaly 1993: 97; Sheppard 2014: 11–13; Webb 2016: 214–16.  9 Webb 2016: 219. Cf. Webb 2009: 110, stating that details such as the garlands would have been ‘so loaded, so telling, for Roman readers like Quintilian that they might not themselves have been fully aware of the amount of decoding involved in their response’. 10 Squire 2016 is a good example and concerns vision.

174

lu UK hui t i nk

a culture in which a strong belief in the power and communicability of mental images and rhetorical training inspired by that belief produced readers who could be counted upon as a matter of course to be ‘constantly engaged in a process of calling up, developing and reflecting on these images’ (Webb 2016: 208).

Some Shortcomings of the Pictorialist Account While the pictorialist account of enargeia is illuminating in certain respects, I think we should be wary of attributing ‘intense’ powers of ‘pictorial visualisation’ to Greek and Roman readers too soon. Apart from the fact that such a move must remain speculative – and on balance it seems implausible that ancient and modern readers had very different cognitive capacities – it risks blinding us to an alternative possibility, namely that in important ways enargeia differs from ordinary conceptions of ‘pictorial vividness’, not in degree, but in kind. For instance, are we really to suppose that Quintilian – of all people, and in a discussion of the ornate style – was less sensitive than ‘we’ are to, say, the string of asyndetic and partly anaphoric clauses in Cicero’s vignette and that these played no role in his choice of this passage as an outstanding example of visually compelling writing? And what are we to make of the fact that enargeia (and ekphrasis, for that matter) is routinely used to characterise not only (not necessarily elaborate) descriptions of static scenery and objects, but also fast-paced narrative passages, such as the Iliadic horse race or Xenophon’s account of Cunaxa?11 Rather than regarding these and other stylistic features of enargeia as irrelevant or as indications of the ease and intensity with which ancient readers allegedly formed mental images, we may wish to ask if mental images are always the best way of thinking about enargeia. When we turn from the stylistic to the experiential side of things, similar questions arise. What are we to make, for instance, of the fact that, as Webb has recently remarked (2016: 213), ancient authors often speak as if enargeia involves the visualisation of a scene, ‘not simply as in a distanced, disembodied photograph, but as if we were present ourselves within the same space, in bodily contact with the place and its happenings’? One may here compare how the scholiast cited above likens the readers of the Iliad to the captivated intradiegetic audience of the horse race for Patroclus, how Plutarch suggests that Xenophon’s enargeia elicits a sense of being physically present in the plain of Cunaxa, or how Quintilian envisages the audience of Cicero’s vignette as having ‘entered’ the banquet hall. And finally, while it is clear that enargeia was supposed by ancient critics to provide certain kinds of understanding which we are not used to putting under the rubric of (mere) ‘sight’, is ‘insight’ always the best term to typify the surplus of ‘meaning’ generated by enar11

As noted by e.g. Calame 1991: 13–14; cf. Vasaly 1993: 91; Manieri 1998: 144–5; Calboli Montefusco 2005: 57. See also Webb (2009: 67–8) on narrative as a common topic of ekphrasis in ancient rhetorical handbooks. In response, Webb (2016: 210) now speaks of a ‘virtual film’ seen in the mind’s eye rather than a ‘picture’, but this reformulation retains the ‘representationalist’ bias which the present chapter aims to qualify.



enargeia , ena ctiv ism an d t he readerl y i m ag i nat i o n 175

geia? Thus, Quintilian’s claim that a speech requires enargeia if it is to ‘dominate’ the judge surely carries a hint that displaying the subject-matter to the mind’s eye is an effective way for an orator to avoid inviting too much contemplation.12 Plutarch’s claim that, due to enargeia, readers of Xenophon will experience the same emotions as the actual participants in the Battle of Cunaxa (‘sharing in the dangers’) does not read as if he thought of himself and others as routinely investing mental images with intellectual and emotional meaning in a gradual and inferential process of sense-making; it is rather suggestive of a more immediate, knee-jerk and visceral, sort of response. In considering such questions, it is important to note that ancient critics themselves appear to have felt the need to redefine the philosophical terms they adopted. Cicero, for one, always translates enargeia as evidentia in philosophical but as inlustratio in literary contexts, which may indicate that for him the ‘clarity’ of the mental images resulting from visual perception was not the same as visual ‘vividness’ in literature.13 Ps.-Longinus (On the Sublime 15.1) presents his emotionally charged definition of phantasia, ‘when under the influence of inspiration and emotion you seem to see what you describe and place it before the eyes of the readers’, as a break with the Stoic usage of the term (see Otto 2009: 91–2). Quintilian (Inst. 6.2.29–30) introduces his discussion of the arousal of emotions through enargeia by providing a double gloss on phantasia: he uses the word not only for discrete mental images (imagines), but also for the human capacity actively to engage the imagination (visiones) and to do so in such a way that at times we no longer seem to ‘think’ (cogitare), but actually to ‘do’ (facere) what we imagine (see Otto 2009: 109–10; cf. Webb 2016: 210). To be sure, formulations like these describe ‘intense’ responses, but they have little to do with philosophical ideas about mental images passively received and contemplated in the theatre of the mind. For such reasons I propose now to abandon the notion of mental images as the explanatory medium and of ‘seeing’ such images in the mind’s eye as the phenomenon to be explained and to examine instead if the non-representationalist enactivist theory of vision and imagination can help us develop a more precise account of some crucial stylistic and experiential aspects of enargeia. On the assumption that the main thrust of enactivism is correct, this is not a priori implausible. As we have seen, the significance of the pictorialist terminology used by ancient critics should not be overestimated; certainly terms like enargeia and phantasia did not retain their full philosophical meanings when they were transferred to a new context. And since rhetorical theory is ultimately a distillation of what worked best in practice, it is not unlikely that where our sources depart from the p ­ hilosophical 12

Cf. Calboli Montefusco 2005: 48; Dozier 2013: 151–2. Cf. Ps.-Longinus, Subl. 15.9–11, arguing that enargeia in rhetoric does not simply ‘persuade’ readers, but ‘enslaves’ (δουλοῦται) them, and that because of it ‘the factual aspect of a case lies buried because it is outshone’ (τὸ πραγματικὸν ἐγκρύπτεται περιλαμπόμενον); on the reasoning here, see Otto 2009: 95–6, 101; Halliwell 2011: 349–51. 13 Otto 2009: 104–5; Bussels 2012: 72–3. It is Quintilian who muddies the waters.

176

lu UK hui t i nk

­ nderstanding of the imagination, both in their analytical remarks and in the u examples which they adduce, they offer glimpses of actual, and cognitively realistic, reader responses.14 Of course, this does not mean that there existed a full-blown conception of the readerly imagination as enactivist in antiquity or that pictorialist preconceptions did not influence critical and readerly practice. It does, however, raise the possibility that ‘pictorialism’ is not the whole story and that the enactivist framework can help us bring out strands of enargeia which have hitherto remained underappreciated.

The Enactivist Account of Visual Experience and Imagination The widespread view that picture-like mental images are indispensable to the proper functioning of the imagination, conceived of as simulated vision, arises in part from the folk-psychological assumption that visual perception itself is like taking detailed, high-resolution ‘snapshots’ of the world (see Jajdelska et al. 2010: 437, 440–1; Cave 2016: 79). Noë’s enactivist theory of visual perception takes issue with this view.15 Noë infers from much-discussed perceptual phenomena such as inattentional and change blindness that the perceived world is not in fact like a gap-free and detailed photograph, and posits instead that vision is, to a substantial degree, selective and attention-dependent.16 The world, that is, is not given, but ‘made available’ by the activity of the observer, who attends to selected aspects of it through appropriate movements of the eyes, head and body (O’Regan and Noë 2001a: 956; Noë 2004: 16–17). One might think here of how, when we attentively follow a tennis match, for instance, we have only a dim visual awareness of the crowd or the stadium. Yet, the world does not ‘feel’ gappy to us, because we know that we can always ‘look’ further as and when we need to, by moving our eyes, head and body to focus on different aspects of the environment. The enactivist account understands vision in terms of an ongoing, attentive interaction between an embodied observer and the environment, mediated by the observer’s mastery of the laws of sensorimotor contingencies, defined as the regularities in how sensory stimulation depends on the movements of the perceiver or the percept (O’Regan and Noë 2001a: 944–5; Noë 2004: 15–20). Our bodies, that 14

Cf. Heath (2009) on the balance between theory and practice in ancient rhetoric. While ancient authors’ reluctance to engage in elaborate analyses of enargeia has been attributed to the fact that the underlying ideas about phantasiai were so widely known that further explication was superfluous (Webb 2016: 208), it is more plausible that enargeia was ultimately a rather elusive concept to them, which captured a phenomenon (or a set of phenomena) which was in itself real enough and prompted by certain stylistic properties of texts, but which could not easily be accommodated in received systems of rhetorical theory and therefore could be only partially explicitly analysed; cf. Vasaly 1993: 90; Innocenti 1994: 360; Manieri 1998: 148. 15 O’Regan and Noë 2001a, 2001b; Noë 2004. For the background of enactivism in the phenomenological tradition associated with, among others, Merleau-Ponty, see Gallagher 2009. 16 Noë 2004: 49–57. For an in-depth treatment of selective visual attention, see Zhang and Lin 2013.



enargeia , ena ctiv ism an d t he readerl y i m ag i nat i o n 177

is, relate to interaction potentials (‘affordances’) of the environment, as is measurable in neuronal activity associated with anticipatory movements that occur during perception in ‘the whole neurally enlivened body’.17 As this sensorimotor feedback gives us information about what it would be like to interact with the environment, we perceive it in these terms. Actual and potential movements – both of the observer’s eyes, head and body and of the observed entities – gain salience for visual perception. To make this tangible, one can think of how a tennis player directs her gaze and whole-body intention to an incoming ball, incorporating its trajectory – she literally moves with the ball and ‘feels’ it approaching – and thereby at the same time already prepares for a well-aimed return.18 The claim that vision is enactive entails the claim that it is ‘smart’, in that it determines the basic way in which we make sense of (‘enact’) the world (Gallagher 2008: 535–6). Thus, the enactivist theory of vision delivers everything we need in order to explain how the tennis player from our example hits the return; there is no need to evoke further cognitive acts on her part, such as her forming a belief about, say, the speed or size of the ball on the basis of ‘snapshots’ of it, and to posit that such ‘higher-order’ mental processes go into planning the return. On an enactivist view, the player’s conscious experience of things like the size, volume and distance of the ball is due to sensorimotor feedback (based on previous visual interactions with this or similar balls), which emerges during, and so is part and parcel of, visual perception (see Gallagher 2001: 88, 2008: 537). A similar case has been made for the ‘emotional colouration’ of visual perception (e.g. the felt excitement and exhilaration involved in preparing and hitting a return); on an enactivist view, the neural activation associated with this is indicative not of a separate cognitive act (by which I infer something from what I see), but of how embodied emotions inform visual perception from the start.19 According to a number of cognitive scientists, finally, the ‘smartness’ of visual perception extends to ‘primary intersubjectivity’, that is, the human ability directly to see other people’s movements in the context of the surrounding world as expressive of specific intentions and emotions.20 On an enactivist view, our interactions with people at this level do not involve advanced inferential cognitive processes based on a ‘Theory of Mind’, but a visually mediated bodily attunement between two or more agents. An example is how our tennis player’s return depends on her ability to incorporate not just the ball’s trajectory, but also the movements of her opponent. Evidence from research into mirror neurons, which shows that the perception of an action activates the same areas of the brain as executing that action, indicates that our direct grasp of what others are doing is facilitated 17

O’Regan and Noë 2001b: 89. For ‘affordances’, see Noë 2004: 21, 103–6. I owe this example to Fuchs and De Jaegher 2009: 473. 19 Gallagher 2008: 538. For embodied emotions, see Colombetti and Thompson 2008. 20 Gallagher 2001, 2008; Fuchs and De Jaegher 2009. Bolens (2012: 1–2), adopting a term from Spolsky, calls it ‘kinesic intelligence’. As Gallagher (2001: 91–2) notes, this does not mean that our efforts to ‘read’ others are never conceptual (‘secondary intersubjectivity’), but in many situations primary intersubjectivity suffices. 18

178

lu UK hui t i nk

by a mechanism which connects their bodies to ours.21 While the interpretation of this evidence is controversial, it seems that our perception of others’ actions always involves a proprioceptive component, and in marked cases – Fuchs and De Jaegher give the example of someone absorbed in watching the salto mortale of an aerial acrobat – our engagement with others’ bodies in action may reach a degree of ‘fascination’, which may cause our lived body to reach toward and ‘conjoin’ with the acrobat’s swinging movements to the extent that we may be prompted to co-movements (Fuchs and De Jaegher 2009: 473–4). If this is what visual perception is like, then what does that entail for the imagination, defined as simulated vision? The enactivist account suggests that the phenomenal similarity between ‘seeing’ and ‘imagining’ is not to be found in their alleged content, namely mental images (for we do not take detailed ‘snapshots’ of the world to begin with), but in the embodied, enactive structure of experience: to imagine something, that is, is to simulate an embodied exploration of what one imagines.22 It is like ‘going through the motions’ of actual visual perception, but in the frame of ‘as if’ and with sensorimotor feedback (often called ‘resonances’ in this context) provided from memory rather than the environment. It is by definition multimodal, involving (like, on the enactivist account, the experience of basic agency itself) the whole sensorimotor array, including the exteroceptive modalities, dominated by vision, and proprioceptive and kinaesthetic modalities (a sensory awareness of bodily postures, movements and embodied aspects of emotions) (Kuzmičová 2012b: 24, 2013: 115). If mental images are involved, they are at best a by-product of our ‘imaginings’, are ‘undergone’ rather than inwardly ‘seen’, and need not specify more features than are relevant to the imagined experience at hand (Thompson 2007a: 156; Troscianko 2013: 186; Caracciolo 2014: 100). For the vividness of the imagination does not depend on the amount of detail ‘seen’, but on a distinct, bodily-perceptual ‘feel’ that is akin to that of actual visual perception. In the following sections I will discuss two complementary ways in which modern critics have applied enactivist insights to quasi-visual reader responses, and which seem to me to resonate with how ancient critics talk about enargeia. The first focuses on how texts can model the ‘probing’ structure of enactive visual experiences and so may prompt vivid (lifelike, intense, smooth) simulations of such experiences. The second is concerned with the story-driven, imaginative equivalent of the real-life phenomenon of ‘fascination’. In making the case for embodied reader responses I will be relying on recent linguistic research which emphasises the embodied aspects of language comprehension, such as that of Zwaan, who understands language as ‘a set of cues to the comprehender to construct an experiential (perception plus 21 22

e.g. Rizzolatti and Craighero 2004. For discussion, cf. Gallagher 2001: 101–3, 2008: 540–2. See Kuzmičová 2012b: 26; Caracciolo 2014: 98–100. Excellent treatments of ‘enactment imagination’ are Thomas 1999; Thompson 2007a. Evidence from eye-tracking experiments shows that imagining involves the same tracing of sensorimotor patterns as actual vision; see Spivey and Geng 2001; Johansson et al. 2006.



enargeia , ena ctiv ism an d t he readerl y i m ag i nat i o n 179

action) simulation of the described situation’, and comprehension as ‘the vicarious experience of the described situation’ (Zwaan 2004: 36). Increasingly, evidence is accumulating that the ‘depth’ of such embodied processing varies depending on the kinds of language used, and that readers’ responses to words through their bodies may lead to fully fledged bodily feelings which can be linked to a sense of ‘presence’ or ‘immersion’ in the storyworld.23

Enacting Narrative Space: Enargeia from ‘Seeing’ to ‘Looking’ When one starts thinking about concrete stylistic differences between the pictorialist and enactivist models of readerly visualisation, some of them come down to the issue of detail. The enactivist approach does not set much store by detailed mental images and even suggests that elaborate descriptions may be baffling rather than stimulating, because they contradict the economy and selectivity intrinsic to our normal enactive-perceptual engagement with the world.24 In fact, descriptions arguably become more vivid if they play on the fallibility of our ‘pictorial’ vision than if they try to fill in all the details (see Troscianko 2014: 144–59). It is interesting to note in this respect that in Cicero’s vignette of the banquet hall, cited by Quintilian as a star example of enargeia, the words ‘I seemed to see’ (videbar videre) indicate that, from the very start, the scene which is about to be described could not be clearly seen. There may, to be sure, have been good reasons to emphasise the point. Perhaps the first-person observer (Cicero himself?) was ashamed to look closely at so abject a scene and kept his eyes turned on the floor, or it was dawn and the first light only made visible the contours of the guests and a few gleaming wet patches on the floor. But for present purposes the relevant point is that to Quintilian’s mind the lack of a ‘clear view’ apparently did not stand in the way of enargeia. So, what could it have been about this passage that he found visually compelling? Part of the answer is, I think, that the passage is structured so as to enable readers to undergo an enactive visual exploration of the represented space. First, the words videbar videre inscribe an internal point of view into the text and suggest that the following description will to some extent be tailored to the body and sensory systems of an embedded human observer.25 To this we can add the present participles (‘coming in’ (intrantis) etc.), which configure the actions they express as ongoing and as currently being perceived, and the way in which the guests’ comings and goings are described in relation to the internal viewpoint.26 Secondly, the description 23

See Zwaan 2008, 2014; Kuijpers and Miall 2011. Sanford and Emmott (2012: 103–31) provide an overview of experimental research in this area. I will refer to some specialised studies below. 24 Cf. Jajdelska et al. 2010: 442–3; Kuzmičová 2012a: 309, 2012b: 36; Troscianko 2013: 188, 2014: 129. 25 Cf. Kuzmičová 2012a: 301–2, 2012b: 40; Troscianko 2014: 168; Caracciolo 2014: 101, 103–5. 26 Manieri (1998: 142) comments on ‘l’immediatezza’ of the present participles in our passage.

180

lu UK hui t i nk

does not simply enumerate sense-data, but conveys the impression that the scene is scanned by someone moving her eyes and head in accordance with the enactivist view of how we ‘look around’ and with the principle of selective visual attention; each of the asyndetic and anaphoric clauses (‘alios . . ., alios . . ., quosdam . . . quosdam . . .’) can even be understood as an instance of ‘looking’ preceded by movement, with the repeated words homing in on the entities to which the observer’s attention is drawn each time.27 The term ‘experiential iconicity’ has been coined for linguistic structures which are iconic, not of the represented entities, but of the perception of those entities (Wolf 2001). It is plausible that Quintilian was sensitive to those features of the passage on which an enactivist approach also homes in. First, embedded viewpoints and the linguistic means of shaping them are often associated with enargeia in our sources. A late Roman treatise on figures of thought actually defines enargeia in terms of the manipulation of deictic elements which establish an internal viewpoint. It commends a line of Vergil that uses the historical present (‘The men are visible (apparent) floating scattered on the vast waters’) for expressing the action ‘as if present’ (quasi praesenti), and one that uses proximal spatial deictics (‘Here (hic) the army of the Dolopes had its camp, here (hic) savage Achilles’) for making it seem as if the entities so pointed to are ‘within our field of vision’ (in conspectu nostro).28 The historical present is also repeatedly associated with enargeia in the scholia,29 and Ps.-Longinus (Subl. 25) praises Xenophon’s and Thucydides’ use of the device, because it makes it seem as if past events ‘were happening now and were present’ (ὡς γινόμενα καὶ παρόντα).30 Quintilian (Inst. 9.2.41–3) himself, finally, knows a figure called ‘transference of time’ (tralatio temporum) or metastasis, which is achieved either by prefacing a scene with explicit references to ‘seeing’ or by using the historical present (cf. Lausberg 1998: 363; Sheppard 2014: 35); this makes it likely that he felt that both ‘videbar videre’ and the present participles in the Ciceronian vignette contributed to enargeia. Furthermore, ancient authors seem to have appreciated the fact that ‘looking’ involves bodily movement. Polybius (3.38.5) significantly instructs his readers that, ‘just as in the case of actual visual perception (ἐπὶ τῆς ὁράσεως) we are used to always turning our faces (συνεπιστρέφειν ἀεὶ τὰ πρόσωπα) towards the object pointed out to us, so too should we mentally (τῇ διανοίᾳ) ever turn and shift our glance (συνδιανεύειν καὶ συρρέπειν) to the geographical locations to which the

27

Cf. Otto (2009: 118), who notes how the successive clauses seem to represent ‘die Bewegung der Augen’. For the general point, see Troscianko 2014: 125–6; Caracciolo 2014: 102. 28 Anon. Schem. Dian. 1 Halm, citing Verg. Aen. 1.118, 2.29. See Ryan (2001: 130–9) for a modern discussion of how the manipulation of spatiotemporal deixis contributes to ‘immersion’, and Allan (2013) for a recent analysis of the Greek historical present along these lines. 29 Meijering 1987: 42–3; Nünlist 2009: 196. Add Donatus, Comm. in Ter. Hec. 295: ‘ “obtrudit” pro “obtrudebat”: ἐνάργεια temporis’ (also Eun. 594, Hec. 174). 30 Note the similar language used by Plutarch of Xenophon; his account of Cunaxa is rich in historical presents.



enargeia , ena ctiv ism an d t he readerl y i m ag i nat i o n 181

story points’.31 The clearest indication that it was known that such movements could be mirrored in certain rhetorical figures and so enhance visual vividness comes in Ps.-Longinus’ analysis of a famous Demosthenic passage which describes the effects of a blow on the victim: πολλὰ γὰρ ἂν ποιήσειεν ὁ τύπτων, ὦ ἄνδρες Ἀθηναῖοι, ὧν ὁ παθὼν ἔνι’ οὐδ’ ἂν ἀπαγγεῖλαι δύναιθ’ ἑτέρῳ, τῷ σχήματι, τῷ βλέμματι, τῇ φωνῇ, ὅταν ὡς ὑβρίζων, ὅταν ὡς ἐχθρὸς ὑπάρχων, ὅταν κονδύλοις, ὅταν ἐπὶ κόρρης. ταῦτα κινεῖ, ταῦτ’ ἐξίστησιν ἀνθρώπους αὑτῶν, ἀήθεις ὄντας τοῦ προπηλακίζεσθαι. (Dem. 21.72) Many things, Athenians, some of which the victim would find it difficult to put into words, may be done by the striker – by gesture, by look, by tone; when he strikes in wantonness or out of enmity; with the fist or on the cheek. These are the things that provoke men and make them beside themselves, if they are unused to insult. Ps.-Longinus (Subl. 20.1–2) mentions ‘the asyndeta in combination with the cases of anaphora and the vivid description’ (ταῖς ἀναφοραῖς ὁμοῦ καὶ τῇ διατυπώσει συναναπεπλεγμένα τὰ ἀσύνδετα) as contributing to a sense that ‘the orator accomplishes the same as the assailant: he strikes the jurors’ minds with blow after blow’. Readers, then, are in the position of the victim of the attack and may experience what Ps.-Longinus tellingly describes as a ‘disturbance and co-movement of the mind’ (φορὰ ψυχῆς καὶ συγκίνησις). In my view, the term συγκίνησις is suggestive of how readers may imaginatively follow the virtual blows, encoded in the text, with their eyes in an effort to duck the blow; the addition ‘of the mind’ refers only to the fact that readers do not actually start turning left and right, but in all other respects what they feel is physical (reliant on sensorimotor resonances).32 While Ps.-Longinus does not discuss this example under the header of enargeia or phantasia, the effect which he describes could clearly be called that.33 The term διατύπωσις is a common synonym of enargeia (and apt here, as it is indicative of the almost physical impression which the virtual blows make), and we will see below that συγκίνησις is closely associated with phantasia in On the Sublime. Other critics also associate rhetorical figures like anaphora and asyndeton with enargeia.34 31

Cf. Meijering 1987: 47–8. Polybius seems to think of his readers as poring over a large map of the Mediterranean in their imaginations (rather than as ‘embedded’ observers). 32 I will return to this term in the next section. On kinēsis-language in ancient criticism, see Meijering 1987: 44–7; Nünlist 2009: 139–40. Both, however, play down its bodily connotations. Pace Russell 1964: 135: ‘συν- . . . merely emphasises the completeness or thoroughness of the disturbance.’ 33 Cf. Anon. Segu. 111, which gives the Demosthenic passage as an example of enargeia. 34 Hermogenes (Id. 303.1–8 Rabe, citing Dem. 19.10) uses the term to describe the effect of epanaphora kata komma, as in the Demosthenic phrase προσιὼν μὲν τῇ βουλῇ, προσιὼν δὲ τῷ δήμῳ (‘coming forward to the Council, and coming forward to the people’). See also Ps.-Demetrius (Eloc. 211–15) on the enargeia of dilogia or ‘figures involving repetition’, with the discussions of Manieri 1998: 134–5; Otto 2009: 80–3.

182

lu UK hui t i nk

Apart from these more or less explicit discussions of relevant stylistic features, it should be noted that examples of visually ‘vivid’ writing adduced in our sources often contain an internal point of view, tied to a character, explicit indications of bodily movement or travel – this maybe reflects a sense that in order to make readers consciously aware of the virtual movements on which vividness depends, it helps textually to specify (and exaggerate) them – and experiential iconicity.35 One relevant example is a celebrated passage from Demosthenes describing the destruction of Phocis: ὅτε γὰρ νῦν ἐπορευόμεθ’ εἰς Δελφούς, ἐξ ἀνάγκης ἦν ὁρᾶν ἡμῖν πάντα ταῦτα, οἰκίας κατεσκαμμένας, τείχη περιῃρημένα, χώραν ἔρημον τῶν ἐν ἡλικίᾳ, γύναια δὲ καὶ παιδάρι’ ὀλίγα καὶ πρεσβύτας ἀνθρώπους οἰκτρούς. (Dem. 19.65) So when we were now travelling to Delphi, necessity compelled us to look upon that scene – homesteads levelled with the ground, cities stripped of their defensive walls, a countryside all emptied of its young men; only women, a few little children, and old men stricken with misery. This passage was much praised for its visual ‘vividness’ in antiquity.36 However, like Cicero’s vignette, it is not rich in details, so that it is hard to see how its enargeia can be bound up with any ‘pictorial’ qualities. Rather, it is notable for combining a reference to human observers on the spot (ἦν ὁρᾶν ἡμῖν), a verb of movement (ἐπορευόμεθα) and experiential iconicity (short clauses and asyndeton).37 The enactivist framework helps us understand how these features contribute to eliciting vivid quasi-visual responses from readers and, I would maintain, why ancient critics were drawn to this passage when looking for examples of enargeia.

Enacting Bodily Movement: Enargeia from Pity to Fear There is one further aspect of Cicero’s vignette which contributes to its visual vividness, and that is that the internal observer’s attention is mostly drawn to aspects of the scene which, on an enactivist view, are cognitively salient. The focus on the floor, for instance, which at first sight seems strange (cf. Otto 2009: 119), in fact gets things just right, both because it highlights a crucial element of rooms, in that 35

For internal viewpoints, cf. Walker 1993; Otto 2009: 89; Webb 2016: 211–13 (though all three with a more ‘psychological’ slant than mine). For ‘travel’, see e.g. Ps.-Longinus, Subl. 26.2 on Hdt. 2.29; Σ bT Il. 14.226–7: the lines enumerate locations past and through which Hera moves on her flight from Olympus to Lemnos; the scholiast comments that ‘the mind (ἡ διάνοια) of the readers travels alongside her (συμπαραθέουσα) and is brought to imagine and visualise (ἐν φαντασίᾳ καὶ ὄψει) the locations’; cf. Meijering 1987: 44; Nünlist 2009: 188. 36 Cf. Webb 2016: 213. Cf. Nik. Progymn. 71.1–5 Felten; Σ Dem. 19.157c Dilts (misplaced from 19.65); Alex. De fig. III.25.13–25 Spengel; Anon. Segu. 111. 37 Also note the ‘presentifying’ deictic νῦν (‘now’), although all sources quote a version of the passage without this word.



enargeia , ena ctiv ism an d t he readerl y i m ag i nat i o n 183

their canonical affordances of reposing and moving about in them depend on it, and because it gives the internal observer whose visual perception the reader simulates a solid surface virtually to stand on, which probably increases the sense of ‘presence’.38 Furthermore, the initial focus on the banqueters’ comings and goings matches the fact that movements are a natural focal point of our selective visual attention and, as readers virtually follow the represented people around, it also evokes a sense of three-dimensional space more effectively than a disembodied description of the room’s dimensions would. Furthermore, since ‘entering’ and ‘exiting’ belong to the most common affordances of rooms, the narrative rendition of these actions is also likely to elicit from readers a more vivid sense of the presence of walls and doors than a detailed description of such furnishings: we know (perceptually, that is, immediately and experientially) how it feels to perform such actions and what they look like to an observer, because it is at the level of such actions that we ourselves perceptually-enactively experience the world and know others to experience it.39 Like seeing such bodily actions being performed in real life, imagining them involves a proprioceptive component; indeed, imagining actions, the vividness of which depends on one’s awareness of sensorimotor processes in one’s own body, may to some extent always cut through an inner-outer dichotomy. This helps us understand, I think, why the focus on bodily actions, which is so prominent a feature of Cicero’s vignette, may do more than just contribute to spatial vividness and also tell us something about what readers may get out of ‘seeing’ this scene. Take the example of the banqueters said almost to ‘keel over’ (vacillantis). As we have seen, on a pictorialist account, this is a ‘telling’ detail from which readers may infer certain things about the physical and psychological state of the people in question. However, the enactivist account of ‘smart’ vision allows for a different approach. On this view, readers do not infer the guests’ predicament as general knowledge, as a conceptual abstraction, but understand it experientially, by ‘incorporating’ the action involved, by catching, as it were, an echo of the described movement in their bones.40 Since this understanding is perceptual-enactive, immediate and prereflective, it can be experienced as visually vivid. This cognitive perspective can be supplemented by a linguistic one. When one thinks about the precise semantic nuances of bodily-action verbs – for instance, 38

For the desirability of stressing canonical affordances, see Kuzmičová 2012b: 31; Troscianko 2014: 137–8, 142; Cave 2016: 46–8. Cf. Kuzmičová 2012b: 28, 2013: 117: ‘The imaginary world is unlikely to feel tangible and present unless physical stimuli that can be interacted with are mentioned (or strongly implied), that is, unless the furnishing of the imaginary world is reached, grasped, manipulated, leaned against, and so forth.’ From a non-enactivist perspective, Scarry (1995) argues that indications of the ‘solidity’ of a room’s surfaces contribute markedly to its imageability. 39 See Grünbaum 2007: 307–9, esp. 308, on how the rendition of bodily movements ‘[i]n a direct and quasi-perceptible (ready for visual imagining) way . . . manifests a spatial situation, which as a whole is strung together by the familiar patterns of movement and orientation’. 40 I owe this formulation to Cave 2016: 63.

184

lu UK hui t i nk

about how ‘keeling over’ subtly differs from, say, ‘tottering’, ‘dropping down’ or ‘foundering’ – it becomes clear that these are difficult to capture in abstract terms; rather, it seems that we distinguish such verbs by their appearance (how they develop in space and time) and how they ‘feel’ in the body (Kemmerer 2006: 350; Cave 2016: 109–10). Bodily-action verbs have therefore been a prime focus of research into how language processing activates regions of the brain associated with perception and kinaesthesia (Zwaan 2009); motor resonances of motion verbs have been posited in readers and also observed directly in neuroimaging studies (Pulvermüller 2005). It has also been shown that readers simulate vertical and horizontal movements as expressed or implied by verb phrases (Richardson et al. 2003), as well as the orientation of objects as implied by the use to which they are said to be put.41 Now in the passage from Cicero, an outside perspective on the guests’ drunken shenanigans probably predominates and the enactment of the movement of ‘keeling over’ only has a modest and nearly imperceptible role to play in readers’ reception of the passage. There are, however, many indications in ancient criticism that ancient readers’ quasi-visual responses included the enactment of bodily actions described in the text.42 In general, it is remarkable how many examples of passages said to possess the quality of enargeia are ‘full of details of human action’,43 and how often readers are implicitly compared to the emotionally involved audiences of sports matches, perhaps the paradigmatic example of ‘fascinated’ spectators in the sense of Fuchs and De Jaegher (see ‘The Enactivist Account of Visual Experience and Imagination’ above).44 Indeed, ‘fascination’ often seems to me as appropriate a way as any to typify the ‘intensity’ of ancient readers’ quasi-visual engagement with literature. The clearest example is Ps.-Longinus’ treatment of phantasia. As we have seen in ‘Some Shortcomings of the Pictorialist Account’ above, according to this critic (Subl. 15.1–2) phantasia comes about ‘when under the influence of inspiration and emotion you seem to see what you describe and place it before the eyes of the readers’. He goes on to specify that phantasiai ‘seek emotion and excitement’ (τό τε ἐπιζητοῦσι καὶ τὸ συγκεκινημένον), and in poetry even aim at the stunning emotional impact he calls ekplēxis. I will return to these terms below. He next refers to several passages in support of these claims, the two most elabo41

Stanfield and Zwaan 2001; Zwaan and Madden 2005. Cf. Bolens 2012: 13: ‘If a text describes a person hammering a nail into a wall, readers simulate a horizontal nail, whereas the nail is simulated as vertical if it is said to be pounded into the floor.’ 42 Kuzmičová (2012b, 2013) speaks of ‘motor enactment’ to typify this response, Bolens (2012) of ‘kinesic empathy’, Caracciolo (2014: 160) of how readers may ‘imaginatively project themselves into the character’s fictional body’. 43 Webb 2009: 109; examples also in Calame 1991; Calboli Montefusco 2005. 44 The Iliadic horse race is commented on by Ps.-Demetrius (Eloc. 210) and various scholia; see Grethlein and Huitink 2017: 76–7. Quintilian (Inst. 8.3.63) gives the boxing match from Verg. Aen. 5.426ff. as his first example of enargeia; in a highly metaliterary passage, Heliodorus (4.3.4) describes the vivid effects of a story about a foot race on the audience; for analysis, see Grethlein 2015a: 270–2.

enargeia , ena ctiv ism an d t he readerl y i m ag i nat i o n 185



rate of which concern excerpts from Euripidean messenger speeches. The first is from Iphigenia in Tauris and relates how Orestes was (actually or in his imagination) attacked by Erinyes:45 κἀν τῷδε πέτραν ἅτερος λιπὼν ξένοιν ἔστη κάρα τε διετίναξ’ ἄνω κάτω κἀπεστέναξεν ὠλένας τρέμων ἄκρας, μανίαις ἀλαίνων, καὶ βοᾷ· ‘Κυνωπίδα, Πυλάδη, δέδορκας τήνδε; τήνδε δ’ οὐχ ὁρᾷς Ἅιδου δράκαιναν, ὥς με βούλεται κτανεῖν δειναῖς ἐχίδναις εἰς ἔμ’ ἐστομωμένη; ἣ ’κ γειτόνων δὲ πῦρ πνέουσα καὶ φόνον πτεροῖς ἐρέσσει, μητέρ’ ἀγκάλαις ἐμὴν ἔχουσα, πέτρινον ὄγκον, ὡς ἐπεμβάλῃ. οἴμοι, κτενεῖ με· ποῖ φύγω; (Eur. IT 281–91) Meanwhile one of the two strangers left the cliff and stood there and tossed his head this way and that. He groaned, his arms trembling to his fingertips, swaying because of the onslaughts of madness, and shouted, ‘Pylades, don’t you see this hound-faced one? And don’t you see this one, a she-dragon from hell, how she wants to kill me, brandishing her fearsome snakes at me? And next to her, one breathing fire and gore beats her wings and holds my mother in her arms, a mass of stone, to hurl at me! Ah, she will kill me! Where can I escape to?’ In response to this passage, Ps.-Longinus writes: ἐνταῦθ’ ὁ ποιητὴς αὐτὸς εἶδεν Ἐρινύας· ὃ δ’ ἐφαντάσθη, μικροῦ δεῖν θεάσασθαι καὶ τοὺς ἀκούοντας ἠνάγκασεν. (Subl. 15.2) Here the poet himself saw Erinyes and as good as compelled his readers to see what he imagined. The second is from Euripides’ Phaethon and concerns the eponymous character’s ill-fated journey through the heavens; the longest continuous portion starts with a line of direct speech containing Zeus’ final instructions and continues with the start of Phaethon’s journey:

45

In fact, Ps.-Longinus (Subl. 15.2) first cites Eur. Or. 255–7, another, and much briefer, vision of Erinyes, which awards them only two descriptive adjectives (αἱματωποὺς καὶ δρακοντώδεις), and then IT 291 (οἴμοι . . . φύγω;). However, ancient quotation practices make it likely that he had the whole passage in view (or at least the entirety of Orestes’ directly quoted speech), and this likelihood is increased because of the structural similarities (of which I will come to speak) between this passage and the one from Phaethon cited below.

186

lu UK hui t i nk

‘ἵει δ’, ἐφ’ ἑπτὰ Πλειάδων ἔχων δρόμον.’ τοσαῦτ’ ἀκούσας παῖς ἔμαρψεν ἡνίας· κρούσας δὲ πλευρὰ πτεροφόρων ὀχημάτων μεθῆκεν, αἱ δ’ ἔπταντ’ ἐπ’ αἰθέρος πτύχας. πατὴρ δ’ ὄπισθε νῶτα Σειρίου βεβὼς ἵππευε παῖδα νουθετῶν· ‘ἐκεῖσ’ ἔλα, τῇδε στρέφ’ ἅρμα, τῇδε.’ ἆρ’ οὐκ ἂν εἴποις, ὅτι ἡ ψυχὴ τοῦ γράφοντος συνεπιβαίνει τοῦ ἅρματος καὶ συγκινδυνεύουσα τοῖς ἵπποις συνεπτέρωται; οὐ γὰρ ἄν, εἰ μὴ τοῖς οὐρανίοις ἐκείνοις ἔργοις ἰσοδρομοῦσα ἐφέρετο, τοιαῦτ’ ἄν ποτε ἐφαντάσθη. (Subl. 15.4, citing Eur. Phaethon, fr. 779 Kannicht) ‘ “Steer steadily towards the seven Pleiads.” The boy listened so far, then seized the reins, whipped the flanks of his winged team and let them go. To heaven’s expanse they flew. His father rode behind on Sirius’ back, giving the boy advice: “That’s your way, there: turn the chariot this way, now this way.”’ Would you not say that the soul of the author also mounts the chariot and, sharing in the danger, takes wing with the horses? For if it had not been carried along keeping pace with those actions in the heavens, he would never have imagined such things. Ps.-Longinus’ response to the Erinyes passage has been thought to give expression to the pictorialist idea that verbal descriptions transfer mental images from author to reader.46 However, the passage does not seem particularly suitable for pictorial visualisation, because it does not offer sufficient descriptive detail for readers to be able to piece together a clear and complete mental image of the Erinyes. It has therefore been argued that the effect of the passage is as much semantic as it is sensory: it depends not so much on readers ‘seeing’, say, a snake-like figure (δράκαιναν) in the mind’s eye (such a thing may well be difficult to imagine anyway) as on them having bad associations with the word ‘snake’.47 The Phaethon passage, which is similarly lacklustre from a pictorialist point of view, has also been thought to be interesting mostly for presenting a glimpse of the divine, a ‘transcendent vision’, designed to ‘help us to understand phenomena that are beyond the normal grasp of an individual human’s perception’ (Webb 2016: 218). In my view, such explanations fail to capture the essence of what is going on here. While neither passage stands out because of its reflections on the supernatural phenomena they describe, they share an emphasis on the simple narration of bodily actions: thus, Phaethon snatches (ἔμαρψεν) the reins, whips (κρούσας) the 46

Cf. Webb 2009: 96; Otto 2009: 97–8. We would prefer to say that Orestes had the vision, but ancient critics did not usually distinguish between poet and character. 47 Otto 2009: 98: ‘Die Beschreibung hat einerseits einen visuellen Aspekt, andererseits aber auch eher einen assoziativen . . . schlangenähnliche weibliche Gestalten . . . sind nicht gut visualisierbar, dafür ist das Wort Schlange mit schlechten Assoziationen besetzt.’

enargeia , ena ctiv ism an d t he readerl y i m ag i nat i o n 187



horses’ flanks and then sets them loose (μεθῆκεν); Orestes gets up (ἔστη), tosses his head up and down (κάρα τε διετίναξ’ ἄνω κάτω), trembles to his fingertips (ὠλένας τρέμων ἄκρας) and hardly keeps his balance as his body precariously ‘sways’ (ἀλαίνων). I suggest that immersed readers may be led imaginatively to enact these movements, especially because in both passages the sensorimotor resonances provoked by the verb phrases are reinforced by characters’ verbalisations of perceptual-enactive experiences in direct speeches, which further encourage readers to enact the scenes from an internal perspective. The jerky movements of Orestes’ head and eyes are continued in the phrase he addresses to Pylades: ‘Do you see this one? And this one, don’t you see her?’ (δέδορκας τήνδε; τήνδε δ’ οὐχ ὁρᾷς;). The progress of Phaethon’s flight is charted in the instructions he is given by Zeus: ‘That’s your way, there: turn here, turn here’ (ἐκεῖσ’ ἔλα, | τῇδε στρέφ’ ἅρμα, τῇδε). Both lines combine instructions to turn one’s head to ‘see’, proximal spatial deictics and verbal repetition of these deictics, with an added asyndeton in the example from Phaethon. The enactivist sensorimotor resonance framework seems to me to fit Ps.Longinus’ markedly non-pictorialist comments on these passages. It is in accordance with his identification of ekplēxis as the chief effect of the cited phantasiai, which according to Russell expresses a ‘surprise or fear which “knocks you out”’ (Russell 1964: 122); the language of compulsion (ἠνάγκασεν) Ps.-Longinus uses in response to Orestes’ vision implies that ekplēxis is neither entirely voluntary nor wholly open to rational scrutiny: ‘the more the poet succeeds in visualizing and transmitting the inner subjectivity of Orestes’ frenzied mental state, the more the audience’s experience . . . will itself approximate a kind of madness.’48 Ekplēxis, then, appears to consist in a largely spontaneous (even irresistible), close emotional identification with the represented characters, Orestes and Phaethon. Furthermore, the term τὸ συγκεκινημένον in Ps.-Longinus’ introductory remark, which literally means ‘movement with/alongside’, suggests that this identification comes about, quite precisely, by (virtual) co-movements with the represented characters which may accompany a state of ‘fascination’. The term is picked up in Ps.-Longinus’ comments on the Phaethon passage, where the author’s (and by implication the reader’s) soul is said to mount the chariot with Phaethon (συνεπιβαίνει), to run risks with him (συγκινδυνεύουσα) and to take wings with the horses (συνεπτέρωται). As in the previous section, I should like to argue that we should not be too quick to dismiss the kinēsis-language as mere metaphor; the response which Ps.-Longinus is after includes, I think, a felt sense of the visceral emotions which come with the imaginative enactment of the described movements, a stomach-churning, sickening fear (in the case of Orestes) or exhilaration (in the case of Phaethon). Halliwell 2011: 348. It is true that the verb used for the audience ‘seeing’ the Erinyes, θεῶμαι, often carries connotations of intellectual contemplation (Liddell et al. 1996: s.v. I.2), but I suspect that it was selected here because of its connection to θεατής (‘spectator’), which plays on the enargeia-trope of turning readers into emotionally involved spectators ‘on the spot’ (cf. the Homeric scholion cited in the ‘Introduction’ above).

48

188

lu UK hui t i nk

I end this section by drawing attention to a problem in the interpretation of Quintilian. We have seen in ‘Some Shortcomings of the Pictorialist Account’ above how Quintilian, in the context of a discussion of enargeia and the arousal of emotions, regards the human capacity to visualise a scene as so powerful that at times people do not merely seem to ‘think’ (cogitare), but even to ‘do’ (facere) what they imagine; he then mentions travelling, sailing, fighting, addressing a crowd and disposing of wealth as examples – schoolboy daydreams indeed! He suggests that budding orators can easily learn to harness this capacity for rhetorical purposes and offers the outlines of a riveting murder story to show how they may do so: hominem occisum queror; non omnia, quae in re praesenti accidisse credibile est, in oculis habebo? non percussor ille subitus erumpet? non expavescet circumventus, exclamabit vel rogabit vel fugiet? non ferientem, non concidentem videbo? non animo sanguis et pallor et gemitus, extremus denique exspirantis hiatus insidet? insequetur ἐνάργεια, quae a Cicerone inlustratio et evidentia nominatur, quae non tam dicere videtur quam ostendere, et adfectus non aliter, quam si rebus ipsis intersimus, sequentur. (Quint. Inst. 6.2.31–2) Suppose I am complaining a man has been murdered. Am I not to have before my eyes all the circumstances which one can believe to have happened during the event? Will not the assassin burst out all of a sudden, and the victim tremble, cry for help, and either plead for mercy or try to escape? Shall I not see one man striking the blow and the other man falling? Will not the blood, the pallor, the groans, the last gasp of the dying be imprinted on my mind? The result will be enargeia, which Cicero calls inlustratio and evidentia, and which makes us seem not so much to be talking about something as exhibiting it. Emotions will ensue just as if we were present at the events themselves. (trans. Russell 2001) The problem is that the introductory remarks appear to presuppose a first-person perspective, while the story which follows is in the third person. Webb therefore felt that the ‘sense of empathy created by the simulacrum of presence is twofold: the orator puts himself in the place of an eyewitness but it is the suffering of the victim that is clearly evoked’ (Webb 2016: 210). The enactivist view allows us to see that there is not necessarily a contradiction: an immersed reader will both ‘see’ and ‘enact’ the actions described in the scene. This has consequences for the emotions (adfectus) which Quintilian thought would automatically ensue. What Quintilian thinks readers will feel is in the first place, I feel, the same sort of fear as the victim, which on an enactivist account is part and parcel of the perceptual experience and in that sense indeed automatic. This does not mean, of course, that in a second instance readers may not also feel pity for the victim and anger with the perpetrator, just like readers of Cicero’s vignette may come to reflect on the behaviour of the drunken guests or readers of Euripides may come to feel an emotional distance from Orestes in his madness or Phaethon in his foolishness. But I would suggest



enargeia , ena ctiv ism an d t he readerl y i m ag i nat i o n 189

that such responses comprise a separate, higher-order cognitive act which does no longer come under the purview of enargeia per se.

Conclusion This chapter started out with a critical review of the prevailing pictorialist account of enargeia. It suggested that this account could not properly explain a number of important stylistic and experiential features of the concept which are repeatedly, if sometimes allusively, flagged up in our sources. It therefore developed an alternative enactivist perspective on enargeia, which is able to account for some important properties of enargeia. Thus, on the stylistic side of things, we have seen that certain non-pictorial stylistic features, such as asyndeton and anaphora, do make a significant contribution to visual vividness and were in all probability regarded by ancient critics as being conducive to enargeia. It has also been shown that there is a positive connection between enargeia and narrative, especially the narration of bodily actions. On the experiential level, the enactivist view of readers’ vivid quasi-visual responses suggests that they always entail an embedded (internal) view of the represented scene and may also elicit a felt sense of bodily involvement, and this is again in accordance with the way in which ancient critics talk about enargeia. Finally, I have argued that enargeia in the first place brings with it an emotional and experiential grasp of what is going on, as opposed to a reflective, intellectual understanding. This, too, fits the way in which ancient critics talk about readers’ quasi-visual responses. I should like to emphasise that I do not claim that an enactivist approach explains everything about enargeia. But if it is the case that readers’ quasi-visual responses to literature originate from a combination of biological and cognitive invariables (which I would configure as enactivist) and socioculturally determined variables (which will have included pictorialist preconceptions), then it is both legitimate and useful to use a ‘second-generation’ cognitive approach to arrive at a more refined picture of what ‘vividness’ entailed for ancient readers. As Kukkonen and Caracciolo state, ‘[h]istorical practices and embodiment as a biological and cognitive condition are not opposed but, on the contrary, caught in a dialectic relationship, so that exploring the background of bodily invariants can improve our understanding of historical specificities, and vice versa’ (2014: 266). It is up to future research into ancient readers’ responses to text to investigate the relationship between the enactivist strands of enargeia laid bare in this chapter, on the one hand, and the folk-psychological and philosophical pictorialist theories and the practice of ekphrasis on the other.

190

f elix budel m ann

11 Gro up M i n ds i n C la s s ica l Athens ? C h o rus a n d D ēm o s a s C a s e Stu d ies of C o l l e ct i v e Cog nition Felix Budelmann

Cognition has reached collectives. It is no longer just sociologists and historians who are trying to understand how groups act, decide, believe or remember. Literary critics of all periods have begun in earnest to study forms of collective cognition in narrative texts.1 ‘Emotional contagion’, the notion that individuals ‘catch’ the emotions of those around them, has become a resonant idea in several disciplines.2 Evolutionary psychologists have been asking with renewed intensity how humans evolved to cooperate on an altogether different scale from primates and why it is that they are so much more ‘group-minded’.3 In the research that inspired the ‘History of Distributed Cognition’ project, it has in particular been philosophers who are interested in the idea of a ‘group mind’. In common-sense terms, it is evident that it is individuals who have minds, not groups. Indeed, social psychologists have long been making a point of critiquing Gustave Le Bon’s classic Psychologie des foules, according to which individuals who come together in a crowd behave as though ‘hypnotised’ (‘comme chez l’hypnotisé’) and are possessed ‘of a sort of collective mind’ (‘une sorte d’âme collective’, Le Bon 1960 [1895]: 31–2, 2). A body of experimental research has shown that group membership, even when temporary, does indeed affect people’s thinking, sometimes strikingly so, but that individuals remain capable of individual thought and are able to distance themselves from the thinking of other members of the group.4 Arguably, however, the plausibility of the idea of the group mind never rested on common-sense empiricism. Certainly in recent philosophy of mind, the term has proved useful in the context of the wider debates about what is gained and lost if the seat of cognition is conceived as broader than the individual, isolated brain. A growing number of philosophers have been discussing whether and how  1

See esp. Palmer 2010, and two special journal issues prompted by Palmer’s work, Style 45.2 (2011) and Narrative 23.2 (2015). The latter includes one article on ancient texts, Grethlein 2015b.  2 For an overview from a social psychology perspective, see Kelly 2001.  3 The evolution of ‘group-mindedness’: e.g. Tomasello 2014, with the review by Heyes 2014.  4 For overviews, with criticism of Le Bon, see Reicher 2001 and, from a classicist’s perspective, Fagan 2011: 83–93. A particularly important experiment is Reicher and Haslam 2006.



gr ou p min ds in class i cal at h ens ? 191

cognition may usefully be conceptualised not just as embodied, or as a system that involves objects such as a note pad or a smartphone on to which some of the thinking is ‘offloaded’, but also as distributed among a set of individuals. Much of the debate turns on the question of how strong a model of the group mind is most plausible. We are all familiar with statements such as ‘Microsoft changed its mind’. Should one interpret such a phrase as an attribution of mental activity to the corporation as a whole, or is it merely a metonymy? The choice is symptomatic of two fundamentally different approaches to theorising cognition among a group of individuals. ‘Methodological collectivists’ start with the group. According to this position, one has to find a way of analysing agency and rationality at group level if one wants to take corporate responsibility and decision-making seriously. The group, for them, has a mind. By contrast, ‘methodological individualists’ argue that collective cognition can without significant loss be disaggregated into its components and prefer to see the group as the sum of individuals who are engaged in a shared endeavour. The former position has the advantage that it takes seriously the fact that the change of direction took place at company level and that not each Microsoft employee changed his or her mind, but introduces a non-standard and therefore more difficult notion of ‘mind’. Robert A. Wilson puts the trade-off clearly (2005: 235): We often naturally talk of groups of people as having intentions, beliefs, goals, and memories. My claim is that what such talk does is add intelligibility to the actions of those groups, and a greater sense of their place in the causal and normative nexus, through their assimilation to human agents. Without the attribution of cognitive agency, the agency of groups can seem puzzling, unclear, and even mysterious. Cognitive agency removes this puzzlement, unclarity, and mystery. One of my main points, however, and one to which there is much sensitivity in the contemporary literature . . ., is that the appeal to cognitive agency at the group-level – to, for example, ‘collective memory’ – can create its own puzzles and mysteries. Needless to say, the debate is less binary than I have made out. A recent encyclopaedia entry, for example, operates a tripartite distinction: ‘coordinated cognition’, ‘collaborative cognition’ and (the most demanding) ‘joint cognition’, and even that is too schematic. Much of the interest of the debate lies in the different accounts of collective cognition it produces and puts to the test. Nevertheless, the basic distinction between methodological collectivism and individualism is important and will turn out to be productive in thinking about classical Athens.5 Groups played an important role in Athens, culturally as well as politically, and this chapter asks how the philosophical concept of the group mind fits two  5

The encyclopaedia entry in question is Wilson and Theiner 2013. For a book-length survey, which itself argues in favour of the view that ‘certain groups are genuine intentional agents’, see Tollefsen 2015 (quotation from p. 4). See also n. 22 below, as well as p. 206 on Rupert 2014.

192

f elix budel m ann

of the most prominent of these groups, the chorus on the dramatic stage and the decision-making populace, the dēmos, in the assembly. Secondly, and vice versa, it asks what the chorus and the dēmos may have to offer to those interested in the idea of group minds.6 In keeping with the conceptual focus of the philosophical literature on which I draw, my primary concern here is not with realities; only in the final section, and only in passing, will I talk about the experience of being in the chorus or the assembly. Rather, I am interested in how Athenian authors thought about the mental activity of these groups. A pretentious way of putting it (far too pretentious for what will be a brief and necessarily cursory discussion of a large and complex topic) would be that this is a study in the history of thought.

The Default: Unitary and Unanimous Groups The single most important observation to make, which will form the backbone of the chapter, is that both the chorus and the dēmos are by default treated as unitary entities. There are significant exceptions, but the normal unit of thought is the collective, not the individuals who comprise the collective. The Athenian citizen body can be spoken of in both the plural and the singular. Historiographers say regularly that ‘the Athenians’ did this or decided that, and orators frequently address the gathered assembly as ‘Athenians’ and such like. When the language of dēmos is used the group becomes a body in the singular. The following passage is one of dozens: ‘For the dēmos realises that it gains more benefit by not holding such offices itself and, instead, leaving them for the most able to hold’ (γιγνώσκει γὰρ ὁ δῆμος ὅτι πλείω ὠφελεῖται ἐν τῶι μὴ αὐτὸς ἄρχειν ταύτας τὰς ἀρχάς, ἀλλ’ ἐᾶν τοὺς δυνατώτατους ἄρχειν, Old Oligarch 1.3, trans. Marr and Rhodes 2008, adapted). There is nothing remarkable about this dual usage; in English, too, singular ‘America’, ‘the American people’ and ‘the Americans’ co-occur in political language, even though the respective nuances are not quite the same as in Greek. What, however, is remarkable is the strictly unitary nature of the dēmos. The phrase from the Old Oligarch is typical. He presents a unitary dēmos squaring off against the (variously conceived) leaders, rather than different members of the dēmos pitted against one another. The well-established two different meanings of dēmos (other than ‘deme’) are interesting in this context. When a conservative writer, such as here the ‘Old Oligarch’, wants to talk about the lower strata, he does not split the dēmos into different classes, but uses the term dēmos to refer to the ‘common people’. The dēmos remains unitary. A useful illustration of the representation of the dēmos as unitary and in unison is provided by the language of decrees. The bestowal of honours on the loyal Samians in 403–2 bce may serve as an example (IG II2 1.40–55 = GHI 2 §2, 403–2 bce, trans. Osborne and Rhodes 2003, adapted).

 6

Chorus and dēmos are juxtaposed in much more detail in a book-length recent study, Visvardi 2015. Her main focus, however, is on group emotion.



gr ou p min ds in class i cal at h ens ? 193 [ἔδοξεν τῆι βολῆι καὶ τῶι δήμωι· Πανδ]ιονὶς ἐπρυτάνευε, Ἀγύρριος Κ[ολλυτ]εὺς | [ἐγραμμάτευε, Εὐκλείδης ἦρχε, Κα]λλίας Ὤαθεν ἐπεστάτει· Κηφισοφῶν [εἶπεν]· | [ἐπαινέσαι μὲν Σαμίος ὅτι ἐσὶν] ἄνδρες ἀγαθοὶ περὶ Ἀθηναίος, καὶ ἅπ[αντα] | [κύρια ε῏ναι ἃ πρότερον ὁ δῆμος] ἐψηφίσατο ὁ Ἀθηναίων τῶι δήμωι τῶι Σα[μίων]. . . . Κηφισωφῶν | [εἶπε· τὰ μὲν ἄλλα κ]αθάπερ τῆι βολῆι· ἐψηφίσθαι δὲ Ἀθηναίων τῶι δήμωι κύρια | [ε῏ναι τὰ ἐψηφισ]μένα πρότερον περὶ Σαμίων καθάπερ ἡ βολὴ προβολεύσασα | [ἐς τὸν δῆμον ἐσ]ήνεγκεν . . . The council and the dēmos decided. Pandionis was the prytany. Agyrrhios of Kollytos was secretary. Eukleides was archon. Kallias of Oa was chairman. Kephisophon proposed: To praise the Samians because they are good men with regard to the Athenians; and everything shall be valid which previously the dēmos of Athens voted for the dēmos of Samos . . . Kephisophon proposed: in other respects in accordance with the council; but the Athenian dēmos shall vote that there shall be valid what was voted previously concerning the Samians, as the council in its probouleuma brought before the dēmos . . .

The phrasing describing the decision-making is formulaic. Both the opening words, recording the decision as one of both the council (which acts as a probouleutic body) and the dēmos, and the phrase introducing Kephisophon’s proposal occur in numerous inscriptions.7 We learn that the dēmos enacted the decree as formulated by the council, except for a rider proposed by Kephisophon. What is not recorded – and this is standard – is the votes, or indeed any aspect of the debate that led up to the voting.8 Factually, the vote was taken by those members of the dēmos who were present on the day but the language of decrees treats those present, in synecdoche, as the whole dēmos and treats them as unitary. We know that Kephisophon persuaded the assembly to adjust the council’s formula, but we do not know how his proposal came about, what sort of discussion he intervened in and how the proposal was carried. The commitment to unity has of course an ideological dimension. Even though debate was fundamental to Athenian democracy, so was unanimity. Two terms appear repeatedly in our texts, ὁμόνοια (‘sameness of mind’, ‘unanimity’) and μία γνώμη (‘single opinion’). Both occur together in this passage, for example, from the orator Dinarchus: ‘How then shall we be of one mind, Athenians? How shall  7

In fact they are here restored on the basis of their formulaic nature. On these formulae, see Rhodes 1972: ch. 2.  8 Votes were recorded only after secret ballots; on the evidence for secret ballots, see most recently Canevaro 2018. Standard shows of hands (cheirotonia) were probably estimated rather than counted; see Hansen 2004. In general on the selective, and as a result depoliticising, nature of decrees, see Osborne 1999.

194

f elix budel m ann

we all agree upon public interests when the leaders and demagogues take bribes and betray their country’s interests . . .’ (πῶς οὖν μίαν γνώμην ἕξομεν ὦ Ἀθηναῖοι, πῶς ὁμονοήσομεν ἅπαντες ὑπὲρ τῶν κοινῇ συμφερόντων, ὅταν οἱ ἡγεμόνες καὶ οἱ δημαγωγοὶ χρήματα λαμβάνοντες προίωνται τὰ τῆς πατρίδος συμφέροντα . . ., 1.99). The ideal explicitly formulated in the orators and historians and implicit in the language of decrees is not one of majority decision-making but of a dēmos deciding with one mind.9 Our historical narratives, too, tend to treat the dēmos as operating as a unit (with exceptions that I shall discuss below). Herodotus, when narrating disagreement, for example in the debate over the wooden walls (7.142–3), presents it as disagreement among speakers more than within the dēmos and he points to homogeneity in statements such as this, of the effect of Aristagoras’ speech before the dēmos: ‘Apparently it is easier to mislead a crowd (“many”) than to mislead an individual, for Aristagoras, who had failed to impose upon [the Spartan] Cleomenes, succeeded with 30,000 Athenians’ (πολλοὺς γὰρ οἶκε εἶναι εὐπετέστερον διαβάλλειν ἢ ἕνα, εἰ Κλεομένεα μὲν τὸν Λακεδαιμόνιον μοῦνον οὐκ οἷός τε ἐγένετο διαβάλλειν, τρεῖς δὲ μυριάδας Ἀθηναίων ἐποίησε τοῦτο, 5.97, trans. De Sélincourt 2003). Thucydides likes to portray the dēmos as fickle, but the dominant pattern is one of the whole dēmos changing its view this way and that (though see n. 18 below). Xenophon offers a particularly neat show of unanimity when he casts his army as a democratic city: ‘ “Anybody in favour”, he said, “should raise his hands.” And they all raised their hands’ (καὶ ὅτῳ δοκεῖ ταῦτ’, ἔφη, ἀνατεινάτω τὴν χείρα. καὶ ἀνέτειναν ἅπαντες, Anabasis 3.2.9).10 The fundamental structure of Athenian politics as understood in our texts was one of (harmonious or troubled) relationships between leaders and a unitary dēmos, not one between factions, parties or individuals within the dēmos. Mutatis mutandis, the chorus presents a similar picture. Both singular and plural language is used. Choruses themselves waver between ‘I’ and ‘we’ in notoriously unpredictable ways; other characters too employ both singular and plural in speaking to and of the chorus.11 Like the dēmos, then, choruses can be thought of both as a set of people and as a single entity. However, they too are by default undivided. We know less about the melodies and choreographies than we would like, but it is widely assumed that dramatic, as indeed non-dramatic, choruses normally sang in unison and danced in a tightly coordinated fashion. Unlike individual actors, some of whom were stars who toured several cities and who therefore may have had limited joint rehearsal time, the citizen choruses rehearsed as a group, intensely and at length. While tragedy in Athens arguably provides less scope for ensem 9

On the ideal of homonoia, see e.g. Ober (1989: 297–9), who lists relevant passages. On the extent to which this ideal reflects reality, see p. 207 below. 10 On Xenophon’s army as a city, see e.g. Hornblower 2011. Xenophon’s notion has a precedent in the Achaean army at Troy, which has often been read in political terms; see e.g. Hammer 2002. 11 For documentation, see Kaimio 1970.



gr ou p min ds in class i cal at h ens ? 195

ble acting than some more recent forms of drama, the chorus is the chief exception.12 Choruses rarely divide into individual voices. When they do, the practice is temporary as well as marked; this will be the subject of the next section. Somewhat more frequently, we encounter pairs of half choruses, though those too usually (re-) unite eventually.13 The most sustained splitting-out concerns the chorus leader, who can speak for the chorus as a whole. Interestingly, this is another form of synecdoche rather than a division and thus underlines the cohesiveness of the dramatic chorus. It is a reasonable hypothesis, almost universally shared, that trimeter lines are usually spoken by the koryphaios rather than the whole chorus, but notably this hypothesis rests solely on assumptions about more and less successful modes of performance. Perhaps because the role of the leader-figure is given to individual characters, the texts themselves hardly ever present a chorus-internal leader/group configuration; the lines belong, simply, to the chorus. (Contrast Alcman’s First Partheneion, in which the chorus address their leaders.) Similarly, we learn from Aristotle that in a chorus there was a παραστάτης, the chorus member positioned next to the leader (Politics 3, 1277a11–12), but again nothing in the dramatic texts indicates an internally structured chorus. A neat expression of this emphasis on coherence is the absence of individualisation. No chorus member in tragedy bears a name. In comedy and satyr plays we occasionally meet with names: charcoal-y names (Aristoph. Acharnenses 609–12), old-men names (Aristoph. Lysistrata 254ff.), satyr or dog names (Soph. Ichneutai, fr. 314.183–4 and 194 Radt), less clearly connotated, possibly real, names (Aristoph. Wasps 230–4, Eccl. 293) and bird species (Aristoph. Birds 297–304).14 But not only is this phenomenon absent from tragedy, but in the majority of cases individualised naming occurs at the moment of entry, as the individuals are melding into a chorus. The corresponding opposite phenomenon occurs on the Pronomos Vase. Here the members of a satyr chorus are individually named (presumably the actual names of the performers) as they doff their masks at the end – as they are ceasing to be a chorus.15 Such naming at the beginning or end of the performance further underlines the unitary nature of the chorus in mid-flow. Costuming, too, contributed to the sense of unity. The satyr chorus on the Pronomos Vase, the Basle dancers in their military get-up and what other iconographic indication we have of choral outfits suggest that near-identical costuming

12

For what little we know of the rehearsal process, see Revermann 2006a: 87–94. The most marked exception is the surviving version of the end of Aeschylus’ Seven Against Thebes. Lammers 1931 provides a survey of split choruses in tragedy. 14 For discussion of the birds chorus, see Dunbar 1995 ad loc. and ad 267–32. See Wilson 1977 for further evidence of individualised comic choruses, and Sutton 1985 for satyr choruses. Φίντων at Aesch. fr. 47a.802, interpreted by Sutton as a personal name, is probably Doric for φίλτων (‘darling’); see Krumeich et al. 1998 ad loc. 15 The names are controversial: see, respectively, Taplin, Wilson (actual names), Junker and Osborne (fictional names) in Taplin and Wyles 2010. 13

196

f elix budel m ann

was standard.16 This is not to say, of course, that there was no space for variation. The masks on the Pronomos Vase, while of a type, are not wholly identical, and some comic choruses, notably that of Birds, are likely to have worn individualised outfits. Much of the attraction of watching a chorus lies in the artful coordination of a set of individuals. For that reason alone it would be mistaken to deny all possibility of diversity and individuality in the dramatic choruses. For all we know there may have been more diversity, visually or aurally, than we can grasp in our evidence. Yet there can be no doubt that the general picture we are getting of Athenian dramatic choruses is one of a unanimous group with only limited internal structuring.

The Exceptions that Illustrate the Norm: Divided Groups One step further from the norm than temporarily individualised choruses are instances of division and disagreement. Such instances are always portrayed as a problem and confirm the unitary default ex negativo. The most celebrated moment of choral division in our tragic corpus is the elders’ reaction to the death cries of Aeschylus’ Agamemnon (1346–71).17 The opening couplet (1346–7), couched in trochaic tetrameters to convey excitement, and possibly the final couplet (1370–1), which like the rest of the passage is in iambics, are presumably spoken by the chorus leader, while the twelve individual members of the chorus have a pair of lines each (text and trans. Sommerstein 2009, trans. adapted): τοὔργον εἴργασθαι δοκεῖ μοι βασιλέως οἰμώγμασιν· ἀλλὰ κοινωσώμεθ᾿ ἤν πως ἀσφαλῆ βουλεύματ᾿ ᾖ. –  ἐγὼ μὲν ὑμῖν τὴν ἐμὴν γνώμην λέγω,     πρὸς δῶμα δεῦρ᾿ ἀστοῖσι κηρύσσειν βοήν. –  ἐμοὶ δ᾿ ὅπως τάχιστά γ᾿ ἐμπεσεῖν δοκεῖ     καὶ πρᾶγμ᾿ ἐλέγχειν σὺν νεορρύτῳ ξίφει. –  κἀγὼ τοιούτου γνώματος κοινωνὸς ὢν     ψηφίζομαι τὸ δρᾶν τι· μὴ μέλλειν δ᾿ ἀκμή. –  ὁρᾶν πάρεστι· φροιμιάζονται γὰρ ὡς     τυραννίδος σημεῖα πράσσοντες πόλει. –  χρονίζομεν γάρ· οἱ δὲ τῆς μελλοῦς κλέος     πέδον πατοῦντες οὐ καθεύδουσιν χερί. –  οὐκ οἶδα βουλῆς ἧστινος τυχὼν λέγω·     τοῦ δρῶντός ἐστι καὶ τὸ βουλεῦσαι †πέρι†. –  κἀγὼ τοιοῦτός εἰμ᾿, ἐπεὶ δυσμηχανῶ     λόγοισι τὸν θανόντ᾿ ἀνιστάναι πάλιν. 16

For a review of the evidence, see Csapo (2010: 5–23), who repeatedly points to the near-identity of costumes. 17 Similarly interesting is the corresponding moment in Eumenides, connecting the chorus and the dēmos, when the group of jurors – not properly a chorus – silently cast their individual votes.



gr ou p min ds in class i cal at h ens ? 197 –      –      –      –      –     

ἦ καὶ βίον τείνοντες ὧδ᾿ ὑπείξομεν δόμων καταισχυντῆρσι τοῖσδ᾿ ἡγουμένοις; ἀλλ᾿ οὐκ ἀνεκτόν, ἀλλὰ κατθανεῖν κρατεῖ· πεπαιτέρα γὰρ μοῖρα τῆς τυραννίδος. ἦ γὰρ τεκμηρίοισιν ἐξ οἰμωγμάτων μαντευσόμεσθα τἀνδρὸς ὡς ὀλωλότος; σάφ᾿ εἰδότας χρὴ τῶνδε μυθεῖσθαι πέρι· τὸ γὰρ τοπάζειν τοῦ σάφ᾿ εἰδέναι δίχα. ταύτην ἐπαινεῖν πάντοθεν πληθύνομαι, τρανῶς Ἀτρείδην εἰδέναι κυροῦνθ᾿ ὅπως.

To judge by the king’s cries, I think the deed has been done. Let us take common counsel (κοινωσώμεθα) and see if there might be any safe plan to follow. – I tell you what my proposal is: to proclaim an urgent call for the citizens to come here to the palace. – I think we should burst in straight away, and get proof of the crime when the blood is flowing freshly from the sword. – I share that opinion, and vote for doing something. This is a moment for not delaying. – You can see that. Their first actions show the behaviour of men giving the signal they mean to be tyrants of this city. – Yes, we’re wasting time, while they are trampling the much-touted virtues of delay into the ground and not letting their hands sleep. – I don’t know what plan to hit on and suggest. One who wants to act must first plan what action to take. – I agree, since I can see no way of bringing the man back from the dead just with words. – Are we really to prolong our lives like this by surrendering to the rulership of these defilers of the royal house? – No, it’s intolerable, and it’s better to die! That’s a less bitter fate than living under tyranny. – Are we to divine that the man is dead just from the evidence of some cries we hear? – We must talk about these things on the basis of firm knowledge. Guesswork is one thing, firm knowledge is another. – I have full support from all sides to approve this proposal, that we must have clear knowledge of how things are with the son of Atreus. The chorus cannot decide whether to take action or to lie low for the time being. There are two complementary ways of looking at this scene. One is to emphasise the disintegration of the chorus. Confronted with the death of their ruler and the murderous Clytemnestra, the chorus lose their ability to operate as a chorus and fragment into individual voices. A unit with a single mind becomes a group of people that deliberate by contributing their individual points of view. ‘The collective

198

f elix budel m ann

voice of the chorus, struggling to make unified sense of the experience of the group, is transformed into a play of separate and opposed voices engaged in a recognizably political argument’ (Gould 1996: 223). Alternatively, one maintains the idea of a collective mind and compares the scene to moments when individual characters articulate conflicting thoughts, such as, most famously, Euripides’ Medea in her great speech at Medea 1019–80. On such a reading, the split voices represent an externalised form of dithering or deliberation. Either way, the individuation of the voices, highly unusual in tragic choruses, marks a moment of intense crisis. Instances of a divided dēmos are more common than divided choruses, but they too are presented as deviations from the norm. Obvious examples could be drawn from Thucydides’ portrayal of post-Periclean Athens as a polis in which the concomitant of inferior leadership was a dēmos in upheaval.18 My example here is Xenophon’s account of the debate in the aftermath of the naval battle of Arginusae in 406 bce, which forms the climax of the first book of the Hellenica.19 The Athenian fleet had defeated the numerically superior Spartans, but a large number of the shipwrecked sailors drowned and a debate ensued over the culpability of the generals in charge. Xenophon has an extensive narrative, relaying a complex process involving three or four meetings of the assembly and creating the overwhelming sense of a situation out of control. It is too long to reproduce in full; a summary and a couple of extracts will be sufficient to convey the flavour. The first meeting, in which the generals are deposed, is treated only briefly and by way of background (1.7.1). The narrative proper begins with the second meeting in which Theramenes and others speak against the generals, who defend themselves by pointing to a storm which impeded the rescue operation (1.7.4–7). They were ‘about to persuade the dēmos’ (τοιαῦτα λέγοντες ἔπειθον τὸν δῆμον), but the meeting was adjourned at nightfall and the council tasked with drawing up a proposal about the manner of the trial. A festival intervened, which (according to Xenophon) was used to whip up fervour against the generals. Kallixeinos was persuaded to get the council to propose a summary vote on the guilt of all the generals, which he does successfully, and a third meeting ensues (1.7.9–34). In this meeting Kallixeinos is himself in turn accused by Euryptolemos and others of having made an unconstitutional proposal. The dēmos is divided, not a divided vote, but still a divided dēmos (1.7.12–14, trans. Brownson 1921, adapted): τοῦ δὲ δήμου ἔνιοι ταῦτα ἐπῄνουν, τὸ δὲ πλῆθος ἐβόα δεινὸν εἶναι, εἰ μή τις ἐάσει τὸν δῆμον πράττειν ὃ ἂν βούληται. καὶ ἐπὶ τούτοις εἰπόντος Λυκίσκου καὶ 18

See especially: 3.49.1 (a divided vote during the Mytilenean debate); 6.24.4 (the unconvinced minority still voting in favour in the Sicily debate); 8.66.5 (mutual suspicion among members of the dēmos under the oligarchic regime). When Thucydides sets out his vision of post-Periclean decline explicitly in chapter 2.65, he talks only vaguely of ‘stumbling over internal differences’ (2.65.12, κατὰ τὰς ἰδίας διαφορὰς περιπεσόντες), and characteristically emphasises above all the relationship between leaders and led, specifically Pericles’ control over the masses. 19 The passage has been much discussed. For bibliography, see Gish 2012 (whose reading, however, is difficult to accept) and Rood 2004: 374–80.



gr ou p min ds in class i cal at h ens ? 199 τούτους τῇ αὐτῇ ψήφῳ κρίνεσθαι ᾗπερ καὶ τοὺς στρατηγούς, ἐὰν μὴ ἀφῶσι τὴν κλῆσιν, ἐπεθορύβησε πάλιν ὁ ὄχλος, καὶ ἠναγκάσθησαν ἀφιέναι τὰς κλήσεις. Some of the people applauded this act, but the greater number cried out that it was monstrous if the dēmos were to be prevented from doing whatever it wished. Indeed, when Lykiskos thereupon moved that these men also should be judged by the very same vote as the generals, unless they withdrew the summons, the mob broke out again with shouts of approval, and they were compelled to withdraw the summonses.

Kallixeinos repeats his proposal, to more shouting. The prytaneis, except for Socrates, agree to put it to the vote. Euryptolemos speaks in defence of the generals (either on the same day or the next), the only speech Xenophon reproduces at length, and proposes to try each of them individually. The response is a fickle toing and froing between different views (1.7.34–5, trans. Brownson 1921, adapted): τούτων δὲ διαχειροτονουμένων τὸ μὲν πρῶτον ἔκριναν τὴν Εὐρυπτολέμου· ὑπομοσαμένου δὲ Μενεκλέους καὶ πάλιν διαχειροτονίας γενομένης ἔκριναν τὴν τῆς βουλῆς. καὶ μετὰ ταῦτα κατεψηφίσαντο τῶν ναυμαχησάντων στρατηγῶν ὀκτὼ ὄντων. ἀπέθανον δὲ οἱ παρόντες ἕξ. καὶ οὐ πολλῷ χρόνῳ ὕστερον μετέμελε τοῖς Ἀθηναίοις, καὶ ἐψηφίσαντο, οἵτινες τὸν δῆμον ἐξηπάτησαν, προβολὰς αὐτῶν εἶναι, καὶ ἐγγυητὰς καταστῆσαι, ἕως ἂν κριθῶσιν, εἶναι δὲ καὶ Καλλίξεινον τούτων. The vote being now taken as between these two proposals, they decided at first in favour of the resolution of Euryptolemos; but when Menekles interposed an objection under oath and a second vote was taken, they decided in favour of that of the council. After this they condemned the generals who took part in the battle, eight in all; and the six who were in Athens were put to death. And not long afterwards the Athenians repented, and they voted that complaints be brought against any who had deceived the dēmos, that they furnish bondsmen until such time as they should be brought to trial, and that Kallixeinos be included among them. Not for nothing is this chapter a favourite of modern scholars who take a critical view of Athenian democracy. Xenophon does his best throughout to present the whole affair as ugly business. He creates the impression of unconstitutional procedure (5, 12), behind-the-scenes lobbying (8) and intimidation (13). His teacher and idol Socrates finds himself standing up alone while the other prytaneis are too frightened (15). With this ugly business comes a dēmos that not only changes its mind but is also divided. The criticism of Kallixeinos’ proposal splits the dēmos and prompts heckling (1.7.12),20 and the counter-proposal by Euryptolemos is voted on twice, 20

For other cases of such split audience response, see Hansen 1987: 160 n. 453.

200

f elix budel m ann

with different results, because the first vote is challenged, possibly an indication of a tight balance (1.7.34).21 Notably, the rhetoric Xenophon reports from the assembly still treats the dēmos as a unit: in the passages cited above, there are the cries ‘that it was monstrous if the dēmos were to be prevented from doing whatever it wished’, and after the event, in a spirit of ‘repenting’, the motion is passed that ‘complaints be brought against any who had deceived the dēmos’. But in (Xenophon’s) reality, the dēmos in this situation is a far cry from the unanimity that constitutes the norm. For all the obvious differences, the chorus of Agamemnon and the dēmos of Xenophon have one essential thing in common: they are groups that break up when things go wrong. An extreme moment of crisis renders a unified experience and unified vote an impossibility.

Combining Individual- and Group-Level View: A Rare Occurrence Much of the current philosophical debate over group minds is concerned with relating the group mind to the individual minds. In a frequently cited book-length case study of distributed cognition, for example, Edwin Hutchins analysed joint navigation work on the bridge of a navy vessel, in which each team member performed their own specific task. Philip Pettit, to give another example, looked at the way a panel of judges can come to different decisions depending on whether they individually consider the relevant criteria and then vote once, or whether they vote on each factor and then pool those decisions.22 Fifth- and fourth-century Athenians, despite their evident willingness to conceive of groups such as the chorus and the dēmos as a single mental unit, did not share these interests. On the whole, our texts focus either on individual minds or on groups, but they do not attempt to analyse a system of individual minds that operate as a group. This is an important point, the implications of which I will address in the final section. For now, however, I wish to look at the very rare exceptions. The single most interesting exception in this regard is a passage from Aristotle’s Politics 3.11. In the context of discussing different constitutional regimes, Aristotle asks how one may justify democracy (1281a40–b10, trans. Rackham 1991, adapted): ὅτι δὲ δεῖ κύριον εἶναι μᾶλλον τὸ πλῆθος ἢ τοὺς ἀρίστους μὲν ὀλίγους δέ, δόξειεν ἂν λύεσθαι καί τιν᾽ ἔχειν ἀπορίαν, τάχα δὲ κἂν ἀλήθειαν. τοὺς γὰρ πολλούς, ὧν ἕκαστός ἐστιν οὐ σπουδαῖος ἀνήρ, ὅμως ἐνδέχεται συνελθόντας εἶναι βελτίους ἐκείνων οὐχ ὡς ἕκαστον ἀλλ᾽ ὡς σύμπαντας, οἷον τὰ συμφορητὰ δεῖπνα τῶν ἐκ μιᾶς δαπάνης χορηγηθέντων· πολλῶν γὰρ ὄντων ἕκαστον μόριον ἔχειν ἀρετῆς καὶ φρονήσεως, καὶ γίνεσθαι συνελθόντας ὥσπερ ἕνα ἄνθρωπον τὸ 21

The grounds on which the challenge is made are disputed: see Hansen 1987: 44; Bleckmann 1998: 567–8; Harris 2013: 342–3. 22 Hutchins 1995; Pettit 2003. For a recent attempt to develop an analysis of collective cognition as a system, see Huebner 2014.



gr ou p min ds in class i cal at h ens ? 201 πλῆθος πολύποδα καὶ πολύχειρα καὶ πολλὰς ἔχοντ᾽ αἰσθήσεις, οὕτω καὶ περὶ τὰ ἤθη καὶ τὴν διάνοιαν. διὸ καὶ κρίνουσιν ἄμεινον οἱ πολλοὶ καὶ τὰ τῆς μουσικῆς ἔργα καὶ τὰ τῶν ποιητῶν· ἄλλοι γὰρ ἄλλο τι μόριον, πάντα δὲ πάντες. But the view that it is more proper for the multitude to be sovereign than the few of greatest virtue might be thought to be explicable, and to raise some difficulty but probably to be true. For it is possible that the many, though not individually good men, yet when they come together may be better, not individually but collectively, than those who are so, just as public dinners to which many contribute are better than those supplied at one man’s cost; for where there are many, each individual, it may be argued, has some portion of virtue and wisdom, and when they have come together, just as the multitude becomes a single man with many feet and many hands and many senses, so also it becomes one as regards character and thinking. This is why the general public is a better judge of the works of music and those of the poets, because different men can judge a different part of the performance, and all of them all of it.

The question of how the dēmos can arrive at good decisions is a difficult one for Aristotle, since for him right and wrong are absolutes rather than mutually agreed in a social contract.23 His solution turns on the pooling of different resources. He uses the comparison of a public meal drawn from many individual contributors and gives the example of publicly judged performances, where each judge contributes different expertise. The passage has been discussed variously for its political as well as philosophical dimensions.24 Here I am concerned only with the central image: ‘Just as the multitude becomes a single man with many feet and many hands and many senses, so also it becomes one as regards character and thinking.’ This is a strange image, so strange in fact that elsewhere Aristotle would classify such a creature as a monster. In the Generation of Animals, he notes that one way in which creatures qualify as monsters is ‘by having a shape of many parts and being born with many feet or many heads’ (4.3, 769b; τὰ δὲ τῷ πολυμερῆ τὴν μορφὴν ἔχειν, πολύποδα καὶ πολυκέφαλα γιγνόμενα). The cause of the strangeness is the unmitigated clash of two levels. On the one hand Aristotle maintains the default view of the dēmos as one being (ἕνα ἄνθρωπον); on the other he tries to account for it in terms of its many members. The result is a monstrosity. Whatever one thinks of Aristotle’s argument about the pooling of varied expertise, his memorable central image is testimony to the difficulty of maintaining two perspectives at the same time, that of the group as one and that of the group as many. So far as I am aware, no other classical text goes quite as far in thinking about the way individual and collective cognition interlock (though Aristotle himself 23

For a discussion of the passage in the context of Aristotle’s ethical thinking, see Gottlieb 2009: 203–7. 24 Philosophy: Waldron 1995, and previous note; politics: several publications by Ober, esp. 1999: 319–24. One might argue that Thuc. 6.38–9 is a rudimentary early precursor of Aristotle’s model.

202

f elix budel m ann

treats related themes elsewhere).25 Unsurprisingly, the chorus offers no proper parallel. There is nevertheless one passage that deserves brief consideration. The author of the Pseudo-Euripidean Rhesus experiments with various forms of innovation in the treatment of the chorus (see Jackson 2014: ch. 2). In the third stasimon, he alternates the kind of individual voices we saw in the Agamemnon with the standard ensemble singing. The chorus of Trojan soldiers on night-watch duty is awaiting the return of Rhesus, who is spying on the Greek camp (Rhes. 527–64, text and trans. Kovacs 2003, adapted): τίνος ἁ φυλακά; τίς ἀμείβει τὰν ἐμάν; πρῶτα δύεται σημεῖα καὶ ἑπτάποροι Πλειάδες αἰθέριαι· μέσα δ᾿ αἰετὸς οὐρανοῦ ποτᾶται. ἔγρεσθε, τί μέλλετε; κοιτᾶν ἔξιτε πρὸς φυλακάν. οὐ λεύσσετε μηνάδος αἴγλαν; ἀὼς δὴ πέλας ἀὼς γίγνεται καί τις προδρόμων ὅδε γ᾿ ἐστὶν ἀστήρ. – – – –

τίς ἐκηρύχθη πρώτην φυλακήν; Μυγδόνος υἱόν φασι Κόροιβον. τίς γὰρ ἐπ᾿ αὐτῷ; – Κίλικας Παίων στρατὸς ἤγειρεν, Μυσοὶ δ᾿ ἡμᾶς. οὔκουν Λυκίους πέμπτην φυλακὴν βάντας ἐγείρειν καιρὸς κλήρου κατὰ μοῖραν;

καὶ μὰν ἀίω· Σιμόεντος ἡμένα κοίτας φοινίας ὑμνεῖ πολυχορδοτάτᾳ γήρυϊ παιδολέτωρ μελοποιὸν ἀηδονὶς μέριμναν. ἤδη δὲ νέμουσι κατ᾿ Ἴδαν ποίμνια· νυκτιβρόμου σύριγγος ἰὰν κατακούω. θέλγει δ᾿ ὄμματος ἕδραν ὕπνος· ἅδιστος γὰρ ἔβα βλεφάροις πρὸς ἀῶ. – – – 25

τί ποτ᾿ οὐ πελάθει σκοπός, ὃν ναῶν Ἕκτωρ ὤτρυνε κατόπτην; ταρβῶ· χρόνιος γὰρ ἄπεστιν. ἀλλ᾿ ἦ κρυπτὸν λόχον ἐσπαίσας διόλωλε; τάχ᾿ ἂν εἴη φοβερόν μοι. αὐδῶ Λυκίους πέμπτην φυλακὴν βάντας ἐγείρειν ἡμᾶς κλήρου κατὰ μοῖραν.

Sung [whole chorus]: Who’s on guard duty? Who will relieve me? The early constellations are setting and the seven-starred Pleiades are aloft. The Eagle flies in mid-heaven. Look lively there! What’s the delay? Out of your beds and to the watch! Don’t you see how the moon shines? Dawn, I tell you, dawn is near, and this star is her harbinger. Anapaests [split voices]: – Who was announced for the first watch? – Mygdon’s son, they tell me, Koroibos. – And who followed him? – The Paeonian contingent woke the Cilicians, and the Mysians woke us. – Should we not then go and wake the Lycians, the fifth watch in the lot’s apportionment? Sung [whole chorus]: Listen. I hear the nightingale. She sits on her bloodstained nest by the Simois, child-slayer she, and in melodious strain sings her musical woe. Already on Ida they are tending the flocks: I hear the buzz of the night-murmuring shepherd’s pipe. Sleep puts its spell on my eyes: most sweetly does it come upon the lids towards dawn. Anapaests [split voices]: – Why is he not coming, the man, Hector sent to spy on the ships? – I’m worried: he has been gone a long time. – Has he blundered into a hidden ambush and been killed? Perhaps this will become I was afraid of. – I say: go and wake the Lycians, the fifth watch in the lot’s appointment. [Exit chorus] As in Agamemnon, the chorus splits into individual voices – in this case couched in anapaests – as the collective tries to decide what to do.26 Again as in Agamemnon, their decision-making is flawed: they decide to seek out the Lycian contingent which is due to take over the watch and as soon as they have left their post the Greek assassins strike. But there is nevertheless a sense in which this is genuine collaborative deliberation. When the group as a whole is uncertain (Who is on duty? Who will relieve me?), one of its members suggests they think systematically (Who had the first watch?). The second and, prompted by the third, the fourth speaker 26

The precise distribution of the lines between speakers is uncertain: see Liapis 2012 on 534–5, 557–64. The chorus split into individual voices again during the fourth stasimon, 692–727.

204

f elix budel m ann

pool their knowledge, permitting the fifth to conclude that it is the Lycians who are up next. In the second part, too, the men bounce thoughts off one another until they arrive at their decision.27 Interpreted generously, the chorus are what those working on distributed cognition call an intermental unit: a group of people, such as the navigation team on the warship or a married couple collectively remembering the past, who (to return to Aristotle) pool knowledge, expertise and, more generally, cognitive resources. This decision-making through cooperation among a set of individuals alternates with self-expression by the chorus in unison. The group as a whole is no less engaged than its individual members, exhorting itself (in the second person plural) to keep watch. But not all is the same. Not only does the chorus-as-a-group sing and is lyrical in its language, but unlike the individual choreuts it does not engage in decision-making. The author of the Rhesus juxtaposes two modes, one that shows a set of individuals in cognitive cooperation, and another in which the unit as a whole expresses its anxieties, sensibilities, thoughts. This is not the systematic exploration familiar from current work on distributed cognition, nor the clashing juxtaposition we saw in Aristotle, but it goes some way in that direction. Both the Aristotle and the Rhesus stasimon are among the very rare passages that look at group cognition both in terms of a set of individuals and in terms of a single unit.

The Extent of Collective Agency While such focus on the how of collective cognition is unusual in our texts, there clearly is an interest in the what. The mechanics of group minds are notable by virtue of their almost complete absence, but there is much evidence of thinking about the kinds of things collective cognition can and cannot achieve, which creates interesting points of contact between the ancient material under discussion in this chapter and contemporary philosophy. In particular, philosophers are debating whether groups as a whole have agency, and if they do what kind of agency this is. The following table sets out in highly simplified form the views put forward in two influential contributions which both argue for the existence of group agency but differ in the way they conceive it.28 French 1995 (on the ethics of corporations)

List and Pettit 2011

1. Intentionality 2. Rationality 3. Responsiveness to criticism Not: beliefs, desires (intentions are laid down in charters, documents, etc.)

1. Beliefs 2. Desires 3. Capacity to act on beliefs 4. Some minimal rationality

27

In the economy of the play as a whole, the chorus’s decision-making is part of a pervasive interest in euboulia: see Hesk 2011: 136–43. 28 I am drawing on the overview in Tollefsen 2015: 53–64.



gr ou p min ds in class i cal at h ens ? 205

Arguably, representations of the chorus and dēmos of classical Athens are closer to List and Pettit than to French, but they are by no means identical. Choruses express their beliefs and desires with such frequency that it is unnecessary to provide examples. Almost any ode could be cited as a case in point. Thought and emotion are the bread and butter of choral utterance. The capacity of choruses to affect the outcome of the play is more circumscribed than that of the solo characters, but they can certainly act on their beliefs, as they exhort and advise characters and indeed act themselves. The chorus of Choephori intercepts of its own accord Clytemnestra’s message to Aegisthus, for example, in furtherance of the revenge plot (766–73), and the chorus of Rhesus, as we saw, leave the stage to seek out the next shift. The dēmos is more difficult since it does not have a literal voice like a chorus. What is very frequent, nevertheless, is the attribution of beliefs and desires to it. In the Xenophon passage above the Athenians ‘repented’ and as a result passed a new motion. Aeschines, to give an example from an orator, is in no way untypical when he says: ‘if the dēmos, gone mad, or forgetful of the existing situation, had actually wished to crown [Demosthenes] at a time so unfitting . . .’ (εἰ καὶ μανεὶς ὁ δῆμος ἢ τῶν καθεστηκότων ἐπιλελησμένος, ἐπὶ τοιαύτης ἀκαιρίας ἐβούλετο στεφανοῦν αὐτόν . . ., 3.211, trans. Adams 1919). Or here is a sentence from Thucydides, again representative: ‘The dēmos was at first displeased when it heard the proposal concerning an oligarchy; but when it had been plainly shown by Peisander that there was no other salvation, through fear and at the same time because it expected to make a change later, it yielded’ (ὁ δὲ δῆμος τὸ μὲν πρῶτον ἀκούων χαλεπῶς ἔφερε τὸ περὶ τῆς ὀλιγαρχίας· σαφῶς δὲ διδασκόμενος ὑπὸ τοῦ Πεισάνδρου μὴ εἶναι ἄλλην σωτηρίαν, δείσας καὶ ἅμα ἐπελπίζων ὡς καὶ μεταβαλεῖται, ἐνέδωκεν, 8.54.1, trans. Smith 1923, adapted). Of course, one can reasonably maintain that this is just figurative speech: I will come back to that shortly. For now what matters is that it is a ubiquitous and intricately developed form of figurative speech. Quantitatively and qualitatively, Athenian texts go very far in using language and dramatic forms that attribute mind states to groups. What about rationality and responsiveness to criticism, which is a major feature of group agency for French but not for List and Pettit? For the dēmos one enters the fraught territory of political partisanship. Our often anti-democratic sources go out of their way to portray the dēmos as irrational. Again and again the dēmos is volatile and in the throes of its emotions. As a body that votes on proposals made by others rather than speaks itself, it makes its views known through shouts rather than reasoned argument; the example in the first Xenophon citation above is not an isolated instance.29 Neither is the topos, in the second quotation, of the dēmos abdicating responsibility by blaming its leaders. According to the Old Oligarch, ‘it is possible for the dēmos to disown whatever agreements it makes, by referring the responsibility to the one man who proposed the motion and the one man who 29

On shouting and other forms of audience participation in assemblies, see Bers 1985, Wallace 2004 and Thomas 2011. Thomas’s article, on the dynamic of mass audiences in democratic Athens, is also more generally relevant to this chapter.

206

f elix budel m ann

put it to the vote’ (ἅσσα δ’ ἂν ὁ δῆμος συνθῆται, ἔξεστιν αὐτῷ, ἑνὶ ἀνατιθέντι τὴν αἰτίαν τῷ λέγοντι καὶ τῷ ἐπιψηφίσαντι ἀρνεῖσθαι, 2.17, trans. Marr and Rhodes 2008).30 The dēmos of our hostile sources certainly does not display French’s notion of responsiveness to criticism. It is a group that is joint in strong views and strong emotion rather than in reason. The reality behind this ideological caricature is difficult to recover and this is certainly not the place to try.31 An interesting comparandum, correcting and corroborating in equal measure, is provided again by the chorus. It would be absurd to deny choruses rationality: more often than not, they are articulate, measured and sensible. Nevertheless, this is an area in which the chorus’s mind is not simply an individual mind writ large. The train of thought of dramatic choruses, as of their extra-dramatic counterparts, is different from that of individuals in ways that are as easy to recognise as they are hard to describe. Choral thought, certainly in the odes, tends to be more allusive and less linear. Choral accountability, too, differs from that of individuals. The limelight, more often than not, is on individual characters: it is individual characters who come to suffer in tragedy and it usually is the individual characters who take the lead.32 In this respect, the anti-democratic caricature of the dēmos finds a partial parallel in the chorus.

Athenian Group Minds? Even in this cursory discussion, the treatment of the chorus and the dēmos reveals a deep interest in collective cognition among Greek authors of the classical period. Athenians readily credited groups with decision-making, with emotion, with thought; they also took an interest in what makes group cognition different. Yet none of this justifies in itself the term ‘group mind’. I noted at the outset that those who balk at the notion of the group mind object that to say a group thinks, feels and so on is only a form of speech, and that there is nothing in collective cognition that is not sufficiently explained at the level of the individual minds. Such a challenge is mounted, for example, in a recent paper by Robert Rupert, who does not think that group minds have properly causal explanatory power. According to Rupert (2014), individual-level explanations are sufficient and more parsimonious. I will end by asking what, if anything, the dēmos and the chorus have to contribute to that problem. First of all, there is much that it does not contribute. Above all, nothing in what the Athenians had to say about the chorus or the dēmos offers material for an argument that counters the challenge that the notion of the group mind is merely figurative language. As we saw, unlike contemporary philosophers, they were on the whole not interested in explaining how the collective of individuals related to 30

For further instances, incl. Thuc. 8.1.1, see Marr and Rhodes 2008: 130. A more nuanced pictured emerges if one considers Demosthenes’ assembly speeches: he regularly styles himself as an advisor who is not afraid of criticising the dēmos; see e.g. 3.30ff., 4.10–12, 9.3–5. 32 Choral passivity and followership gives rise to its own caricature: see Pl. Prt. 315b, Euthyd. 276b. 31



gr ou p min ds in class i cal at h ens ? 207

the group as a unit, and in the rare instances where they did broach that issue, the difficulty of combining the two levels emerges clearly. (Though it is hardly a coincidence that the writer who went furthest in this direction is a philosopher, Aristotle.) Nevertheless the dēmos and chorus have something important to offer to those interested in group minds. They provide an example of a culture that, in certain specific instances, appears to be more extensively committed than we are today to treating the group rather than the individual as the default level at which mental activity is situated. This commitment to the chorus and the dēmos as the base unit has been manifest throughout, but may usefully be illustrated one more time by thinking about the symmetry of chorus and dēmos. As a matter of fact, chorus and dēmos were very different kinds of groups. Dramatic choruses are much smaller, consisting of twelve or fifteen performers for tragedy and twenty-four for comedy, and therefore are considerably closer to the kinds of groups that have been studied as models of intermental units by philosophers. It is thoroughly credible that these small-scale groups, which rehearse together for a long time, do indeed begin to operate in very close coordination, with a part-intuitive understanding of one another’s movements and articulation. The same is not the case for the hundreds or thousands of citizens who come together in any one assembly, but arguably the difference is not absolute. In an important article, Mirko Canevaro has challenged the view that consensus in the assembly is merely an ideological construct. Pointing to the unanimity or minimal disagreement in the records of the forty or so secret ballots in deliberative settings that survive from antiquity, to the role of the proedroi in deciding what proposal to put to the vote and when, and to the tendency of open ballots to produce more coherent results, he argues that processes in the assembly were designed to create consensus: ‘We should entertain . . . the possibility that strong disagreements existed and were expressed during debate and deliberation, and that debate and deliberation, institutionally directed and facilitated, could ultimately produce . . . consensus’ (Canevaro 2018: 144). If Canevaro’s argument points in the right direction, as seems likely, it strengthens the parallel between dēmos in the assembly and chorus. Both are groups capable of genuine cohesion. However, the differences remain fundamental. Even if the final assembly vote is consensual, consensus-building will often be a complex process and some dissent will often remain. The sheer size of the assembly must have strictly limited the degree to which individuals within it operated as a coordinated group, even in what was more of a face-to-face society than most modern cities.33 Yet, and this is the striking point, these differences in internal dynamics are of little interest to our sources. Much more often than not, chorus and dēmos alike

33

It is notable that Ober 2008, which argues that the institutions of Athenian democracy were such that they used citizens’ knowledge to best effect, puts considerable emphasis on smaller units (boulē, prytaneis, etc.).

208

f elix budel m ann

are groups operating with a single mind.34 These single minds are essentially black boxes, viewed from outside, their different make-up and their internal cognitive processes invisible. Even though there is as far as I can see no cogent riposte to the charge that ‘group minds’ are a form of figurative language rather than a reality, most Athenians would probably not want to put it that way where chorus and dēmos are concerned. In ancient Greece in general and democratic Athens in particular, groups were prominent in all aspects of life. Not only will many Athenian males have had first-hand experience of both dancing in choruses and voting in the assembly or the law courts,35 but group experiences are not confined to the two case studies chosen here. Many citizens served on triremes, where they sat together on crammed benches, pulling on the same oar, to a beat given by a piper. Festival processions, too, were regular occurrences, and of course the musical and poetic competitions, conducted before large audiences and stretching over several days, many hours per day, were themselves communal events. It is to be expected that a culture in which group experiences were a more integral part of life than they are in many cultures today would more naturally and more extensively think not just of individuals but also of certain groups as cognitive units.36 As befits a chapter in a History series, the discussion, which started from a body of literature in philosophy, ends with a statement about the importance of cultural context, but perhaps it is nevertheless a statement with at least some relevance to the current debate. If we took the Athenian chorus and dēmos as our basis for thinking about group minds, we might arrive at the following position. Whether one accepts that there is such a thing as group minds depends on one’s cultural starting point. To deduce the existence of group minds by logical argument while operating according to an ontology that is based on individuals and in which groups are a secondary phenomenon is bound to be a major challenge (and as such the term ‘group mind’ may itself be a hindrance, as it is so clearly rooted in the notion of the – individual – mind). Might it be the case that what makes the concept flourish is a context in which a particular group is considered the intuitively appropriate starting point in the first place? Might it be the case that if there is a natural home for group minds it is not in thinking about ‘distributed cognition’, an approach that builds on the base model of the individual mind even though it then seeks to look beyond it, but rather in settings in which certain groups come first and their members second?37

34

And see, again, Visvardi 2015 for other aspects of this parallel treatment. On the more specific, related issue of the representation of the dēmos in drama, embodied in the chorus and by other means, see Carter 2010. 35 For the large-scale participation in dramatic choruses, see Revermann 2006b. 36 Cultural neuroscience suggests that cultural differences can even influence, and/or be influenced by, social cognitive and social affective brain processes; see e.g. the meta-analysis in Han and Ma 2014. But the field is still in its infancy, as well as controversial. 37 I am grateful to Tim Rood, Mirko Canevaro, Guy Westwood, the editors and anonymous ­referees, and the members of the HDC workshop for helpful advice.



one sou l in tw o b o di es 209

12 O n e S o ul i n T w o Bod ies : D i s t ri but e d C o gnition a nd An ci e n t Gre e k Friend s hip David Konstan

It is said that when Aristotle was asked, ‘What is a friend?’ (τί ἐστι φίλος;), he replied: ‘One soul dwelling in two bodies’ (μία ψυχὴ δύο σώμασιν ἐνοικοῦσα). So wrote Diogenes Laertius (5.1.20), who composed biographies of the ancient Greek philosophers in the third century ce. A similar definition is found in Cicero’s essay On Friendship (81), when he says that a human being naturally ‘both loves himself and seeks out another whose mind [animus] he may so mingle with his own as almost to make one mind out of two’ (‘et se ipse diligit et alterum anquirit, cuius animum ita cum suo misceat ut efficiat paene unum ex duobus’).1 Cicero, with Roman practicality, inserted the qualifier ‘almost’ (paene), but even so the idea is startling if taken literally, as I believe that Aristotle, at least, intended. Similarly, the formula ‘another self’, which Aristotle applies to a friend at least half a dozen times in his treatises and is also echoed by Cicero (again modified by the unreal conjunction: ‘as though the friend is both another and the same’, ‘est enim is qui est tamquam alter idem’, 80), suggests that friends, although they are clearly distinct entities (hence ‘another’), are in some sense nevertheless one and the same (the meaning of idem). In this chapter, I argue that Aristotle’s account of friendship in fact anticipates, and in certain ways even goes further than, aspects of the theory of distributed cognition. As Yvonne Rogers, in ‘A brief introduction to distributed cognition’ (1997), explains: ‘A general assumption of the distributed cognition approach is that cognitive systems consisting of more than one individual have cognitive properties that differ from those individuals that participate in those systems.’ The notion of distributed cognition has been applied to tasks that are undertaken jointly by a group, for example piloting a ship, where no one person may have access to all the data needed to determine the proper course, and to the use of external props in making calculations, as when we do arithmetic sums or division on paper or on a blackboard that we could not do just in our heads (Rogers provides a crisp summary of such cases). The cognitive process at work in these cases can be thought of as ranging over the collective mental activities of the group or, in the latter  1

Cf. Horace’s description of Vergil as animae dimidium meae or ‘half of my soul’ (Carm. 1.3.8).

210

dav id k o ns t an

instance, as including the passive instrumentality of the chalkboard, which allows us to externalise our thought processes and without which the relevant conclusions could not have been reached. Friendship is yet another type, one in which the joint processes of two or more minds are constitutive of the relationship as such. But I should like first to illustrate a curious case in which intellectual props of a very particular sort are explicitly treated as extensions of a subject’s cognitive capacities, although the claim is also subjected to an ironic critique which nevertheless falls short of dislodging it. In the 27th epistle in Seneca’s Letters to Lucilius (sometimes entitled Epistulae morales), Seneca encourages his young addressee to turn away from guilty pleasures and pursue a life devoted to philosophy. As he puts it: ‘Virtue alone affords everlasting and peace-giving joy; even if some obstacle arise, it is but like an intervening cloud, which floats beneath the sun but never prevails against it’ (‘sola virtus praestat gaudium perpetuum, securum; etiamsi quid obstat, nubium modo intervenit, quae infra feruntur nec umquam diem vincunt’, 27.3, trans. Gummere 1917). This kind of effort, moreover, ‘cannot be delegated to someone else’ (‘delegationem res ista non recipit’, 27.4). In this respect, it differs, according to Seneca, from other kinds of literary activity, which do allow for external aids. By way of illustrating this claim, Seneca cites the example of a contemporary of his, a wealthy man by the name of Calvisius Sabinus,2 who had, Seneca writes with haughty disdain, ‘the bank-account and the brains of a freedman’ (‘et patrimonium habebat libertini et ingenium’, 27.5). This Sabinus was plagued by so poor a memory that he might even forget such household names as Ulysses, Achilles and Priam, names which, Seneca avers, ‘we know as well as we know those of our own teachers’ (‘quos tam bene quam paedagogos nostros novimus’, ibid.). Indeed, a command of the Iliad and Odyssey, along with other classics of Greek literature, was a crucial mark of education and a sign of elite status in Rome’s hierarchical culture; not to be able to recall them was taken as the sign of a seriously deficient intellect. Seneca insists that even a decrepit nomenclator, that is, a slave assigned to remind his master of the names of his acquaintances and clients (who might number in the thousands), would not suffer lapses like those that afflicted Sabinus in regard to the great heroes of mythology. Sabinus availed himself of a remedy, however, to compensate for his shortcoming, modelled on the common figure of the nomenclator or ‘namer’: he purchased at inordinate prices slaves who had mastered the entire texts of Homer and Hesiod, as well as the other major poets, and if he could not find appropriately trained individuals he had them educated to the task himself. In itself, this is no cause for surprise; after all, highly educated slaves often served as teachers for the children of the aristocracy and memorisation was among the fundamental talents they fostered. For his part, Sabinus kept his retinue of slaves near his couch at dinner parties, and whenever he faltered in citing a verse of poetry, it was their task to prompt him so that he might complete the citation – a routine that Seneca says made life intolerable for his guests (27.6).  2

Cf. Short, this volume, pp. 97–9.



one sou l in tw o b o di es 211

Up to this point, we may take it that Sabinus’ slaves functioned as little more than note cards and formed no part of his own mental apparatus. But Sabinus himself regarded the matter differently. As it happened, a certain Satellius Quadratus, who conformed to the type of the ‘parasite’, that is, a clever and witty but impoverished fellow who fawned on the rich in order to cadge a meal but also, in the way of such chaps, might secretly mock them for their ignorance and boorishness, remarked that Sabinus could have bought himself a library full of books for far less than the cost of all those slaves. ‘But’, writes Seneca, ‘Sabinus held to the opinion that what any member of his household knew, he himself knew also’ (‘ille tamen in ea opinione erat, ut putaret se scire, quod quisquam in domo sua sciret’, 27.7). Clearly, Sabinus could not open a papyrus roll at the right place in order to find the appropriate passage, nor could he memorise the contents of his bookshelf: it is for this very reason that he requires the assistance of his slaves, who are continuously available and provide him with precisely the citations he needs. The slaves function as living props, in constant interaction with the master. They are his external hard drive, as it were, storing information that he can tap into at any moment. Their role is very much that of extensions of Sabinus’ cognition. Seneca does not let the matter rest here. Satellius goes on to suggest that Sabinus take lessons in wrestling. When Sabinus demurs on the grounds that he is sickly and unfit, Satellius retorts: ‘Don’t say that, I implore you; consider how many perfectly healthy slaves you have!’ (‘ “Noli, obsecro te”, inquit, “istuc dicere; non vides, quam multos servos valentissimos habeas?”’ 27.8). The point is witty, but it misses the mark: there is a sense in which Sabinus would be physically better off, to the extent that his slaves perform labour for him, in much the same way that a person with a prosthetic limb or denture is. But whatever the force of the analogy to distributed cognition, Satellius’ jibe does not reflect Seneca’s own position. Seneca adduced the example of Sabinus to show that one can avail oneself of another’s learning in literary matters; it is only in philosophy that one is required to do all the work oneself. As Seneca concludes: ‘No man is able to borrow or buy a sound mind; in fact, as it seems to me, even if sound minds were for sale, they would not find buyers. Depraved minds, however, are bought and sold every day’ (‘bona mens nec commodatur nec emitur. Et puto, si venalis esset, non haberet emptorem. At mala cotidie emitur’).3 William Fitzgerald (2000: 13) speaks of ‘the symbiosis of master and slave’ in ancient Rome, ‘a paradoxical symbiosis between the master and his “separate part” that expresses itself in complementarities, reversals and appropriations’. Fitzgerald illustrates this interdependency by way of two poems. The first is an epitaph for a certain Xanthias, a descendant of Cicero’s famous scribe Tiro, who invented the technique of stenography in order to record the great orator’s words. Xanthias died prematurely and the commemorative inscription (CIL 13.8355) notes that he had already mastered shorthand and could read expertly; it concludes:

 3

For further discussion, see Seal 2009: 74–87.

212

dav id k o ns t an

iam voce erili coeperat ad omne dictatum volans aurem vocari at proximam. heu morte propera concidit arcana qui solus sui sciturus domini fuit. Already he had begun to be summoned to be his master’s closest ear, flying to every dictate of his master’s voice. Alas, he succumbed to hasty death, he who alone would have known his master’s intimate thoughts.4 The second is a brief lyric by the fourth-century ce poet Ausonius, addressed to his secretary (notarius). Ausonius affirms that the boy (or slave: puer) knows his thoughts and transcribes them on to the wax tablet even before they have been uttered (Ephemeris 7.16–17); god and nature, Ausonius concludes, must have granted him the gift to know in advance what he was going to say and to wish just what he wished (‘idemque velles quod volo’, 33–6; cf. Fitzgerald 2000: 16–17). Sabinus’ conviction that what his slaves know counts as his own knowledge reflects the tendency on the part of Roman masters to treat as instrumental possessions not just the bodies of their slaves but their minds as well. However interdependent the master and slave might be, the power relation was clearly unequal and it was the master who absorbed the intellect of his amanuensis, not vice versa (though Ausonius’ poem suggests a certain anxiety about the slave pilfering his master’s thoughts: cf. furta, 25). Friends, on the contrary, were equals and shared cognition could not take the one-sided, incorporative form of the master-slave relation. Friendship would thus have resembled rather the kind of bond that Thomas Fuchs and Hanne De Jaegher call ‘enactive intersubjectivity’, which in one respect may be understood as the ‘interaction and coordination of two embodied agents’, while on the ‘phenomenological’ level it takes the form of ‘a mutual incorporation, i.e. a process in which the lived bodies of both participants extend and form a common intercorporality’ (Fuchs and De Jaegher 2009: 465). This is a radical formula, which calls into question the idea of an entirely autonomous self and the concomitant assumption that it is impossible to bridge the gap between one’s own consciousness and that of another – premises that have been central to philosophical and social scientific analysis ever since Descartes (the view is sometimes identified as ‘Cartesian’) and which have only recently come to be challenged by various currents of thought loosely grouped under the label ‘postmodern’ (cf. Hermans and Hermans-Konopka 2010; Rudder Baker 2011). But Aristotle and his contemporaries were in an important respect ‘pre-modern’ and  4

Trans. Courtney 1995; cited in Fitzgerald 2000: 14.



one sou l in tw o b o di es 213

they did not necessarily subscribe to this kind of dualism. Given Aristotle’s uncanny rigour and insight and the vibrant philosophical context in which he wrote and on which he in turn exercised a massive influence, it is reasonable to expect an approach to self and consciousness that provides a usable philosophical framework for investigations of intersubjective cognition – all the more so, in that the thoughtworld of classical Greece promoted what Christopher Gill has described as an intersubjective or collectivist approach to mind and understanding (Gill 1996). The idea that friendship in some sense constitutes a union of selves finds expression in a variety of rituals that are intended to confirm and at the same time symbolise the relationship. One such practice was described by the Greek historian Herodotus (3.8.1–2, my translation): σέβονται δὲ Ἀράβιοι πίστις ἀνθρώπων ὅμοια τοῖσι μάλιστα, ποιεῦνται δὲ αὐτὰς τρόπῳ τοιῷδε: τῶν βουλομένων τὰ πιστὰ ποιέεσθαι ἄλλος ἀνήρ, ἀμφοτέρων αὐτῶν ἐν μέσῳ ἑστεώς, λίθῳ ὀξέι τὸ ἔσω τῶν χειρῶν παρὰ τοὺς δακτύλους τοὺς μεγάλους ἐπιτάμνει τῶν ποιευμένων τὰς πίστις, καὶ ἔπειτα λαβὼν ἐκ τοῦ ἱματίου ἑκατέρου κροκύδα ἀλείφει τῷ αἵματι ἐν μέσῳ κειμένους λίθους ἑπτά: τοῦτο δὲ ποιέων ἐπικαλέει τε τὸν Διόνυσον καὶ τὴν Οὐρανίην. ἐπιτελέσαντος δὲ τούτου ταῦτα, ὁ τὰς πίστις ποιησάμενος τοῖσι φίλοισι παρεγγυᾷ τὸν ξεῖνον ἢ καὶ τὸν ἀστόν, ἢν πρὸς ἀστὸν ποιέηται: οἱ δὲ φίλοι καὶ αὐτοὶ τὰς πίστις δικαιεῦσι σέβεσθαι. The Arabs respect a pledge of good faith [pistis, Ionic plural] like those who do so to the greatest degree. They make these pledges in the following way. When two men wish to make a loyal bond [ta pista], another man standing between the two cuts with a sharp stone the part inside the hand by the thumbs of those who are making the pledge, and then, taking a patch from the cloak of each, smears with the blood seven stones that are lying in the middle, and while he is doing this he summons Dionysus and Urania. When he has finished doing these things, the man who has made the pledge commends the foreigner [xeinos] or townsman [astos], if he is making it with a townsman, to his friends [philoi], and the friends themselves think it right that the pledges be respected. Rituals of this type are familiar from many societies and friends who pledge themselves in this way are sometimes referred to as ‘blood brothers’, perhaps with the idea that the exchange of blood renders them kinsmen. But if we step back from the language of sacraments, we may see in this performance an intimation of extended cognition, in which friendship is not only understood as involving the minds or intentions of the individual partners but is expressed also as an act that includes human witnesses, the gods, bodily fluid and inanimate material. I do not believe that Greeks in the classical period inaugurated or signified friendships by means of such rituals (contra Herman 1987; but see Konstan 1997: 36–7, 83–7). But they did regard friendship as an intense and indeed transpersonal relation, as suggested by the language of a single soul in two bodies and a friend as

214

dav id k o ns t an

another self.5 Friendship is an ideal instance of the social distribution of cognition, for it is by nature a relationship that depends on a special kind of reciprocal consciousness and is not intelligible solely as a representation in the mind of a single individual. I believe that it was treated as such by ancient thinkers and above all by Aristotle, who produced the most detailed and illuminating analysis of the subject. In so doing, he placed on a philosophical footing the intuition of a coalescence of selves, for all that it seems foreign to the individualism characteristic of modern thought. In the Nicomachean Ethics, Aristotle offers a preliminary definition of friendship as follows (Eth. Nic. 8.2, 1155b31–1156a5, my translation): τῷ δὲ φίλῳ φασὶ δεῖν βούλεσθαι τἀγαθὰ ἐκείνου ἕνεκα. τοὺς δὲ βουλομένους οὕτω τἀγαθὰ εὔνους λέγουσιν, ἂν μὴ ταὐτὸ καὶ παρ᾽ ἐκείνου γίνηται: εὔνοιαν γὰρ ἐν ἀντιπεπονθόσι φιλίαν εἶναι. ἢ προσθετέον μὴ λανθάνουσαν; πολλοὶ γάρ εἰσιν εὖνοι οἷς οὐχ ἑωράκασιν, ὑπολαμβάνουσι δὲ ἐπιεικεῖς εἶναι ἢ χρησίμους: τοῦτο δὲ τὸ αὐτὸ κἂν ἐκείνων τις πάθοι πρὸς τοῦτον. εὖνοι μὲν οὖν οὗτοι φαίνονται ἀλλήλοις: φίλους δὲ πῶς ἄν τις εἴποι λανθάνοντας ὡς ἔχουσιν ἑαυτοῖς; δεῖ ἄρα εὐνοεῖν ἀλλήλοις καὶ βούλεσθαι τἀγαθὰ μὴ λανθάνοντας δι᾽ ἕν τι τῶν εἰρημένων. They say that one must wish good things for a friend for his sake, and they call those who wish good things in this way well disposed [eunous], even if the same does not arise on the part of the other; for goodwill [eunoia] in those who feel it mutually is philia. We must add that it must not escape notice, for many are well-disposed toward people they have not seen, but believe that they are decent and worthy, and one of them might feel the same thing toward him. These then would seem to be well disposed toward each other, but how might one call them friends if they escaped [the other’s] notice of how they were disposed toward one another? It is necessary, then, that they be well disposed toward one another and wish good things, not escaping [the other’s] notice, in regard to some one of the abovementioned kinds [i.e. usefulness, pleasure or goodness, which are the qualities that are lovable in themselves]. Now, emotions generally involve an object and more specifically this is often another human being, and in this sense the emotion may be said to extend beyond the mental functions of a single individual: we pity or envy others and cannot imagine either sentiment in a wholly asocial context. But we can nevertheless describe these emotions as interior processes in which the other person is represented ­­ in one’s  5

Aristotle applies the language of identity also to those related by blood, and especially the connection between fathers and offspring; fathers, he avers, love their children because they are part of themselves: they are somehow, Aristotle says, one, even though they are separate individuals (Eth. Nic. 8.12, 1161b16–33). See Gill 1996: 340–4, on the ideal of a shared life in Aristotle and classical culture generally.



one sou l in tw o b o di es 215

own consciousness.6 We take it for granted that other people are real and have minds of their own, but when it comes to explaining how pity arises, for example, we can set aside any reflections that those who are pitied may have on their own situation, or if we do take their thoughts into account, it is as one more objectively perceived datum. Shame, as Aristotle understands it, is perhaps a more complex matter, since it depends at least in part on how others view us: for example, we feel shame before stern or forbidding individuals if they judge our behaviour disapprovingly, but not in the presence of more lax or intimate associates who regard our actions indulgently, or (Aristotle maintains) in the presence of slaves, whose opinions free men were habituated to discounting. But even with shame, the judgements of others are distinct from our own; that is why our response can vary in accord with how we regard their view of us. Our shame is not the product of a joint cognitive process, or the ‘coordination of intentional activity in interaction’ (De Jaegher and Di Paolo 2007: 497). Friendship is a different matter. We may see why by examining the relationship between love, which Aristotle treats as an emotion in his treatise on rhetoric, and friendship. The question is tricky because the Greek word for friendship, philia, is the same as that for love more generally (though it is distinct from erōs, that is, erotic passion) and it is only the context that disambiguates it. When pressed to make the distinction clear, Aristotle resorts to an expression which, translated literally, means ‘the “to love”’ (to philein), the Greek way of creating a verbal noun equivalent to the English ‘loving’. Aristotle defines τὸ φιλεῖν (‘loving’) as ‘wishing for someone the things that he deems good, for the sake of that person and not oneself and the accomplishment of these things to the best of one’s ability’ (τὸ βούλεσθαί τινι ἃ οἴεται ἀγαθά, ἐκείνου ἕνεκα ἀλλὰ μὴ αὑτοῦ, καὶ τὸ κατὰ δύναμιν πρακτικὸν εἶναι τούτων, 2.4, 1380b36–1381a1).7 There is perhaps an acknowledgement of the irreducibly transpersonal nature of love in the stipulation that we must take into account the friend’s values and not just our own, in our desire to serve him or her in the way that love requires, but here again, as in the case of shame, this awareness can be construed as a representation within our own mind of the other’s preferences, as opposed to a common process of cognition. Indeed, the very stipulation that we consider the friend’s own values suggests that they may differ from ours, which is a sign of an admirable moral modesty but not of distributed cognition as such. Having defined ‘loving’, Aristotle then offers a second definition (2.4, 1381a1–2): ‘A philos [that is, a friend] is a person who is both loving [philōn] and  6

Distributed cognition differs crucially from this kind of internal representation; as Rogers (1997) explains, ‘the distributed cognition approach aims to show how intelligent processes in human activity transcend the boundaries of the individual actor. Hence, instead of focusing on human activity in terms of processes acting upon representations inside an individual actor’s heads the method seeks to apply the same cognitive concepts, but this time, to the interactions among a number of human actors and technological devices for a given activity.’  7 Cope, in his great commentary (1877: 42), translates ‘whatever we think good’, but the singular οἴεται goes better with the antecedent τινι; those who love, according to Aristotle, do not ­necessarily impose upon their friends their own idea of what is good.

216

dav id k o ns t an

loved in return [antiphiloumenos]’ (φίλος δέ ἐστιν ὁ φιλῶν καὶ ἀντιφιλούμενος). Then he adds: ‘Those who believe that they are so disposed toward one another believe that they are philoi [that is, friends]’ (οἴονται δὲ φίλοι εἶναι οἱ οὕτως ἔχειν οἰόμενοι πρὸς ἀλλήλους). This account of friends is plainly similar to the one Aristotle gives in the Nicomachean Ethics and shows that, within the category of loving, friends are a special case in which the sentiment is mutual; nevertheless, there is a subtle but, I think, significant difference between the two accounts, to which no one, so far as I know (myself included), has called attention. In the Ethics, as we have seen, Aristotle defines philia or friendship and insists that friends must not be unaware of one another’s goodwill; in the Rhetoric, however, he defines not friendship but a friend and is content with the epistemologically more limited stipulation that people who believe that their love is reciprocated believe that they are friends. Why this hint of subjectivity? I am now inclined to think that it is because Aristotle is concentrating here on love as an emotion rather than on friendship as a relationship: to believe that one is your friend, it suffices that you have a mental representation or belief in the other’s attitude toward you; but for friendship actually to exist, it is necessary that both parties in fact feel the same way and that both know this to be the case. The latter is a state of affairs that extends to two minds, not just one: in friendship proper, cognition is distributed.8 When Aristotle wishes to distinguish one-way affection from philia in the Nicomachean Ethics, he coins the term philēsis (8.2, 1155b27–34), at least in regard to inanimate things such as wine (as when we say ‘I really love wine’), since in this case there is neither reciprocal affection (antiphilēsis) nor even the wish for the thing’s good for its own sake. When it comes to liking people who do not like us in return, Aristotle provisionally adopts the term eunous, ‘well disposed’ or ‘bearing goodwill’, since in this case we do wish good things for the other’s sake, even if our sentiment is not reciprocated. As such, it corresponds to to philein or ‘loving’ as Aristotle defines it in the Rhetoric. Eunous is a stop-gap term, designed to distinguish unilateral affection from friendship proper, for which Aristotle here reserves the word philia.9 The reason why Aristotle bothers to invent or employ various expressions  8

When Aristotle treats philia as a virtue, that is, as a disposition rather than an emotion, he defines it as the mean between two extremes: flattery, which is the excess, and enmity or hostility (ekhthra), which is the deficiency. Here philia pertains to an individual philos – that is, one who behaves as a philos ought (Mag. mor. 1.31.1–2); cf. 2.11.6–7, 1208b36–7, where Aristotle argues that philia is used of our feelings toward god or inanimate things, such as wine, but adds that he is investigating not this type, but rather ‘that toward animate things, and moreover those capable of feeling philia in return [antiphilein]’ (τὴν πρὸς τὰ ἔμψυχα, καὶ πρὸς ταῦτα ἐν οἷς ἐστι τὸ ἀντιφιλεῖν). Aristotle may also, however, have used the language of belief in the Rhetoric because an orator needs to plant in the audience the conviction that the speaker is disposed to them in a friendly way, whatever the reality, whereas in the ethical treatises what matters is whether people in fact reciprocate friendship. (I am grateful to one of the referees for this suggestion.)  9 Aristotle could have employed here the participle ho philōn (or rather the plural, hoi philountes), as he did in the Rhetoric; compare Eudemian Ethics 7, 1236b3–5: ‘What is loved is dear to the one who loves it, but a friend is dear to the one who is loved and loves in return’ (φίλον μὲν γὰρ τὸ φιλούμενον τῷ φιλοῦντι, φίλος δὲ τῷ φιλουμένῳ καὶ ἀντιφιλῶν). Later in the Nicomachean Ethics



one sou l in tw o b o di es 217

(including to philein) to signify one-way affection is that classical Greek lacked, as I have said, a noun that uniquely designated friendship. So if he wants philia here to mean the relation we call ‘friendship’, he needs other terms for love as an emotion. All this has been said to show that Aristotle thinks of friendship as a phenomenon involving two minds – I say ‘phenomenon’ rather than ‘relationship’ because the latter suggests a self-subsisting entity beyond what the individuals think and feel, what we might call a status, as in the case of kinship. Martha Nussbaum has argued that, for Aristotle, ‘love, while an emotion, is also a relationship’. As she explains: ‘I may feel love for someone, or be in love with someone, and that love is itself an emotion . . .; but there is another sense in which love is present only if there is a mutual relationship.’ What is meant by love being ‘present’ is not entirely clear, but Nussbaum further observes that Aristotle uses the term ‘love’ or philia equivocally, ‘to name both an emotion and a more complex form of life’ (Nussbaum 2001: 473–4). This is on the right track, I think, and points toward the kind of participatory interaction that subtends a certain form of distributed cognition. But when Nussbaum goes on to affirm that ‘lovers will have emotions toward their relationship itself, and the activities it involves’, it appears that philia in the sense of friendship is assuming the status of an entity in its own right, however abstract (474).10 Aristotle, however, does not speak of friendship in this way and for good reason. In the Categories, Aristotle speaks of relations such as half and double, or greater and smaller; things can also stand in a spatial relation to one another. As he puts it: ‘Those things are called relative, which, being either said to be of something else or related to something else, are explained by reference to that other thing. For instance, the word “superior” is explained by reference to something else, for it is superiority over something else that is meant’ (πρός τι δὲ τὰ τοιαῦτα λέγεται, ὅσα αὐτὰ ἅπερ ἐστὶν ἑτέρων εἶναι λέγεται, ἢ ὁπωσοῦν ἄλλως πρὸς ἕτερον, οἷον τὸ μεῖζον τοῦθ᾿ ὅπερ ἐστὶν ἑτέρου λέγεται· τινὸς γὰρ λέγεται μεῖζον, Cat. 7, 6a38–40, trans. Edghill 1928). But relations are not substances, which are characterised by the fact that they have no contrary and do not admit of degree (Cat. 5). Friendship, however, does have a contrary, namely enmity, and friendship itself admits of degree, since it can be more or less intimate. It is hard to imagine experiencing love for a relation, as Aristotle understands the term. I myself once wrote that ‘the mutual love that obtains between philoi is better described, I believe, as a state of affairs, consisting simply in the fact that each loves the other. As a name for simple love, that is, the altruistic wish for the good of another, philia, like to philein, is a pathos [i.e. an emotion]; as a state of affairs obtaining between friends, it consists of two pathē, one for each philos’ (Konstan (9.5, 1166b30–1167a21), Aristotle contrasts eunoia with both philia and philēsis as rather a dispassionate form of affection (cf. Eth. Eud. 7, 1241a3–14). 10 Such a relation differs, I think, from an ‘encounter’, in which, as De Jaegher and Di Paolo (2007: 492) observe, ‘the agents sustain the encounter, and the encounter itself influences the agents and invests them with the role of interactors. The interaction process emerges as an entity when social encounters acquire this operationally closed organisation. It constitutes a level of analysis not reducible, in general, to individual behaviours.’

218

dav id k o ns t an

2006: 178). But I now think that this is inadequate as an account of Aristotle’s view. For Aristotle, friendship is primarily an activity, not just a state. It is true, perhaps, that people can in a sense be friends while asleep, just as one can be virtuous while asleep, as Aristotle remarks in the Nicomachean Ethics (1.5, 1095b30–1096a1): ‘possession of virtue seems actually compatible with being asleep, or with lifelong inactivity’ (δοκεῖ γὰρ ἐνδέχεσθαι καὶ καθεύδειν ἔχοντα τὴν ἀρετὴν ἢ ἀπρακτεῖν διὰ βίου). Aristotle adds that it is also compatible ‘with the greatest sufferings and misfortunes; but a man who was living so no one would call happy’ (κακοπαθεῖν καὶ ἀτυχεῖν τὰ μέγιστα: τὸν δ᾽ οὕτω ζῶντα οὐδεὶς ἂν εὐδαιμονίσειεν, trans. Ross 1925). But as Aristotle notes soon afterwards, virtue finds its expression in action (1.8, 1098b31–1099b2): ‘to this [i.e. virtue] belongs virtuous activity’ (ταύτης γάρ ἐστιν ἡ κατ᾽ αὐτὴν ἐνέργεια). What is more, it matters greatly, Aristotle affirms, ‘whether we place the chief good in possession or in use, in state of mind or in activity. For the state of mind may exist without producing any good result, as in a man who is asleep or in some other way quite inactive, but the activity cannot; for one who has the activity will of necessity be acting, and acting well’ (ἐν κτήσει ἢ χρήσει τὸ ἄριστον ὑπολαμβάνειν, καὶ ἐν ἕξει ἢ ἐνεργείᾳ. τὴν μὲν γὰρ ἕξιν ἐνδέχεται μηδὲν ἀγαθὸν ἀποτελεῖν ὑπάρχουσαν, οἷον τῷ καθεύδοντι ἢ καὶ ἄλλως πως ἐξηργηκότι, τὴν δ᾽ ἐνέργειαν οὐχ οἷόν τε: πράξει γὰρ ἐξ ἀνάγκης, καὶ εὖ πράξει, trans. Ross 1925). In the Eudemian Ethics (2.1, 1219a24–36, trans. Inwood and Woolf 2013), Aristotle states that ἔτι ἔστω ψυχῆς ἔργον τὸ ζῆν ποιεῖν, τοῦ δὲ χρῆσις καὶ ἐγρήγορσις: ὁ γὰρ ὕπνος ἀργία τις καὶ ἡσυχία. ὥστ᾽ ἐπεὶ τὸ ἔργον ἀνάγκη ἓν καὶ ταὐτὸ εἶναι τῆς ψυχῆς καὶ τῆς ἀρετῆς, ἔργον ἂν εἴη τῆς ἀρετῆς ζωὴ σπουδαία. τοῦτ᾽ ἄρα ἐστὶ τὸ τέλεον ἀγαθόν, ὅπερ ἦν ἡ εὐδαιμονία. δῆλον δὲ ἐκ τῶν ὑποκειμένων ἦν μὲν γὰρ ἡ εὐδαιμονία τὸ ἄριστον, τὰ δὲ τέλη ἐν ψυχῇ καὶ τὰ ἄριστα τῶν ἀγαθῶν, αὐτὴ δὲ ἢ ἕξις ἢ ἐνέργεια, ἐπεὶ βέλτιον ἡ ἐνέργεια τῆς διαθέσεως καὶ τῆς βελτίστης ἕξεως ἡ βελτίστη ἐνέργεια, ἡ δ᾽ ἀρετὴ βελτίστη ἕξις, τῆς ἀρετῆς ἐνέργειαν τῆς ψυχῆς ἄριστον εἶναι. ἦν δὲ καὶ ἡ εὐδαιμονία τὸ ἄριστον. ἔστιν ἄρα ἡ εὐδαιμονία ψυχῆς ἀγαθῆς ἐνέργεια. the function of the soul [is] to make a thing be alive, and . . . the function of being alive [is] a using and a being awake – sleep is a kind of idleness and rest. Hence, given that the function of the soul and of its virtue must be one and the same thing, its virtue’s function would be an excellent life. This, then, is the complete good, which is what happiness is. This conclusion is clear from what we have laid down, namely that happiness is the best thing, the ends are in the soul and the best of goods and the things in the soul are either a state or an activity. So since the activity is better than the disposition and the best activity belongs to the best state, it is clear from what has been laid down that the activity of the soul’s virtue is the best thing. And the best thing is also happiness. Happiness, then, is the activity of the good soul.



one sou l in tw o b o di es 219

Friendship, according to Aristotle, is either a virtue or is accompanied by virtue (Eth. Nic. 8.1, 1155a3–4) and in this respect is also actualised only in activity or performance. For Aristotle, friendship is something that one does – ‘to be a friend’ is not simply equivalent to the English ‘I am a friend’ or ‘we are friends’; rather, it is more like ‘I am being your friend’, which sounds odd in English but captures more precisely Aristotle’s meaning.11 Thus, Aristotle assumes that friends wish to spend the day together (sunēmereuein) and he regards living together (suzēn) precisely as the actualisation or activity (energeia) of friendship (Eth. Nic. 9.12, 1171b35). This seems like an inordinately demanding condition for being friends, but what Aristotle means is that we are only truly being friends when we are in one another’s company and this is what friends most want to do. This kind of shared life is what sustains the mutual awareness of affection that is the ground of friendship. In turn, the shared sense of love manifests itself in the desire to do all one can to realise for the other all that he (or she) may wish or think good. There is a constant back and forth, a continual reinforcement or expression of the friendship in which the relation in fact consists. Friendship does not reside in the consciousness or awareness of one of the friends, or even of both individually; it is rather a matter of joint awareness manifested through the flow of acts of affection that continually reaffirm and sustain the love. It is just for this reason that philia cannot survive long periods of separation. As Aristotle puts it (Eth. Nic. 8.5, 1157b6–14): ὥσπερ δ᾽ ἐπὶ τῶν ἀρετῶν οἳ μὲν καθ᾽ ἕξιν οἳ δὲ κατ᾽ ἐνέργειαν ἀγαθοὶ λέγονται, οὕτω καὶ ἐπὶ τῆς φιλίας: οἱ μὲν γὰρ συζῶντες χαίρουσιν ἀλλήλοις καὶ πορίζουσι τἀγαθά, οἱ δὲ καθεύδοντες ἢ κεχωρισμένοι τοῖς τόποις οὐκ ἐνεργοῦσι μέν, οὕτω δ᾽ ἔχουσιν ὥστ᾽ ἐνεργεῖν φιλικῶς: οἱ γὰρ τόποι οὐ διαλύουσι τὴν φιλίαν ἁπλῶς, ἀλλὰ τὴν ἐνέργειαν. ἐὰν δὲ χρόνιος ἡ ἀπουσία γίνηται, καὶ τῆς φιλίας δοκεῖ λήθην ποιεῖν: ὅθεν εἴρηται “πολλὰς δὴ φιλίας ἀπροσηγορία διέλυσεν”. As in regard to the virtues some men are called good in respect of a state of character, others in respect of an activity, so too in the case of friendship; for those who live together delight in each other and confer benefits on each other, but those who are asleep or locally separated are not performing, but are disposed to perform, the activities of friendship; distance does not break off the friendship absolutely, but only the activity of it. But if the absence is lasting, it seems actually to make men forget their friendship; hence the saying ‘out of sight, out of mind’. The interpretation I am advancing here is similar to that proposed by Zena Hitz, who writes: ‘By calling the friend another self, Aristotle is not appealing to a general, implausible analogy between oneself and one’s friend. Rather, he understands 11

One might say that in ordinary English being a friend is perfective in aspect, describing an achieved condition; in this respect, it is like the verb ‘to know’: one cannot say ‘I am knowing’, which is continuative (the same is true for the Greek verb meaning ‘to know’, which is in fact a perfect form, oida).

220

dav id k o ns t an

friendship to involve collaborative activity and so the friend is another self in the sense of being a helper rather than a mirror’ (2001: 10). As Hitz explains, Aristotle ‘understands friendship to involve collaborative activity’, that is, ‘collaborative thinking and perceiving’ (10–11). Thus, ‘Living together means shared activity’ (12). This, in turn, is the basis of the shared sense of self; as Hitz puts it, ‘This is not to say that our selves are conflated in some woolly pseudo-mystical way; our friends are us in a concrete sense, in that they share the activities that constitute what we are and help us with them. So Aristotle emphasises repeatedly by connecting friends with what it is to be or to einai [i.e. ‘the very being’] for a person’ (17). Aristotle argues further that friends also provide an essential means of acquiring self-knowledge, based precisely on the fact that a friend is another self. Thus, in the Magna moralia, Aristotle affirms that it is ‘a most difficult thing . . . to attain a knowledge of oneself’ (ἐπεὶ οὖν ἐστι καὶ χαλεπώτατον . . . τὸ γνῶναι αὑτόν), and yet ‘we are not able to see what we are from ourselves (and that we cannot do so is plain from the way in which we blame others without being aware that we do the same things ourselves)’ (αὐτοὶ μὲν οὖν αὑτοὺς ἐξ αὑτῶν οὐ δυνάμεθα θεάσασθαι (ὅτι δ᾿ αὐτοὶ αὑτοὺς οὐ δυνάμεθα, δῆλον ἐξ ὧν ἄλλοις ἐπιτιμῶμεν, αὐτοὶ δὲ λανθάνομεν ταὐτὰ ποιοῦντες)). Aristotle continues: ‘as then when we wish to see our own face, we do so by looking into the mirror, in the same way when we wish to know ourselves we can obtain that knowledge by looking at our friend. For the friend is, as we assert, another I’ (ὥσπερ οὖν ὅταν θέλωμεν αὐτοὶ αὑτῶν τὸ πρόσωπον ἰδεῖν, εἰς τὸ κάτοπτρον ἐμβλέψαντες εἴδομεν, ὁμοίως καὶ ὅταν αὐτοὶ αὑτοὺς βουληθῶμεν γνῶναι, εἰς τὸν φίλον ἰδόντες γνωρίσαιμεν ἄν· ἔστι γάρ, ὡς φαμέν, ὁ φίλος ἕτερος ἐγώ, 2.15, 1213a13–24, trans. Stock 1925, slightly modified). Aristotle does not make clear here just why we require a mirror in order to know ourselves, but in the Eudemian Ethics he seems to address this question, albeit in terms that are exceedingly obscure and variously interpreted by modern commentators. To begin with, Aristotle asserts: ‘It is manifest that [life] is perception and knowledge, and that consequently social life is perception and knowledge in common’ (φανερὸν οὖν ὅτι τὸ αἰσθάνεσθαι καὶ τὸ γνωρίζειν, ὥστε καὶ τὸ συζῆν τὸ συναισθάνεσθαι καὶ τὸ συγγνωρίζειν ἐστίν, trans. Rackham 1991).12 Aristotle goes on to claim that ‘the known and the perceived are generally speaking constituted by their participation in the “determined” nature, so that to wish to perceive oneself is to wish oneself to be of a certain character’ (τὸ γνωστὸν καὶ τὸ αἰσθητόν ἐστιν ὡς ὅλως εἰπεῖν τῷ κοινωνεῖν τῆς ὡρισμένης φύσεως: ὥστε τὸ αὑτοῦ βούλεσθαι αἰσθάνεσθαι τὸ αὑτὸν εἶναι τοιονδὶ βούλεσθαι ἐστίν). This explanation is far from perspicuous, but Aristotle attempts to clarify it by stating that ‘when perceiving one becomes perceived by means of what one previously perceives, in the manner and in the respect in which one perceives it, and when knowing one becomes known’ 12

For the textual problems, and an alternative reading to Rackham’s, see Stern-Gillet 1995: 54–7. Stern-Gillet concludes that for Aristotle, ‘the formation of primary friendship involves the joint becoming of each other’s self, while its continuing practice ensures a form of noetic communion’ (58).



one sou l in tw o b o di es 221

(αἰσθανόμενος μὲν γὰρ αἰσθητὸς γίνεται ταύτῃ καὶ κατὰ τοῦτο, καθὰ πρότερον αἰσθάνεται, καὶ ᾗ καὶ οὗ, γνωστὸς δὲ γινώσκων). He then specifies that ‘ “friend” really denotes, in the language of the proverb, “another Hercules” – another self’ (ὁ γὰρ φίλος βούλεται εἶναι, ὥσπερ ἡ παροιμία φησίν, ἄλλος Ἡρακλῆς, ἄλλος αὐτός), and he adds, ‘a friend really means as it were a separate self. To perceive and to know a friend, therefore, is necessarily in a manner to perceive and in a manner to know oneself’ (αὐτὸς διαιρετὸς εἶναι ὁ φίλος. τὸ οὖν τοῦ φίλου αἰσθάνεσθαι τὸ αὑτοῦ πως ἀνάγκη αἰσθάνεσθαι εἶναι, καὶ τὸ τὸν φίλον γνωρίζειν τὸ αὑτόν πως γνωρίζειν, 7.12, 1244b24–6, 1245a2–5, 7–10, 29–31).13 It is impossible to enter into the issues raised by this aspect of Aristotle’s thought here, but the passages cited above offer an idea of how his understanding of thinking is related to his idea of friendship. Catherine Osborne, for example, rejects what she calls ‘the fashionable view that Aristotle thinks that the good man gains self-knowledge from having friends’, and argues rather that ‘the value of friends lies in looking out together at a shared world of experience. A friend, I suggest, is an extended self, because he stands alongside me and together we become enlarged by appreciating what is good and suffering what is bad’ (2009: 1). As a result, ‘ “living together” is going to be a matter of shared feeling and shared observation’ (7). Osborne suggests further that ‘Aristotle does not think that there is a determinate but hidden self there to be discovered by some means’; rather, ‘the self becomes actual and determinate only when the agent or subject actualises their agency or subjectivity’ (10). We might say that the self is itself an emergent property of human interaction. The idea that cognition might extend beyond the boundary of our minds or brains would not have seemed as strange to thinkers in the classical world as it may to us. Many philosophers, including Aristotle himself, thought of perception, for example, as a two-way process (both intromission and extramission), in which the perceived object was as much affected as the perceiver.14 This is why Aristotle could seriously entertain the seemingly absurd notion that when a menstruating woman looked in a mirror, the mirror acquired a reddish tinge (On Dreams 2, 459b24–8 and 460a24–7). As Aristotle reports: ‘The cause is . . . that the eye is not only affected by the air but also has an effect upon it and moves it . . . The air . . . has a certain effect on the air on the surface of the mirror . . . and the air on the mirror affects the surface of the mirror’ (αἴτιον δ᾿ . . . ὅτι οὐ μόνον πάσχει τι ἡ ὄψις ὑπὸ τοῦ ἀέρος, ἀλλὰ καὶ ποιεῖ τι καὶ κινεῖ . . .· ὁ δ᾿ ἀὴρ . . . καὶ τὸν ἐπὶ τῶν κατόπτρων ἀέρα . . . ποιόν τινα ποιεῖ . . .· ὁ δὲ τοῦ κατόπτρου τὴν ἐπιφάνειαν, 13

See also Eth. Nic. 9.9, 1170a13–b19, on which Price (1989: 123–4) comments that friends ‘articulate [one’s] own thoughts’, just as by their actions they ‘realise [one’s] own preferences’; the result is that ‘each reveals the mind of the other’ in a way that they could not have achieved on their own. 14 Aristotle himself wavered on the matter of extramission, that is, the projection of visual rays in vision, but he endorses the view at Mete. 3.4, 373a35–b15 and various other places (e.g. Cael. 2.8, 290a17–24, Gen. an. 5.2, 780b33–781a13). Alcmaeon, Empedocles and Plato (Ti. 45b–c, 67d–e) subscribed to one or another version of extramission as well (see Rudolph 2016).

222

dav id k o ns t an

trans. Beare 1931).15 Similarly, we may suppose that friends are reciprocally aware of loving and being loved. As an activity, loving has an effect upon its object, which is in turn perceived by the one who loves; the exchange is continual. Aristotle’s reasoning in the Eudemian Ethics becomes, if not wholly intelligible, at least more pertinent if we take his ideas on perception into account. For Aristotle love is not simply a feeling, as modern definitions suggest, as in the second edition of Webster’s New International Dictionary (1959), which defines ‘love’ as ‘a feeling of strong personal attachment’ and ‘ardent affection’. Aristotle says nothing of the sort, but mentions instead benevolent intentions and, to the extent possible, acts in accord with those intentions. Rather than frame friendship as an interior sentiment, or even one that is experienced simultaneously by two discrete individuals, Aristotle sees it as a function of two reciprocally aware people who are constantly engaged in manifesting their love through actions.16 Given that friends, according to Aristotle, share tasks and achieve goals together, we might compare their interaction to group cognition, as in the case of multiple members of a crew piloting a ship.17 But a closer analogy is Andy Clark’s notion of ‘glue and trust’ to characterise bonds that obtain between people who are reliably available to one another and have full confidence in each other; these are the conditions, he argues, under which an external agent may be conceived of as part and parcel of one’s extended cognition.18 Where Clark points to a long-term marriage as the example of such a relationship, Aristotle would have thought first and foremost of friends, that is, one’s most intimate companions: they are always there (they spend their days together) and one trusts them implicitly. Trust, too, had a wide and, one might say, transpersonal sense in classical Greek. Consider the procedure by which the Arabs pledged trust and friendship, according to Herodotus. The Greek word pistis, rendered above as ‘a pledge of good faith’, has a variety of connotations: it can designate the trust or confidence we place in others, or, contrariwise, it may signify trustworthiness in the sense of the honesty or reliability that inspires others to trust in us. It can also refer to a pledge or guarantee, or to the thing that is given in trust or entrusted to another. In a slightly extended sense, it can refer to a logical or mathematical proof or confirmation (all these senses pertain also to the Latin fides).19 Rather than register surprise at the breadth of its 15

For further discussion, see Preus 1968; Kent Sprague 1985; Van der Eijk 1994: 167–93. Farenga (2006: 347) speaks of the intersubjective identity that constitutes the citizen self in ancient Athens, bound by ties of love: the citizen subject is ‘not a corporeal individual but an intersubjective creature who is plural and decorporealised’. See further Budelmann, this volume. 17 This example, along with cooperation in the cockpit of an aeroplane, was proposed by Hutchins (1995: ch. 3), and in various articles. 18 See Clark 2014; cf. Clark and Chalmers 1998: 17: ‘In an unusually interdependent couple, it is entirely possible that one partner’s beliefs will play the same sort of role for the other as the notebook plays for Otto [Otto suffers from Alzheimer’s and enters information in a notebook that he always carries with him]. What is central is a high degree of trust, reliance, and accessibility.’ 19 On pistis in classical Greek social and economic relations, see Johnstone 2011; for the early Roman Empire, see Morgan 2015: esp. 55–9 on fidelity between friends. 16



one sou l in tw o b o di es 223

usages, or treat the word as polysemous, subdividing its senses under various subheadings as dictionaries do, it may be preferable to see a semantic dispersion of the term pistis that corresponds to the amplitude of the idea of friendship, of which trust is the hallmark. The connotative penumbra of the word is really more a matter of perspective or focalisation: just as a friend is one who loves and is loved in return, so trust is embodied in both parties to a friendship, and in the loving acts that they continually perform as the active manifestation of their affection. Seen this way, it may seem less strange that when people pledge friendship a third party should anoint stones with their blood, and that each of the avowers (as I take Herodotus to mean) should include the other among his group of friends. The practice acknowledges the social character of friendship, to be sure, but it also implicitly recognises that friendship does not take place in the mind of an individual but in a collective awareness of the bond, the stones being as much a part of this cognition as the pencil or blackboard is of the thought processes of mathematicians as they compute the solution to a problem. This interpretation is preferable, I think, to older tendencies to view such rituals as establishing a fictive kinship or symbolising objective reciprocal obligations, irrespective of the feelings of the parties pledging their good faith. That love may be understood as uniting people in a common sensibility that seems to transcend individual identity has always been a theme of literature and religion. In the modern world, the idea has found expression in various ways. Axel Honneth, Professor of Philosophy at the University of Frankfurt and Research Director of the Institute for Social Research (the centre of activity of the so-called ‘Frankfurt School’), notes that the American philosopher John Dewey ‘employed the term “interaction” to indicate that our everyday activity is not characterised by a self-centred, egocentric stance but by the effort to involve ourselves with given circumstances in the most frictionless, harmonious way possible’, and adds that ‘human behavior is distinguished by the communicative stance achieved through taking over a second person’s perspective’ (2005: 111, 114). So, too, Angelika Krebs surveys a variety of theories concerning love and concludes that it is best understood as a dialogue (along the lines of Martin Buber’s ‘I and thou’): ‘A person loves another when he shares feelings and actions with her in her entire particularity as an end in herself.’20 Krebs concludes her paper with a citation from Rainer Maria Rilke’s ‘Liebeslied’ (‘Love Song’, vv. 8–13, trans. Crego 1999): Doch alles, was uns anrührt, dich und mich, nimmt uns zusammen wie ein Bogenstrich, der aus zwei Saiten eine Stimme zieht. Auf welches Instrument sind wir gespannt? Und welcher Geiger hat uns in der Hand? Oh, süßes Lied. 20

Krebs 2009: 741: ‘Eine Person liebt eine andere, wenn sie Gefühle und Handlungen mit ihr in ihrer ganzen Besonderheit selbstzweckhaft teilt.’

224

dav id k o ns t an

And yet everything which touches us, you and me, takes us together like a single bow, drawing out from two strings but one voice. On which instrument are we strung? And which violinist holds us in his hand? O sweetest of songs. The theory of distributed cognition goes beyond such formulas, suggestive as they are, and seeks to provide a framework for understanding cognition as ‘an emergent property of interaction’, where interaction takes the form of ‘coordination and cognitive coupling’.21 I believe that Aristotle’s treatment of friendship can plausibly count as an early case study in distributed cognition, one that emerged in a social context and philosophical tradition different from our own, but which perhaps for that reason was more free to see beyond the conception of thought as a property of isolated minds that operate on nothing more than mental representations. On this view, indeed, virtue itself may ultimately be a function of transpersonal cognition.

21

Liu et al. 2008: 1177; cf. De Jaegher and Di Paolo 2007: 493 on social interaction as ‘the regulated coupling between at least two autonomous agents, where the regulation is aimed at aspects of the coupling itself so that it constitutes an emergent autonomous organisation in the domain of relational dynamics, without destroying in the process the autonomy of the agents involved’.



d ist rib uted c ognit ion and i t s di s co nt ent s 225

13 Di st r i but e d C o g n i t i o n and its D is contents : A   Di a l og ue a c ro s s Hi s t o ry a nd Artis tic G enre Thomas Habinek and Hector Reyes

Writing a history of distributed cognition requires equal attention to cognition and history. Theories of cognition are diverse and contested, in antiquity as today; so are theories and practices of history. Loosely speaking, we might differentiate between history as teleological narrative of actions and events from past to present, and history as retrospective identification and analysis of salient aspects of earlier human experience. If we adopt the latter approach, we are not out of step with our respective disciplines of Classics and Art History, which have incorporated the study of ideological and other frameworks that condition prior instances of engagement with an earlier past. However, this general disciplinary project has come to something of an impasse, leading either to discontinuous accounts of discrete instances of ‘reception’ or to the implicit adoption of past intellectual models, recapitulating the nineteenth-century ‘isms’, such as formalism, historicism and empiricism, that have long structured our respective disciplines. One way out of this impasse is to embrace a new type of intellectual history that analyses the encoding of knowledge that simultaneously produced the objects of our study and the analytical tools used to study those objects. It is our contention that a history of distributed cognition can contribute to this form of intellectual history by asking what key versions of cognition and knowledge production become apparent to the later observer who seeks them out, and why they take the shape they do.1 There are at least three methodological benefits for the historian in treating cognition as distributed, that is, not limited to discrete agents, artefacts or temporalities. First, and most important, with cognitive or intellectual agency understood as distributed, intellectual history is freed from the expectation to privilege a single analytical framework as the lens through which materials must be studied. Neither  1

Although it is not our goal to adjudicate among the various versions and aspects of theories of distributed cognition, we would like to acknowledge here the benefit we have derived from the following works: Hutchins 1995 and Friedenberg and Silverman 2016 on distributed cognition in general; Donald 2001, Theiner 2011 and Gallagher 2013 on social cognition; Spivey 2007 on continuity between brain, body and environment; Hutto and Myin 2012 on non-propositional thought; Sterelny 2010 on environmentally supported cognition.

226

thoma s habin ek a nd h ect o r reyes

the analytical framework of the period being studied, nor the nineteenth-century frameworks that founded the discipline, nor contemporary models of thought need take priority in the analysis of cultural objects. Secondly, and following this first point, a broadly diachronic analysis that incorporates the study of different types of materials from different historical moments becomes possible: diverse ‘sources’, in all their historical specificity, can be treated as multiple and interconnected approaches to a single cognitive problem rather than one-off moments of reception. And lastly, the historian can offer to scientists a model of putting into scholarly practice the very theories of cognition that they have developed – as evidenced by our own thoroughgoing collaboration in preparation of this chapter. To exemplify our approach, we pose the question: how and why, in the classical European tradition, did cognition ever come to be considered as non-distributed? What displacements and exclusions took place to permit the conceptualisation of thought as a singular, disembodied process? How, in particular, did an understanding of ancient cognitive theory as presupposing singular agents take hold? The sources we draw on in our inquiry are visual as well as verbal (and could be expanded to incorporate embodied practices as well), and with respect to both we give as much attention to the affordances provided by artistic form as to propositional content – in line with our own understanding of cognition as at least potentially cross-modal and non-propositional. We treat two early modern paintings (Raphael’s School of Athens and Poussin’s Landscape with Numa Pompilius and the Nymph Egeria) and the ancient texts with which each is in dialogue (including works by Plato, Cicero, Vergil, Ovid and Servius) as constituting a network of knowledge production for the contemporary historian, displacing the unique agency often assigned to each and allowing the system to generate knowledge inaccessible to any individual participant within it. To put it differently, we seek a distributed history of ancient cognition in its distributed and non-distributed forms.

Raphael, The School of Athens Painted between 1509 and 1511 and occupying almost an entire wall in the Stanza della Segnatura in the Vatican, Raphael’s fresco The School of Athens (Plate 3) depicts fifty-eight figures from the history of philosophy clustered into a number of distinct groups in an architectural setting that evokes various classical and classicising structures and motifs.2 Although not all of the individual figures have been securely identified, the visually dominant central group is organised around depictions of Plato and Aristotle, while one group in the lower left has Pythagoras at its centre and one in the lower right depicts several figures, perhaps including Euclid, concentrating on a sketchboard while others look at, and perhaps speak to, one another while holding globes. The painted human figures function both as analogue for an extra-pictorial reality, in the sense that they represent individual  2

For date, circumstances and relationship to broader artistic programme of the Stanza, see JoostGaugier 2002. For literary and architectural prototypes, see Most 1996.



d ist rib uted c ognit ion and i t s di s co nt ent s 227

philosophers, and as allegory, in that collectively in their disposition around Plato and Aristotle they become the concrete manifestation of an abstract idea of philosophy as an enterprise uniquely associated with a particular intellectual genealogy and individual agents within it.3 Even the corporeal dispositions of the individual figures are intended as visual equivalents to discursive propositions: for example, Plato’s upward gesture can be understood to refer to the metaphysical affirmations that were thought to characterise his major dialogues, while the downward gesture of Aristotle evokes the empiricism of his treatises. Despite the lively and communal nature of the philosophical activity depicted in the fresco and the inclusion of external props for cognition, such as books, sketchpads and globes, the painting presents philosophical thought as an enterprise detached from specific human contexts and carried on primarily as a transmission of propositional knowledge from teacher to student or leader to follower. To the extent that cognition is depicted as having an externalised component, it is of a purely solipsistic sort, with the individual thinker using the external prop to advance his own internalised cognitive process.4 As has long been noted, few if any of the thinkers depicted by Raphael could have been in the same place at the same time, with identifiable figures ranging in date from Pythagoras (sixth century bce) through to, on some accounts, Averroës (twelfth century ce). Moreover, while communication – sometimes oral, sometimes visual – is depicted within each cluster of figures, each cluster as a whole is isolated from all the others. Intellectual history is imagined as taking place as much through individualised engagement with the textualised circulation of ideas dislodged from space and time as in the impossible conversations of historically disembedded actors. From an aesthetic standpoint, an aura of intellectual abstraction is made visually manifest, permeating the entire piece from its elevation above the viewing line of the perceiver to the smoothness of its finish. But most of all, it is the circumscription of forms – the linear outlines that describe the boundaries of the bodies, the intensity of the hues that differentiate and harmonise different groupings of figures, the compositional containment of so many bodies within the semi-circular space – that is most impressive. It is not novel to say that Raphael has demonstrated that disegno, both the abstract principle of design and the artistic practice of linear drawing, could participate within courtly forms of knowledge. But Raphael does more than demonstrate the compatibility of modes of knowledge; Raphael calls attention to the homogeneous and analogical work of circumscription across different formal registers in order to make a statement about the artistic act. It is a self-conscious performance that reorganises the viewer’s attention. The recognition by the viewer of the singular act of visual binding will ultimately correspond to the work’s intended use as a visual archive of information about philosophy to be drawn upon, in particular, by the humanists affiliated with the court of Pope Julius II. Such viewers  3

On the displacement of an older allegorical depiction of philosophy as a single, emblematic female figure, see Most 1996.  4 See Tollefsen 2006 on solipsistic versus collective coupling of internal and external.

228

thoma s habin ek a nd h ect o r reyes

are likely to have taken great pleasure in their ability to identify the figures within the painting (that is, to demonstrate their mastery of the archive) and in the ways that their own courtly configurations mirrored the circumscribed groupings of the esteemed philosophers.5 Despite this archival function, the painting nonetheless reinforces the ‘illusory conceit’ of consciousness as singular and unidirectional6 and promotes an understanding of thought as a collection of propositions rather than a dynamic process of engagement.

Plato, Protagoras The preference for internalised cognition characteristic of The School of Athens assumes greater significance when we consider the painting in relation to a likely pre-text, namely the gathering of sophists imaginatively recounted in Plato’s Protagoras (Most 1996). There a specific locale is identified (the house of Callias) along with a specific occasion (Protagoras’ visit to Athens), with the audience of the dialogue asked to imagine that the numerous named figures were all present in the same place at the same time – a collocation that classical scholars have judged chronologically plausible (Nails 2002). In other words, the form of the dialogue allows for at least the possibility of knowledge construction as a situated, embodied practice. When Socrates and his unnamed companion arrive, they enter directly into the scene of the action and observe the participants while standing on the same level. Although the sophists and their admirers initially fall into three distinct clusters viewed with wonder by Socrates – the perambulating Protagoras and his entourage, the seated Hippias and his listeners, and Prodicus, still in bed, communing with his followers – they swiftly reorganise themselves into a single group in order to attend to the discussion between Socrates and Protagoras (Pl. Prt. 317c). Nor do they remain silent observers, but instead interpose questions and comments throughout, with an unusually large number of characters having a speaking role in the dialogue. None of which is to suggest that Plato in the Protagoras is any more an advocate of distributed cognition than Raphael is. Although there is lively conversation involving multiple characters, the core discussion between Protagoras and Socrates founders on each party’s refusal to permit the style of argument preferred by the other. In cognitive terms, each fails ‘to maximally exploit the particular capacities of the other’.7 When Hippias brokers a compromise, approved by all present, to the effect that each will speak in a modified version of his preferred style, with an  5

On the humanist context of the Julian court that would have influenced the programme, see JoostGaugier 2002: 81–114.  6 Anderson (2015: 6) writes of ‘how the illusory conceit of a single stream of consciousness is linked to the illusion of subjectivity as transparent and fixed’.  7 We borrow a phrase from Clark (1997: 220), who refers to the relationship between the brain and external media more generally.



d ist rib uted c ognit ion and i t s di s co nt ent s 229

arbiter to be selected who will curb any excesses (Pl. Prt. 337e–338b), Socrates balks, turning a disagreement about discursive form into an ethical and ontological struggle by asserting that the wise man should never have to submit to one who is less wise. In effect, Socrates denies one of the basic principles of distributed cognition, namely that a system may well have properties not held by any individual within it (Hutchins 1995). His stubborn refusal is noteworthy, since it comes after, without explicitly confronting, Protagoras’ ‘magnificent performance’ (Socrates’ own description, Pl. Prt. 328d), a spellbinding speech in which the sophist asserts that the gods Zeus and Hermes gave the capacity for political wisdom and civic virtue to all human beings, and that deliberation about public affairs therefore requires consultation of the many as well as the few (Pl. Prt. 322c–323a). In effect, the Protagoras depicts a struggle between a speech-making show-off, who nonetheless stresses the teachability of virtue and the benefits of a distributed system of information-gathering,8 and a seeming lover of conversation and dialogue who in fact only seeks knowledge along a single pathway – his own. Yet it is a struggle made possible by the historically specific, even if fictional, gathering of sophists and their admirers, all of whom communicate engagingly with one another. Intentionally or not, Raphael’s School of Athens eliminates even those traces of distributed cognition that survive Socrates’ repudiation by transferring the energy and lively appeal of the sophists’ colloquy at the house of Callias to a retrospective philosophical genealogy that all but erases the sophists from the great fresco of history. A painting such as Raphael’s cannot be reduced to a propositional claim about the nature of philosophical activity; but precisely because of its vividness, memorability and accrued cultural authority, it shapes the viewer’s perceptions and thoughts about cognition in specific ways even as it seeks to advance a particular version of the history of ideas. Such distributed cognition as is present in the Protagoras disappears in Raphael’s extraordinary depiction of the transhistorical force of individualised thought communicated through unidirectional instruction, all in a painting remarkable for, among other things, its strong single-point, linear perspective that converges on a vanishing point just behind the dominant figures of Plato and Aristotle. The painterly project of corporeal disposition ‘distributes’ its cognising figures across the fresco in a uniform, even and coherent manner, effectively occluding other potential forms of artistic distribution that are neither linear nor strictly circumscribed. Raphael’s representation of Plato and Aristotle’s conversation, around which all bodies, ideas, objects and space have been organised, replaces the unresolved debate between Protagoras and Socrates represented by Plato. And with that erasure and substitution, Socrates is granted a victory withheld in the more tentative denouement of Plato’s dialogue.

 8

On this aspect of classical Athenian democracy, see Ober 2008. Cf. Budelmann, this volume (esp. pp. 200–1 on Arist. Pol. 3.11, 1281a40–b10).

230

thoma s habin ek a nd h ect o r reyes

Poussin, Landscape with Numa Pompilius and the Nymph Egeria A radically different approach to cognition underlies both the representational content and the formal strategies found in an early work of the French peintre-­ philosophe Nicolas Poussin, known as Landscape with Numa Pompilius and the Nymph Egeria (Plate 4).9 Painted between 1627 and 1630, probably for Poussin’s patron Cassiano dal Pozzo while the painter was still resident in Rome, the modest-sized canvas (77 cm wide and 105 cm high) evokes multiple ancient narratives that have the second, peace-loving king of Rome consulting with the water nymph Egeria, who assists him in the task of developing religious rituals and other institutions for the benefit of the emerging city-state. Although the story of Numa and Egeria was recounted or alluded to by Livy, Ovid, Plutarch, Juvenal and other classical authors, Poussin’s composition corresponds to no single version. Indeed, it is impossible to read the painting as illustration of any given moment in the shared, underlying narrative. Is Numa dead or alive? Is this the king’s first encounter with Egeria or one of many? Is one figure greeting, or saying farewell to, the other; or are they engaged in a moment of shared contemplation? Why is there a naked shepherd playing pan-pipes when no such figure is mentioned in the ancient sources pertaining to Numa?10 What does the story of Numa have to do with the branch he is shown as touching or grasping in the painting? The indeterminacy of the narrative content of the painting invites the viewer to enter into a network of possible associations that emerges from other aspects of Poussin’s artistry. For example, the setting depicted is clearly pastoral, despite alternative classical localisations of Numa and Egeria’s assignations in a more urban context. By evoking rural Aricia, Poussin allows the viewer to associate Egeria not only with Numa, but also with Virbius, or the reborn Hippolytus, who appears to her there, according to both Vergil and Ovid, and, citing his own experience of rebirth, exhorts her not to grieve at Numa’s death. Does the theme of return from the world of the dead thus associated with Egeria help to explain the ghostly glow emitted by Numa in the painting and his grasp of a gold-coloured bough, itself an ancient talisman of transit between the land of the living and the abode of the dead? Or is it the branching ‘y’ at the base of the bough that should draw the viewer’s attention? According to Servius, the y-shape of the golden bough is a symbol of the teachings of Pythagoras, not least of which was that of the near-endless recycling or transmigration of souls across generations. And indeed, Pythagoras’ great speech in Ovid’s Metamorphoses is inserted within the poem’s narrative of Numa and Egeria, where it is offered as a response to Numa’s quest for understanding of the nature  9 10

Basic information and bibliography on the painting at Wright 2007: 63. Häfner (2011: 23–37) argues that the figure depicts Hippolytus, or Virbius, but other than the critic’s desire to find a one-to-one association of image and object of representation, there seems no basis for this singular identification.



d ist rib uted c ognit ion and i t s di s co nt ent s 231

of reality (quae sit rerum natura, Ov. Met. 15.6). According to one scholar, the leafy object depicted in Poussin’s painting is not the golden bough but an olive branch, in reference to a passage in Ovid’s Fasti that has Numa recruited to Rome from ‘olive-bearing fields’ (oliviferis . . . ab arvis, Ov. Fast. 3.151). Yet the same passage offers Pythagoras and Egeria as equally likely sources of Numa’s wisdom. Whether golden bough or olive branch, the object in question raises issues of life after death and the cyclical, as opposed to unidirectional, nature of history. Moreover, both the golden bough, as discussed by Servius, and the olive branch proposed by Häfner figure in narratives of kingship – the sacral king who must fight for his life in the mysteries at Aricia, or the legendary king Numa, summoned to Rome by an eager populace. Poussin’s depiction of a naked shepherd playing on pipes, nowhere mentioned explicitly in the ancient sources pertaining to Numa and Egeria, is a further invitation to seek subtle connections within the variegated record of antiquity. The story of Numa and Egeria, as it turns out, is also one of entry into a world of music. Egeria is sometimes one of the Muses, sometimes their companion. As a water goddess she herself makes sweet sound,11 much like her fellow Roman water goddesses, the Camenae, who come to be seen as indigenous Muses.12 In addition, the rituals founded by Numa include musical practices: Ovid and Livy have him establishing the annual dance of the Salian priests (a late grammarian even says he led the dancers himself), while Dionysius of Halicarnassus describes Numa as having ‘attuned the whole body of the people, like a musical instrument, to the sole consideration of the public good’.13 On this matter, as in the case of the golden bough, Roman ritual and legend converge with the Greek philosophising of Pythagoras, whose theoretical understanding of music in relationship to the movements of the cosmos was well known in seventeenth-century Italy as in antiquity.14 With his introduction of a pastoral musician, Poussin again invites the viewer to participate in an assemblage of knowledge about antiquity from different sources and to reflect on the different means through which knowledge is constructed and shared, including ritual practice, legend, music, philosophising, quiet contemplation and, of course, painting. In this respect, he revives a Ciceronian understanding of cognition as an inherently cross-modal process in which artists, musicians and users of language participate equally.15 If the representational content of Poussin’s painting signals the fluid and dynamic nature of thought, its more strictly formal characteristics extend the invitation to participate in a distributed set of perceptions and recognitions. The three central figures in the painting – Numa, Egeria and the musician – yield no 11

Olli respondit suavis sonus Egeriai, Ennius 113 (Skutsch), as cited by Varro, De lingua latina 7.42. For comprehensive discussion of sources pertaining to the Camenae, see Hardie 2015. 13 Ov. Fast. 3.259ff.; Livy 1.20; Diomedes, Grammatici Latini 1.476 (Keil); Dion. Hal. Ant. Rom. 2.62.5. See also Zorzetti 1991; Habinek 2005b: 8–33. 14 McClary 1995. On interest in Pythagoras among acquaintances of Poussin in Rome, see Dempsey 1966; Häfner 2011. 15 For further discussion, see Habinek 2017. 12

232

thoma s habin ek a nd h ect o r reyes

single focus. The distinctive brushwork used to differentiate them from their surroundings invites reflection on their substantive or material qualities; their visual relationship is unambiguous, the nature of their commerce open to interpretation. In contrast to the single-point, linear perspective of Raphael, Poussin employs so-called ‘perspective aérienne’, in which colours become less saturated and begin to mingle with those of the atmosphere as objects become more distant from the viewer. The technique allows for a perception of depth and recession without a focus on any single figure or object or an evident vanishing point. Recession of space is here achieved in a counter-intuitive manner, with a brilliant but pale white indicating sunlight falling on the distant mountains and fields, while a purer but less bright mixture of green and brown tints makes up the small hill that establishes the middle ground, and rich dark browns constitute the ground plane in the foreground. Only by reversing expectations of bright foreground and dark background can the viewer make sense of the depicted space. With regard to form, as with content, we are invited to think and perceive along with the painting. Against the backdrop of receding space, Poussin establishes a unified composition by isolating three subtly similar visual and tactile events, which form their own triangular pattern. Numa touches the bough, in a gesture that constitutes the primary action and narrative focus of the painting. In the exact centre of the painting, bright light coming from the horizon caresses overhanging branches, calling attention to their ever-so-slight separation from a shape to be understood as the rim of a distant mountain. Similarly, Egeria wears a crown of cool, jade-coloured leaves that, absent the cognitive effort of reading them differently, seem to dangle from among the slightly darker, sage-coloured, matte leaves of the bush behind her. Just as the visual content of Poussin’s painting, especially in relation to its classical sources, refuses to privilege any one version of the relationship between Numa and Egeria, so too his formal strategies – including a multiply focused array of figures, shimmering play between distance and proximity, and a painterly investigation of the relative significance of touch and sight in the identification of objects – all situate painter, painting and viewer in a perceptual process that is dynamic, interactive and variously distributed. Unlike Raphael’s School of Athens, where perception is expected to lead to propositional knowledge, and brilliant painterly technique serves to crystallise the relationship between the two, in Poussin’s Numa, the embodied processes of painting, touching and seeing are all treated as part of an extended cognitive ensemble. The viewer’s tools for understanding are activated by the painting, but no decisive version of the depicted scene is on offer. Is Egeria grieving, or at peace? Has Numa taken the bough yet? Is this a pastoral scene or a Pythagorean universe? The painting is not an archive of facts or propositions, but a participant in the recalibration of a cognitive system that embraces as many viewers as avail themselves of its affordances. Like Numa’s tenuous grasp of the talismanic bough, the delicate shades and subtle arrangement of the painting as a whole leave its full significance in suspense, as if waiting for the viewer to grasp back.



d ist rib uted c ognit ion and i t s di s co nt ent s 233

Cicero, De oratore If Raphael’s School of Athens and Poussin’s Numa Pompilius and the Nymph Egeria can be said to represent and prompt contrasting modes of cognition, Cicero’s De oratore, like Plato’s Protagoras, acknowledges the co-existence of distributed and solitary endeavours, albeit with a distinct tilt in the direction of distribution. In effect, the De oratore16 embodies a paradox: on the one hand, it presents views about oratory and the education of the orator as emerging only through the dialogic, collaborative process of its numerous interlocutors; on the other hand, the treatise takes as its goal the training of an idealised individual orator who has a comprehensive understanding of all matters that could possibly pertain to the art of public speaking (‘omnium rerum magnarum atque artium scientiam consecutus’, Cic. De or. 1.20). Rediscovered in 1421, De oratore became a crucial text in humanist education and in time provided an important model for Poussin in the articulation of his famous theory of the modes of musical and artistic composition.17 As much as any ancient text, De oratore laid the groundwork for a still influential conception of training in the liberal arts as politically, socially and ethically beneficial. Its internal tensions both illustrate and authorise the conflicting understandings of cognition and knowledge acquisition that shape the European intellectual tradition. Reciprocally, attention to the competing models of cognition present within De oratore enriches our understanding of the overall aim and function of the surviving treatise. De oratore introduces the theme of cognition with its opening phrase: cogitanti mihi, ‘as I am thinking’. Cicero is thinking, and asks his brother and his readers to think along with him, about men of old, who were able to lead busy lives without risk or peaceful lives without loss of dignity. Cicero reports that he himself had hoped to enjoy a sort of leisure that would allow him to revisit, together with his brother, the investigations that had jointly concerned them when they were young (‘ad eas artes quibus a pueris dediti fuimus celebrandas inter nosque recolendas’, Cic. De or. 1.2). But historical circumstances intervened, keeping Cicero so busy that he can only with difficulty eke out the time to write, albeit in response to his brother’s exhortations. Solitary, individualised writing is thus presented as less desirable than shared attention and exchange of ideas, even as such writing introduces and generates the conversations among men of an earlier generation that occupy the bulk 16

On the date of the dialogue and the identity of the interlocutors, see the concise discussion of Mankin 2011. Hall 1994 is helpful in untangling the complexity of the argumentative structure of De oratore, but underestimates the role of collaboration and distribution in producing the overall impact of the treatise. Like many scholars, he seems to assume that Cicero must be presenting a singular authorial argument rather than illustrating a shared mode of cognition. More recently, Kenty has argued that in De oratore Cicero uses ‘the form and content of his dialogue to portray eloquence as a subject of conversation and debate celebrating the polymorphic and, so to speak, interdisciplinary nature of eloquence’ (2017: 355). 17 On the importance of De oratore in early modern Italy, see Reeve 1996; Baker 2015. On Poussin’s use of De oratore and other works by Cicero in his theory of the modes, see Reyes 2009.

234

thoma s habin ek a nd h ect o r reyes

of the three-book dialogue. The value of shaping ideas through a single authorial will or intellect is thus cast into some doubt already at the opening of De oratore. Indeed, Cicero proceeds to explain that he will limit his subject matter to topics deemed relevant ‘by the shared judgment of the most eminent men, after investigation and extensive discussion’ (Cic. De or. 1.22). Nor will he organise his work as ‘a list of precepts’ or resort to ‘his own interpretation’ of Greek texts (Cic. De or. 1.23). Instead, he will structure the treatise as a conversation among individuals from the Roman past who were both eloquent and distinguished (Cic. De or. 1.23). In effect, he will minimise the disadvantages of solitary cognition by remembering (memini, Cic. De or. 1.24) what Gaius Cotta told him about a conversation held by Crassus, Antonius, Scaevola and others during a retreat at Crassus’ country house at Tusculum. Although the suggestion that a recollected (and no doubt highly embroidered) conversation can somehow claim the advantages of real-time, distributed thinking may seem an artificial or disingenuous gesture on Cicero’s part, it is less artificial than our own assumptions about the solitary nature of authorship would lead us to believe. As recent scholarship indicates, there is no clear reason to think that the treatises of Cicero were in fact composed by a single individual operating as isolated author (Gurd 2012). Apart from the distributed nature of elite Roman personhood, which often enfolds the thoughts and actions of slaves or other unspecified assistants into those of the master, Cicero goes so far as to argue that writing without openness to revision by others is the act of a tyrant and leaves in his ­­letters numerous indications of his own preference for collaborative authorship.18 Indeed, the characters in De oratore seem to articulate Cicero’s own anxieties about solitary or undistributed thought and to share his preference for collaborative intellectual effort. No sooner has Crassus opened the discussion by proclaiming the magnificence of the solitary speaker (unius, Cic. De or. 1.31) who transforms the human capacity for language into something wondrous, his unique ability ‘to bring help to suppliants, to raise up the afflicted, to bestow security’ and so forth (Cic. De or. 1.32), than Scaevola challenges his declaration on a number of fronts: was it eloquence, or perhaps wisdom, that allowed great individuals like Romulus, Numa and Servius Tullius to establish the Roman state? Indeed, was it individuals at all, or the knowledge condensed in ancient laws and customs (leges veteres moresque maiorum, Cic. De or. 1.39), or perhaps the practices and ceremonies of religion (religiones et caerimoniae, Cic. De or. 1.39), or civil procedure (Cic. De or. 1.39)?19 Is it even possible to imagine that a single orator would have the knowledge necessary to address a given subject ‘in complete fullness’ (copiosissime, Cic. De or. 1.41)? Could he, for example, be expected to know all of philosophy or mathematics? 18

On distributed authorship, see brief remarks of Habinek 2005a; see also his book-in-progress tentatively titled Personhood and Authorship in Ancient Rome. On solitary writing as a mark of the tyrant, see Gurd 2007 and 2012: 49–78. 19 Scaevola thus anticipates by two millennia Gallagher’s acute discussion of the legal ‘system’ as a potential technology of cognition: see Gallagher 2013.



d ist rib uted c ognit ion and i t s di s co nt ent s 235

Crassus refuses to back down, but in his response to Scaevola yields enough ground to move the discussion forward and refine the argument. He acknowledges the distinguished intellectual lineage behind Scaevola’s position, but warns against defining eloquence so narrowly that it only consists of good verbal style or should only be acquired by public speakers. Philosophers, he points out, are more persuasive when eloquent, but orators are raving madmen if they lack knowledge of their subject matter. Without repudiating his prior insistence on the importance of the solitary orator, Crassus in effect acknowledges that the knowledge he employs can never be his alone. He even imagines the orator consulting specific, named individuals about topics such as warfare, law or philosophy in preparing his case (Cic. De or. 1.66). What is more, his substantive claim is reinforced by Cicero’s literary representation of his thoughts and arguments as being altered under the pressure of engagement with Scaevola. The sequence of statement, supplementation and revision by multiple speakers continues throughout the first two books of the treatise. Antonius, the second most important character in the dialogue, at times takes direct issue with Crassus, at times issues mild corrections or revisions, and frequently brings new material to bear on the discussion. Despite their sometimes polemical nature, his interventions always manage to expand the range of topics covered by the treatise as a whole. The same is true of lesser characters. Scaevola, after making some important points about the orator’s need for legal knowledge, disappears from the latter portions of the dialogue. Catulus and Caesar arrive on the second day of the conversation, at the beginning of the second book, but rather than slowing the conversation, their presence moves the collective discussion forward as they bring new information and new perspectives to bear on the central topics, Catulus with his special interest in the niceties of language, and Caesar with his reputation for and scholarly understanding of wit. And even the latter, after regaling the others with examples of various orators’ witty remarks, only manages to assemble his thoughts into teachable form (Cic. De or. 2.229) under questioning by Crassus and Antonius. Here, too, the form imposed on the discussion by Cicero as author corresponds to the collaborative nature of knowledge production engaged in by his interlocutors. The lengthy discussion of wit in book 2 of De oratore is generally regarded as a distinctively Ciceronian contribution to the subject matter of classical rhetoric, yet within the dialogue it is placed in the mouth of a minor character whose presence at the discussion is all but accidental and who has the opportunity to share his knowledge thanks to the curiosity and courtesy of the other participants.20 Despite its fluid and collaborative nature, the discussion at Tusculum recounted by Cicero via Cotta eventually converges on a lengthy monologue by Crassus that occupies most of book 3. Cicero implicitly justifies the change in format by explaining, at the opening of the final book, that the events he is describing took place shortly before Crassus’ untimely death. In a sense, the last day of the discussion is 20

Leeman et al. (1989: 172–4) suggest that the topic of humour is included in De oratore partly in defence of Cicero’s own controversial use of jesting in his speeches.

236

thoma s habin ek a nd h ect o r reyes

the last chance for all involved, whether at Tusculum or among Cicero’s readership, to have the benefit of Crassus’ experience and insight. Although the timing of a dialogue shortly before the death of a key speaker is a device that dates at least to Plato, it seems to have special force in a dialogue that includes men of several generations and that repeatedly addresses continuities and changes between the dramatic occasion of the conversation and Roman political culture of earlier and later eras. As in a number of Ciceronian treatises, memory, especially memory embedded in and transmitted through collective rituals, is an important feature of both form and content.21 So too the Academic scepticism often imputed to Cicero seems less a philosophical commitment than a cognitive style, one that he regards as especially appropriate for leading men in a republic. In De oratore, then, the culminating attention granted to Crassus is not a crown awarded to a victor in debate but a recognition of the intellectual loss incurred by his absence from later deliberations. Somewhat paradoxically, Cicero’s ‘spokespeople’, as they are often called, do less to determine the structure of conversation and knowledge production than Plato’s Socrates, despite the latter’s repeated insistence on knowing nothing. Although the emphasis in our discussion of De oratore thus far has been on the tension between the collaborative nature of cognition about eloquence and the seeming autonomy of the orator that at least some interlocutors expect such a process to produce, it is also worth briefly noting ways in which the content of the instruction presented in the treatise presupposes that thought consists of interaction between brain, body and environment. For example, book 2 of De oratore preserves an important account of the ancient system of recollection based on loci, or places, whereby the speaker associates arguments or words to be remembered with sharply outlined images placed in clearly defined localities of which he has prior experience (Cic. De or. 2.351–60).22 Not surprisingly, Antonius, who takes on the task of elucidating the system in order to save Crassus the burden of doing so (Cic. De or. 2.350), attributes it to several prior thinkers, namely Simonides, Charmadas of Athens and Metrodorus of Skepsis, and likens one aspect of it to the work of painters who modify the shapes of objects in order to indicate their position (Cic. De or. 2.358). For Antonius, the system is both externalised, in its use of a sequence of places as prompts or scaffolding for recollection, and cross-modal, in so far as it entails shifting aural input into visual form, and visual form into verbal output (Cic. De or. 2.357).23 As he explains, the system allows its users ‘to grasp by seeing’ what they ‘can scarcely embrace by thinking’ (‘ea quae cogitando complecti vix possemus intuendo quasi teneremus’, Cic. De or. 2.357). Externalised aspects of cognition also become apparent in the treatise’s d ­­ iscussion 21

See, for example, the opening paragraphs of Cicero, De amicitia. For discussion of this and other relevant passages, see Small 1997; Onians 1999: 162–216; Sutton 2010; Short (this volume). 23 On scaffolding, see e.g. Clark 1997; Tribble 2005; Malafouris 2013. As Susi (2011: 188) puts it, scaffolding ‘allows epistemic actions that offload cognitive effort’. Sterelny (2010) argues that such scaffolding is but a specific instance of a more general process of co-constitution between internal and external. Cf. Spivey 2007. 22



d ist rib uted c ognit ion and i t s di s co nt ent s 237

of the act of writing. Writing is described not as a means of preserving or transmitting thought, but as a tool for generating the ability to speak well.24 Crassus asserts that ‘the most brilliant thoughts and expressions necessarily emerge at the point of the stilus’ (Cic. De or. 1.151); practice in writing makes an extemporaneous speech sound polished (Cic. De or. 1.152); the use of a written script for part of a speech makes the rest of a discourse proceed in the same style, like a boat driven forward by momentum even after the crewmen leave off rowing (Cic. De or. 1.152–3).25 For Antonius, the alphabet supplies an analogy for the use of commonplaces (i.e. generalised sayings applicable to a wide variety of particular circumstances). Just as a literate person has no need to think about the letters he uses to produce a word, so for the experienced speaker the commonplace utterances necessary for explicating the case ‘instantly come to the fore’ (Cic. De or. 2.130). In employing the spatial memory system developed by Simonides and others, Antonius advises, we are to regard the loci or places as a waxed tablet, the clearly circumscribed images as letters impressed upon them (Cic. De or. 2.354, 2.360). The analogy with the waxed tablet, which would have been scraped clean to allow for reuse, makes clear that here as elsewhere in De oratore writing is treated less as archive or offline storage than as one component of a process linking internal and external. Apart from the use of memory-places, probably the best-known system of scaffolding provided by ancient rhetoric is that of status- or stasis-theory, a system in which the orator prepares the arguments in a case by asking a set of increasingly refined questions that allow him to determine how to relate the particulars of his case to more familiar issues and arguments.26 Antonius shows familiarity with an early version of the system, that of topics, when he explains how he determines the issue in doubt (causa ambigendi, Cic. De or. 2.104), the reality of the situation under discussion (res necne, Cic. De or. 2.106), the quality of the relevant action (qualis sit, Cic. De or. 2.106), and its definition (cum quo verbo, Cic. De or. 2.107), and again when he lists the sources of argument that flow either from within the facts of the case itself (as with quality, definition and division, Cic. De or. 2.163–4) or from extrinsic sources, such as the reliability of testimony or the content of documents (Cic. De or. 2.173).27 What is striking, however, is Antonius’ insistence that such scaffolding is only really useful for a speaker who has had extensive prior experience or who has 24

i.e. it falls under the rubric of what Clark and Chalmers (1998: 9) call ‘active externalism’. Cf. Roby, this volume. 25 On writing as a tool for improvisation in Roman rhetorical practice, see Gurd 2012: 25–47. See Short (this volume) on writing as an aid to numerical calculation in the Roman world. 26 On the relationship between Antonius’ discussion of rhetorical invention and the approaches of Aristotle and Hermagoras, see Leeman et al. 1989: 20–4. 27 More detailed applications of the system are discussed in Rhetorica ad Herennium 1.18–27. There the author explains how one asks of a case first whether it is an issue of fact, law or justice. If law, for example, then one further asks whether the issue is applicability? definition? ambiguity? conflict between laws? If applicability, is it a matter of direct applicability, or indirect through inference? and so on.

238

thoma s habin ek a nd h ect o r reyes

­ repared himself through imitation of other excellent speakers. Antonius is somep times depicted by scholars as an advocate of natural talent over and against Crassus’ insistence on formal training. In fact, he is more appropriately viewed as a proponent of non-propositional or inductive thought, in contrast to Crassus’ willingness to give deduction and abstract method a greater role in the education of the orator. Finally, both Antonius and Crassus continually draw analogies between oratory and other arts such as painting, sculpture and music. Their claims suggest both the enactive nature of cognition, grasping and shaping objects in the world, and the collective nature of thought that emerges from the activities of diverse agents working in seemingly distinct media. Crassus is especially interested in the role of concept formation in and through all the arts and in the profound compatibility of their diverse contributions to human understanding, while for Antonius art demonstrates how an experienced craftsman can resolve an issue without explicit cogitation, in the sense of manipulation of propositions. As he explains, once Polyclitus determined how to depict Hercules, he had no difficulty fashioning his lion’s skin or the water serpent he defeated, even though he had no training in depicting such figures (Cic. De or. 2.70; cf. 2.73 on Phidias and Minerva). Even as he demonstrates familiarity with the rationalised methods of Greek teachers, who subdivide every case into discrete and minute problems to be resolved through explicit – that is, verbalised – step-by-step cogitation, Antonius himself argues that the analogue practice of imitation of a well-chosen exemplary speaker is more likely to produce outstanding results.28 In terms of models of distributed cognition, we might say that Antonius promotes a type of embodiment in which thinking oneself into another’s way of thinking29 allows a speaker to generate utterances he would not otherwise have produced, whereas Crassus assigns greater value to an explicit system of doctrine and method. In either case the inevitably distributed nature of human cognition is taken for granted, even as the ideal of an autonomous, fully comprehending, solo orator is held forth as incentive for achievement. Cicero’s De oratore thus contains within it the divergent approaches to cognition that characterise The School of Athens, Numa and Egeria and the Protagoras of Plato. Together, the four works remind us of the complexity of models of cognition at any given point in history, even when not explicitly articulated as such, and the diversity of disciplinary skills required to make them accessible to contemporary readers. All four works concern themselves with the nature of cognition, whether internal, external, singular, distributed, propositional or enactive, both in the content of their representations and in their formal strategies of communication. Conversely, typologies of cognition articulated as such by contemporary theorists provide a useful lens through which to view traditional works anew; or, perhaps 28

Notice also his preparation for a legal appearance: he and his client take turns arguing the two sides of the case, then he alone plays the parts of opposition counsel, judge and himself (Cic. De or. 2.102). 29 Or, to follow Sutton’s phrasing (2010: 206), Antonius regards other people’s ‘dynamic engrams [as] potentially our exograms’.



d ist rib uted c ognit ion and i t s di s co nt ent s 239

better, a means for engaging with them in order jointly to generate new knowledge. The history necessary to tell the story of distributed cognition can only be constructed through open-ended interaction with the texts, artefacts and agents who participate in it. No one text, artefact or temporal moment can teach us as much as the network of relations among them.

240

n otes on cont ri b ut o rs

N o t e s o n C o ntrib u tors 

Miranda Anderson is a research fellow at the University of Edinburgh. Her research focuses on cognitive approaches to literature and culture. She is the author of The Renaissance Extended Mind (2015) and she has recently published articles in Narrative and Poetics Today. Felix Budelmann teaches Classics at the University of Oxford. He specialises in Greek literature and has a particular interest in cognitive approaches to literature. He is the author of The Language of Sophocles (2000) and Greek Lyric: A Selection (2018). Douglas Cairns is Professor of Classics in the University of Edinburgh. His most recent books are Sophocles: Antigone (2016), Emotions in the Classical World (with Damien Nelis, 2017), Greek Laughter and Tears (with Margaret Alexiou, 2017) and Seneca’s Tragic Passions (with Damien Nelis, 2017). Christopher Gill has held academic posts at Yale, Bristol, Aberystwyth and Exeter. His research is centred on ancient philosophy, especially psychology and ethics, including concepts of personality and self, and on the relationship between ancient and modern thought in these areas. His books include Personality in Greek Epic, Tragedy, and Philosophy (1996), The Structured Self in Hellenistic and Roman Thought (2006) and Naturalistic Psychology in Galen and Stoicism (2010). Thomas Habinek is Professor of Classics and Director of the Society of Fellows in the Humanities at the University of Southern California, with primary interests in ancient Greek and Roman intellectual, literary and cultural history. Luuk Huitink is currently employed as a research fellow at Leiden University. He has previously been a  postdoctoral researcher on the ERC Project ‘Ancient Narrative’ at Heidelberg University, Leventis Research Fellow in Ancient Greek at Merton College, Oxford, and Spinoza Visiting Fellow at Leiden University.



n otes on contri b ut o rs 241

George Kazantzidis is Assistant Professor of Latin Literature at the University of Patras, Greece. His interests lie in the history of mental illness and the history of emotions. He is currently working on a book project provisionally entitled ‘Ancient Paradoxography: Medicine, Horror and the Sublime’. David Konstan is Professor of Classics at New York University. Among his publications are Greek Comedy and Ideology; Friendship in the Classical World; Pity Transformed; The Emotions of the Ancient Greeks; Before Forgiveness: The Origins of a Moral Idea; and Beauty: The Fortunes of an Ancient Greek Idea. He is a past president of the American Philological Association (now Society for Classical Studies), a fellow of the American Academy of Arts and Sciences, and honorary fellow of the Australian Academy of the Humanities. Peter Meineck is Professor of Classics in the Modern World at New York University and the founder of Aquila Theatre. Recent publications include Theatrocracy: Greek Drama, Cognition and the Imperative for Theatre (2017) and Combat Trauma and the Ancient Greeks (with David Konstan, 2014). Diana Y. Ng is Associate Professor of Art History at the University of MichiganDearborn. She is co-editor of Reuse and Renovation in Roman Material Culture: Functions, Aesthetics, Interpretations (forthcoming from Cambridge University Press). She has published essays on Roman sculpture in Asia Minor, public architecture and sculpture as expressions of civic identity, elite benefaction and commemoration, and processes of learning through spectacle and the display of images in the Roman theatre at Ephesus. Hector Reyes is Assistant Professor (Teaching) at the University of Southern California. He specialises in early modern French painting understood in relationship to broader currents in European intellectual life. Andrew M. Riggsby is Lucy Shoe Meritt Professor in Classics at the University of Texas at Austin. In addition to work on Roman politics, law and culture, his book Mosaics of Knowledge: Representing Information in the Roman World is forthcoming from Oxford University Press. Courtney Roby is Associate Professor of Classics at Cornell University. Her first book (Technical Ekphrasis in Ancient Science and Literature: The Written Machine between Alexandria and Rome, 2016) traces the techniques used to represent technological artefacts in Greco-Roman literature. Her previous work on cognitive linguistics and cognitive science in Greek and Roman technical literature includes projects on embodied cognition in Hero of Alexandria’s Dioptra and embodiment and semantics in the Roman land-surveyors. William Michael Short is Lecturer in Classics at the University of Exeter. Prior to this, he was Associate Professor at the University of Texas at San Antonio. He

242

n otes on cont ri b ut o rs

specialises in Latin language and literature, Roman cultural history, cognitive semantics and cultural anthropology. His research interests rest at the intersection of language, culture and cognition, and in this area he has been pioneering an approach to ancient Roman culture inspired by Lakoffian conceptual metaphor theory. Mark Sprevak is Senior Lecturer in Philosophy at the University of Edinburgh. His primary research interests are in philosophy of mind, philosophy of science and metaphysics, with particular focus on the cognitive sciences. He has published articles in, among other places, The Journal of Philosophy, The British Journal for the Philosophy of Science, Synthese, Philosophy, Psychiatry & Psychology, and Studies in History and Philosophy of Science. His book The Computational Mind is forthcoming from Routledge. Michael Wheeler is Professor of Philosophy at the University of Stirling. His primary research interests are in philosophy of science (especially cognitive science, psychology, biology and artificial intelligence) and philosophy of mind. He also works on Heidegger and on developing ideas at the interface between the analytic and the continental philosophical traditions. His book Reconstructing the Cognitive World: The Next Step was published by MIT Press in 2005.



bibliogra p h y 243

B i bl i o g r ap hy 

Adams, C. D. (ed.) (1919), Aeschines: Speeches, Cambridge, MA: Harvard University Press. Adams, F. and K. Aizawa (2008), The Bounds of Cognition, Malden, MA, and Oxford: Blackwell. Alföldi-Rosenbaum, E. (1971), ‘The finger calculus in antiquity and in the Middle Ages’, Frühmittelalterliche Studien 5: 1–9. Alibali, M. W., R. Boncoddo and A. B. Hostetter (2014), ‘Gesture in reasoning: An embodied perspective’, in L. Shapiro (ed.), The Routledge Handbook of Embodied Cognition, London: Routledge, pp. 150–9. Allan, R. J. (2013), ‘History as presence: Time, tense and narrative modes in Thucydides’, in A. Tsakmakis and M. Tamiolaki (eds), Thucydides between History and Literature, Berlin: De Gruyter, pp. 371–89. Allington, D. (2011), ‘ “It actually painted a picture of the village and the sea and the bottom of the sea”: Reading groups, cultural legitimacy, and description in narrative (with reference to John Steinbeck’s The Pearl)’, Language and Literature 20: 317–32. Alvarez, G. and P. Cavanagh (2004), ‘The capacity of visual short-term memory is set both by visual information load and by number of objects’, Psychological Science 15.2: 106–11. Alverson, H. (1994), Semantics and Experience: Universal Metaphors of Time in English, Mandarin, Hindi, and Sesotho, Baltimore: Johns Hopkins University Press. Anderson, M. (2015), The Renaissance Extended Mind, Basingstoke and New York: Palgrave Macmillan. Annas, J. (1992), Hellenistic Philosophy of Mind, Berkeley and Los Angeles: University of California Press. Antović, M. and C. Pagán Cánovas (eds) (2016), Oral Poetics and Cognitive Science, Berlin: De Gruyter. Appadurai, A. (2011), The Social Life of Things, Cambridge: Cambridge University Press. Armstrong, C. (2009), Figures of Memory: Poetry, Space, and the Past, Basingstoke: Palgrave Macmillan.

244

bibliograp h y

Arnau, E., A. Estany, R. González del Solar and T. Sturm (2014), ‘The extended cognition thesis: Its significance for the philosophy of (cognitive) science’, Philosophical Psychology 27.1: 1–18. Arnold, M. (1960), Emotion and Personality, London: Cassell and Co. Arroyo, S., R. Lesser, B. Gordon, S. Uematsu, J. Hart, P. Schwerdt, K. Andreasson and R. Fisher (1993), ‘Mirth, laughter and gelastic seizures’, Brain 166: 757–880. Assmann, J. (1992), Das kulturelle Gedächtnis: Schrift, Erinnerung und politische Identität im frühen Hochkulturen, Munich: Beck. Assmann, J. (2006), Religion and Cultural Memory, Palo Alto: Stanford University Press. Atkinson, R. and R. Shiffrin (1968), ‘Human memory: A proposed system and its control processes’, in K. Spence and J. Spence (eds), The Psychology of Learning and Motivation, vol. 2, New York: Academic Press, pp. 89–195. Aujac, G. (1993), La Sphère: Instrument au service de la découverte du mond: d’Autolycos de Pitanè à Jean de Sacrobosco, Caen: Paradigme. Baddeley, A. (1998), Human Memory: Theory and Practice, Boston: Allyn & Bacon. Baddeley, A. and G. Hitch (1974), ‘Working memory’, Psychology of Learning and Motivation 8: 47–89. Baddeley, A., V. Lewis, M. Eldridge and N. Thomson (1984), ‘Attention and retrieval from long-term memory’, Journal of Experimental Psychology: General 113: 518–40. Bagnall, R. and P. Derow (1981), Greek Historical Documents: The Hellenistic Period, Chico, CA: Society of Biblical Literature. Bahrick, H. (1983), ‘The cognitive map of a city: Fifty years of learning and memory’, Psychology of Learning and Motivation 17: 125−63. Bahrick, H. (1984), ‘Semantic memory content in permastore’, Journal of Experimental Psychology: General 113: 1−29. Baker, P. (2015), Italian Renaissance Humanism in the Mirror, Cambridge: Cambridge University Press. Barker, A. (1989), Greek Musical Writings, vol. 2, Cambridge and New York: Cambridge University Press. Barker, A. (2006a), ‘Ptolemy’s musical models for mind-maps and star-maps’, in C. Cheyne and J. Worrall (eds), Rationality and Reality, Dordrecht: Springer Netherlands, pp. 273–91. Barker, A. (2006b), Scientific Method in Ptolemy’s Harmonics, Cambridge: Cambridge University Press. Barnes, J. (1971–2), ‘Aristotle’s concept of mind’, Proceedings of the Aristotelian Society 72: 101–14. Barnes, J. (ed.) (1984), The Complete Works of Aristotle: The Revised Oxford Translation, Princeton: Princeton University Press. Barrett, L. F. and M. Bar (2009), ‘See it with feeling: Affective predictions during object perception’, Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences 364.1521: 1325–34.



bibliogra p h y 245

Barrett, N. F. (2017) ‘The normative turn in enactive theory: an examination of its roots and implications’, Topoi 36.3: 431–43. Barthes, R. (1985), The Responsibility of Forms, trans. R. Howard, Berkeley: University of California Press. Bartos, H. (2015), Philosophy and Dietetics in the Hippocratic On Regimen: A Delicate Balance of Health, Leiden and Boston: Brill. Battezzato, L. (2009), ‘Techniques of reading and textual layout in ancient Greek texts’, Cambridge Classical Journal 55: 1–23. Battig, W. and F. Bellezza (1979), ‘Organisation and levels of processing’, in C. Puff (ed.), Memory Organisation and Structure, New York: Academic Press, pp. 321–46. Baylor, A. L. (2009), ‘Promoting motivation with virtual agents and avatars: Role of visual presence and appearance’, Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences 364.1535: 3559–65. Beach, K. (1988), ‘The role of external mnemonic symbols in acquiring an occupation’, in M. M. Gruneberg and R. N. Sykes (eds), Practical Aspects of Memory, vol. 1, New York: Wiley, pp. 342–6. Beacham, R. (1992), The Roman Theater and its Audience, Cambridge, MA: Harvard University Press. Bean, G. (1980), Turkey Beyond the Meander, New York: Norton. Beard, M., J. North and S. Price (1998), Religions of Rome, 2 vols, Cambridge: Cambridge University Press. Beare, J. I. (trans.) (1931), Aristotle: On Dreams, in The Works of Aristotle, vol. 3, ed. W. D. Ross, Oxford: Oxford University Press. Bejjani, B.-P.,  P. Damier, I. Arnulf, L. Thivard,  A.-M. Bonnet,  D. Dormont, P. Cornu, B. Pidoux, Y. Samson and Y. Agid (1999), ‘Transient acute depression induced by high-frequency deep-brain stimulation’, New England Journal of Medicine 340: 1476–80. Bejor, G. (1991), Hierapolis III: Le statue, Rome: Giorgio Bretschneider. Bennetto, L., B. Pennington and S. Rogers (1996), ‘Intact and impaired memory functions in autism’, Child Development 67.4: 1816–35. Benson, P. J. and D. I. Perrett (1994), ‘Visual processing of facial distinctiveness’, Perception 23: 75–93. Berridge, K. C. (2003), ‘Comparing the emotional brain of humans and other animals’, in R. J. Davidson, K. R. Scherer and H. H. Goldsmith (eds), Handbook of Affective Sciences, Oxford: Oxford University Press, pp. 25–51. Berridge, K. C. and P. Winkielman (2003), ‘What is an unconscious emotion? The case for unconscious “liking”’, Cognition and Emotion 17: 181–211. Berry, J. and P. Dasen (eds) (1973), Culture and Cognition, London: Methuen. Bers, V. (1985), ‘Dikastic thorubos’, in P. Cartledge and F. D. Harvey (eds), Crux: Essays Presented to G.E.M. de Ste. Croix, London: Duckworth, pp. 1–15. Berti, A. and F. Frassinetti (2000), ‘When far becomes near: Re-mapping of space by tool use’, Journal of Cognitive Neuroscience 12: 415–20. Bettini, M. (2011a), The Ears of Hermes, Columbus: Ohio State University Press. Bettini, M. (2011b), ‘Miti di memoria’, in J.-P. Brun, A. Schnapp and M.

246

bibliograp h y

­Segonds-Bauer (eds), L’histoire comme impératif, ou, La ‘volonté de comprendre’, Naples: D’Auria, pp. 129−44. Bieber, M. (1961), The History of the Greek and Roman Theater, Princeton: Princeton University Press. Billing, C. M. (2008), ‘Representations of Greek tragedy in ancient pottery: A theatrical perspective’, New Theatre Quarterly 24.3: 229–45. Billings, J., F. Budelmann and F. Macintosh (eds) (2013), Choruses, Ancient and Modern, Oxford: Oxford University Press. Bingen, J., A. Bülow-Jacobsen, W. Cockle, H. Cuvigny, F. Kayser and W. Van Rengen (1997), Mons Claudianus: Ostraca Graeca et Latina II, Cairo: Institut français d’archéologie orientale du Caire. Bleckmann, B. (1998), Athens Weg in die Niederlage: Die letzten Jahre des Peloponnesischen Kriegs, Stuttgart and Leipzig: Teubner. Blunt, A. (ed.) (1964), Nicolas Poussin: Lettres et propos sur l’art, Paris: Hermann. Boden, M. (2008), Mind as Machine: A History of Cognitive Science, Oxford: Oxford University Press. Bolens, G. (2012), The Style of Gestures: Embodiment and Cognition in Literary Narrative, Baltimore: Johns Hopkins University Press. Bolens, G. (2017), ‘Mind in movement’, History of Distributed Cognition Project, (last accessed 10 January 2018). Bolzoni, L. (2001), The Gallery of Memory, Toronto: University of Toronto Press. Bommas, M. (ed.) (2011), Cultural Memory and Identity in Antiquity (Cultural Memory and History in Antiquity), London and New York: Continuum. Boroditsky, L. (2000), ‘Metaphoric structuring: Understanding time through spatial metaphors’, Cognition 75: 1–28. Bowman, A. K. and G. Woolf (eds) (1994), Literacy and Power in the Ancient World, Cambridge: Cambridge University Press. Bradley, M. M. and P. J. Lang (2000), ‘Measuring emotion: Behavior, feeling, and physiology’, in R. D. Lane and L. Nadel (eds), Cognitive Neuroscience of Emotion, Oxford: Oxford University Press, pp. 242–76. Bremmer, J. N. (1994), Greek Religion, Oxford: Oxford University Press. Bremmer, J. and N. Horsfall (1987), Roman Myth and Mythography, London: University of London Press. Brennan, S. E. (2007), ‘Caricature generator: The dynamic exaggeration of faces by computer’, Leonardo 40.4: 392–400. Brennan, T. (2003), ‘Stoic moral psychology’, in B. Inwood (ed.), The Cambridge Companion to the Stoics, Cambridge: Cambridge University Press, pp. 257–94. Broadbent, D. (1975), ‘The magic number seven after fifteen years’, in A. Kennedy and A. Wilkes (eds), Studies in Long-Term Memory, New York: Wiley, pp. 3–18. Brodersen, K. (1995), Terra cognita: Studien zur römischen Raumerfassung, Spudasmata 59, Hildesheim and New York: Olms. Brownson, C. L. (ed.) (1921), Xenophon, vol. 2, Cambridge, MA: Harvard University Press.



bibliogra p h y 247

Budelmann, F. and P. Easterling (2010), ‘Reading minds in Greek tragedy’, Greece & Rome 57: 289–303. Budelmann, F. and P. LeVen (2014), ‘Timotheus’ poetics of blending: A cognitive approach to the language of the New Music’, Classical Philology 109: 191–210. Burgess, S. (2000), ‘How to build a human body: An idealist’s guide’, in M. R. Wright (ed.), Reason and Necessity: Essays on Plato’s Timaeus, London: Duckworth and Classical Press of Wales, pp. 43–58. Burnyeat, M. (1992), ‘Is an Aristotelian philosophy of the mind still credible? A draft’, in M. C. Nussbaum and A. O. Rorty (eds), Essays on Aristotle’s De Anima, Oxford: Oxford University Press, pp. 15–26. Burnyeat, M. (2000), ‘Plato on why mathematics is good for the soul’, in T. Smiley (ed.), Mathematics and Necessity, Proceedings of the British Academy 103: 1–81. Burton, L., L. Rabin, S. Vardy, J. Frohlich, G. Wyatt, D. Dimitri, S. Constante and E. Guterman (2004), ‘Gender differences in implicit and explicit memory for affective passages’, Brain and Cognition 54.3: 218–24. Bussels, S. (2012), The Animated Image: Roman Theory on Naturalism, Vividness and Divine Power, Berlin: Akademie-Verlag. Butler, J. (1998), ‘Imitation and gender insubordination’, in Julie Rivkin and Michael Ryan (eds), Literary Theory: An Anthology, Oxford: Blackwell Publishers, pp. 722–30. Cabrera, M., K. Novak, D. Foti, R. Voyles and J. Wachs (2017), ‘What makes a gesture a gesture? Neural signatures involved in gesture recognition’, arXiv.org, (last accessed 6 February 2018). Cacioppo, J. et al. (1993), ‘The psychophysiology of emotion’, in M. Lewis and J. Haviland (eds), Handbook of Emotions, New York: Guilford Press, pp. 119–42. Cairns, D. L. (2005), ‘Bullish looks and sidelong glances: Social interaction and the eyes in ancient Greek culture’, in D. Cairns (ed.), Body Language in the Greek and Roman Worlds, Swansea: Classical Press of Wales, pp. 123–55. Cairns, D. L. (2013), ‘A short history of shudders’, in A. Chaniotis and P. Ducrey (eds), Unveiling Emotions II: Emotions in Greece and Rome: Texts, Images, and Material Culture, Stuttgart: Steiner, pp. 85–107. Cairns, D. L. (2014), ‘Psyche, thymos, and metaphor in Homer and Plato’, Les Études Platoniciennes 11: 1–37. Cairns, D. L. (2015), ‘The horror and the pity: Phrike as a tragic emotion’, Psychoanalytic Inquiry 34: 75–94. Cairns, D. L. (2016a), ‘Clothed in shamelessness, shrouded in grief: The role of “garment” metaphors in ancient Greek concepts of emotion’, in G. Fanfani, M. Harlow and M.-L. Nosch (eds), Spinning Fates and the Song of the Loom: The Use of Textiles, Clothing, and Cloth Production as Metaphor, Symbol, and Narrative, Oxford: Oxbow, pp. 25–41. Cairns, D. L. (2016b), ‘Metaphors for hope in early Greek literature’, in R. R. Caston and R. A. Kaster (eds), Hope, Joy, and Affection in the Classical World, New York: Oxford University Press, pp. 13–44. Cairns, D. L. (2017a), ‘Horror, pity, and the visual in ancient Greek aesthetics’,

248

bibliograp h y

in D. L. Cairns and D. P. Nelis (eds), Emotions in the Classical World: Methods, Approaches and Directions, Stuttgart: Steiner, pp. 53–77. Cairns, D. L. (2017b), ‘Mind, metaphor, and emotion in Euripides (Hippolytus) and Seneca (Phaedra)’, in D. L. Cairns and D. P. Nelis (eds), Seneca’s Tragic Passions, special edition of Maia 69.2: 247–67. Cairns, D. L. (2017c), ‘The tripartite soul as metaphor’, in P. Destrée and R. G. Edmonds (eds), Plato and the Power of Images, Leiden: Brill, pp. 219–38. Cairns, D. L. (forthcoming), ‘Phaedra’s fantasy other: Phenomenology and the enactive mind in Euripides’ Hippolytus’, in A. Gartziou-Tatti, M. Liatsi and E. Karakasis (eds), Ethics in Ancient Greek Literature: From Homer to Epicurus, Berlin: De Gruyter. Cairns, D. L. and D. P. Nelis (2017), ‘Introduction’, in D. L. Cairns and D. P. Nelis (eds), Emotions in the Classical World: Methods, Approaches and Directions, Stuttgart: Steiner, pp. 7–30. Calame, C. (1986), ‘Facing otherness: The tragic mask in ancient Greece’, History of Religions 26.2: 125–42. Calame, C. (1991), ‘Quand dire c’est faire voir: l’évidence dans la rhétorique antique’, Etudes de lettres 4: 3–22. Calame, C. (2009), Poetic and Performative Memory in Ancient Greece, Washington, DC: Center for Hellenic Studies. Calboli Montefusco, L. (2005), ‘Ἐνάργεια et ἐνέργεια: L’évidence d’une démonstration qui signifie les choses en acte’, in M. Armisen-Marchetti (ed.), Demonstrare: Voir et faire voir: forme de la démonstratio à Rome, Toulouse: Presses universitaires du Mirail, pp. 43–58. Cameron, A. (2004), Greek Mythography in the Roman World, Oxford: Oxford University Press. Cancik, H., H. Schneider, C. Salazar and D. Orton (eds) (2002), Brill’s New Pauly: Encyclopaedia of the Ancient World: Antiquity (English edn), Leiden and Boston: Brill. Canevaro, M. (2018), ‘Majority rule vs consensus: The practice of democratic deliberation in the Greek poleis’, in M. Canevaro, A. Erskine, B. Gray and J. Ober (eds), Ancient Greek History and Contemporary Social Science, Edinburgh: Edinburgh University Press, pp. 101–56. Caracciolo, M. (2014), The Experientiality of Narrative: An Enactivist Approach, Berlin and Boston: De Gruyter. Carawan, E. (2008), ‘What the mnemones know’, in E. A. Mackay (ed.), Orality, Literacy, Memory in the Ancient Greek and Roman World, Leiden: Brill, pp. 163–84. Carroll, J. (1994), Evolution and Literary Theory, Columbia: University of Missouri. Carruthers, M. (1990), The Book of Memory, Cambridge: Cambridge University Press. Carter, D. M. (2010), ‘The demos in Greek tragedy’, Cambridge Classical Journal 56: 47–94. Carter, J. B. (1987), ‘The masks of Ortheia’, American Journal of Archaeology 91: 355–83. Cataldi, S. (1996), Emotion, Depth and Flesh, Albany: State University of New York Press.



bibliogra p h y 249

Cave, T. (2016), Thinking with Literature: Towards a Cognitive Criticism, Oxford: Oxford University Press. Cermak, L. and F. Craik (1979), Levels of Processing in Human Memory, Hillsdale, NJ: Erlbaum. Chaniotis, A. (2007), ‘Theater rituals’, in P. Wilson (ed.), The Greek Theatre and Festivals, New York: Oxford University Press, pp. 48–66. Chaniotis, A. (2011a), ‘Emotional community through ritual: Initiates, citizens, and pilgrims as emotional communities in the Greek world’, in A. Chaniotis (ed.), Ritual Dynamics in the Ancient Mediterranean: Agency, Emotion, Gender, Representation, Stuttgart: Steiner, pp. 264–90. Chaniotis, A. (2011b), Ritual Dynamics in the Ancient Mediterranean, Stuttgart: Steiner. Chaniotis, A. (2012), ‘Introduction’, in A. Chaniotis (ed.), Unveiling Emotions: Sources and Methods for the Study of Emotions in the Greek World, Stuttgart: Steiner, pp. 11–36. Chemero, A. (2009), Radical Embodied Cognitive Science, Cambridge, MA: MIT Press. Chrisomalis, S. (2009), ‘The cognitive and cultural foundations of numbers’, in E. Robson and J. Stedall (eds), The Oxford Handbook of the History of Mathematics, Oxford: Oxford University Press, pp. 495–517. Clark, A. (1997), Being There: Putting Brain, Body, and World Together Again, Cambridge, MA: MIT Press. Clark, A. (2003), Natural-Born Cyborgs: Minds, Technologies, and the Future of Human Intelligence, Oxford: Oxford University Press. Clark, A. (2008a), ‘Pressing the flesh: A tension in the study of the embodied, embedded mind?’, Philosophy and Phenomenological Research, 76.1: 37–59. Clark, A. (2008b), Supersizing the Mind: Embodiment, Action, and Cognitive Extension, Oxford: Oxford University Press. Clark, A. (2009), ‘Spreading the joy? Why the machinery of consciousness is ­(probably) still in the head’, Mind 118.472: 963–93. Clark, A. (2010), ‘Material surrogacy and the supernatural: Reflections on the role of artefacts in “off-line” cognition’, in L. Malafouris and C. Renfrew (eds), The Cognitive Life of Things: Recasting the Boundaries of the Mind, Cambridge: McDonald Institute Monographs, pp. 23–8. Clark, A. (2013), ‘Gesture as thought?’, in Z. Radman (ed.), The Hand, an Organ of the Mind: What the Manual Tells the Mental, Cambridge, MA: MIT Press, pp. 255–68. Clark, A. (2014), ‘Extended mind’, History of Distributed Cognition Project, (last accessed 6 February 2018). Clark, A. and D. Chalmers (1998), ‘The extended mind’, Analysis 58.1: 7–19. Coldiron, M. (2004), Trance and Transformation of the Actor in Japanese Noh and Balinese Masked Dance-Drama (No. 30), Lewiston, NY: Mellen. Cole, M. and S. Scribner (1974), Culture and Thought, New York: Wiley. Collini, S. (2009), ‘Impact on humanities: Researchers must take a stand now or be judged and rewarded as salesmen’, Times Literary Supplement, 13 November 2009, pp. 18–19.

250

bibliograp h y

Colombetti, G. (2014), The Feeling Body: Affective Science Meets the Enactive Mind, Cambridge, MA: MIT Press. Colombetti, G. (2015), ‘Enactive affectivity, extended’, Topoi 54: 1–11. Colombetti, G. and T. Roberts (2015), ‘Extending the extended mind: The case for extended affectivity’, Philosophical Studies 172.5: 1243–63. Colombetti, G. and E. Thompson (2008), ‘The feeling body: Towards an enactive approach to emotion’, in W. F. Overton, U. Müller and J. Newman (eds), Developmental Perspectives on Embodiment and Consciousness, New York: Erlbaum, pp. 45–68. Connerton, P. (1989), How Societies Remember, Cambridge: Cambridge University Press. Conway, A., C. Jarrold, M. Kane, A. Miyake and J. Towse (eds) (2007), Variation in Working Memory, Oxford: Oxford University Press. Conway, M., G. Cohen and N. Stanhope (1992), ‘Very long-term memory’, Applied Cognitive Psychology 6.6: 467–82. Cooper, J. M. (1996), ‘An Aristotelian theory of emotion’, in A. Rorty (ed.), Essays on Aristotle’s Rhetoric, Berkeley: University of California Press, pp. 238–57. Cope, E. M. (1877), The Rhetoric of Aristotle, with a Commentary, vol. 2, rev. and ed. J. E. Sandys, Cambridge: Cambridge University Press. Corbeill, A. (2004), Nature Embodied: Gesture in Ancient Rome, Princeton: Princeton University Press. Courtney, E. (1995), Musa Lapidaria: A Selection of Latin Verse Inscriptions, Atlanta: Scholars Press. Cowan, N. (1988), ‘Evolving conceptions of memory storage, selective attention, and their mutual constraints within the human information processing system’, Psychological Bulletin 104: 163–91. Cowan, N. (2001), ‘The magical number 4 in short-term memory: A reconsideration of mental storage capacity’, Behavioral and Brain Sciences 24.1: 87–114. Craik, F. and R. Lockhart (1972), ‘Levels of processing: A framework for memory research’, Journal of Verbal Learning and Verbal Behaviour 11: 671–84. Creese, D. E. (2010), The Monochord in Ancient Greek Harmonic Science, Cambridge: Cambridge University Press. Crego, C. (trans.) (1999), The Poetry of Rainer Maria Rilke, available online at (last accessed 6 February 2018). Crosby, A. (1997), The Measure of Reality: Quantification and Western Society, Cambridge: Cambridge University Press. Csapo, E. (1997), ‘Riding the phallus for Dionysus: Iconology, ritual, and gender-role de/construction’, Phoenix 51.3–4: 253–95. Csapo, E. (2010), Actors and Icons of the Ancient Theater, Chichester: Wiley-Blackwell. Csapo, E., H. R. Goette, J. R. Green and P. Wilson (eds) (2014), Greek Theatre in the Fourth Century bc, Berlin: De Gruyter. Csapo, E. and W. J. Slater (1995), The Context of Ancient Drama, Ann Arbor: University of Michigan Press.



bibliogra p h y 251

Cui, X., C. B. Jeter, D. Yang, P. Read Montague and D. M. Eagleman (2007), ‘Vividness of mental imagery: Individual variability can be measured objectively’, Vision Research 47: 474–8. Cuomo, S. (2001), Ancient Mathematics, London: Routledge. D’Andria, F. (2003), Hierapolis of Phrygia (Pammukale), trans. P. Arthur, Istanbul: Ege Yayinlari. D’Andria, F. and T. Ritti (1985), Hierapolis II: Le Sculture del Teatro, Rome: Giorgio Bretschneider. Damasio, A. (1994), Descartes’ Error: Emotion, Reason and the Human Brain, New York: Avon. Damasio, A. (1999), The Feeling of What Happens: Body, Emotion and the Making of Consciousness, New York: Harcourt Brace and Company. Damasio, A. (2003), Looking for Spinoza: Joy, Sorrow, and the Feeling Brain, London: Heinemann. Damasio, A. (2010), Self Comes to Mind: Constructing the Conscious Brain, London: Heinemann. Damasio, A. and G. B. Carvalho (2013), ‘The nature of feelings: Evolutionary and neurobiological origins’, Nature Reviews Neuroscience 14.2: 143–52. Danesi, M. and P. Perron (1999), Analyzing Culture, Bloomington: Indiana University Press. De Bernardi Ferrero, D. (1966), Teatri classici in Asia Minore, vol. 1, Rome: ‘L’Erma’ di Bretschneider. De Bernardi Ferrero, D. (1970), Teatri classici in Asia Minore, vol. 3, Rome: ‘L’Erma’ di Bretschneider. De Chaisemartin, N. and D. Theodorescu (1991), ‘Recherches préliminaries sur la frons scaenae du théâtre’, in R. R. R. Smith and K. T. Erim (eds), Aphrodisias Papers 2, Journal of Roman Archaeology Supplement 2, Ann Arbor: University of Michigan Press, pp. 29–65. De Gelder, B. (2009), ‘Why bodies? Twelve reasons for including bodily expressions in affective neuroscience’, Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences 364.1535: 3475–84. De Gelder, B., J. van den Stock, H. Meeren, C. Sinke, M. Kret and M. Tamietto (2010), ‘Standing up for the body: Recent progress in uncovering the networks involved in the perception of bodies and bodily expressions’, Neuroscience and Biobehavioral Reviews 34.4: 513–27. De Jaegher, H. (2009), ‘Social understanding through direct perception? Yes, by interacting’, Consciousness and Cognition 18.2: 535–42. De Jaegher, H. and E. Di Paolo (2007), ‘Participatory sense-making: An enactive approach to social cognition’, Phenomenology and the Cognitive Sciences 6.4: 485–507. De Lacy, P. (ed.) (1978), Galen: On the Doctrines of Hippocrates and Plato, Berlin: Akademie-Verlag. De Sanctis, G. (2014), ‘Spazio’, in M. Bettini and W. M. Short (eds), Con i Romani: Un’antropologia della cultura antica, Bologna: Il Mulino, pp. 143–66. De Sélincourt, A. (trans.) (2003), Herodotus: The Histories, Harmondsworth: Penguin.

252

bibliograp h y

Deckert, F., B. Seidler, M. Ebbecke and M. Gillmann (2011), ‘Table content understanding in SmartFIX’, in 2011 International Conference on Document Analysis and Recognition (ICDAR), pp. 488–92. Deignan, A. (2003), ‘Metaphoric expressions and culture’, Metaphor and Symbol 18: 255–71. DeLoache, J. S. (1991), ‘Symbolic functioning in very young children: Understanding of pictures and models’, Child Development 62.4: 736–52. Dempsey, C. (1966), ‘The classical perception of nature in Poussin’s earlier works’, Journal of the Warburg and Courtauld Institutes, 29: 219–49. Dennett, D. (1991), Consciousness Explained, Boston: Little, Brown & Company. Detlefsen, M., D. Erlandson, J. Heston and C. Young (1975), ‘Computation with Roman numerals’, Archive for History of Exact Sciences 15: 141–8. Di Paolo, E. (2005), ‘Autopoiesis, adaptivity, teleology, agency’, Phenomenology and the Cognitive Sciences 4.4: 429–52. Di Paolo, E. (2009), ‘Extended life’, Topoi 28: 9–21. Dilke, O. (1971), The Roman Land Surveyors: An Introduction to the Agrimensores, Newton Abbot: David and Charles. Dilke, O. A. W. (1985), Greek and Roman Maps, Ithaca: Cornell University Press. Donald, M. (1991), Origins of the Modern Mind: Three Stages in the Evolution of Culture and Cognition, Cambridge, MA: Harvard University Press. Donald, M. (2001), A Mind So Rare: The Evolution of Human Consciousness, New York: W. W. Norton. Donald, M. (2006), ‘Art and cognitive evolution’, in M. Turner (ed.), The Artful Mind, Oxford: Oxford University Press, pp. 3–20. Dow, J. (2015), Passions and Persuasion in Aristotle’s Rhetoric, Oxford: Oxford University Press. Dozier, C. (2013), ‘Blinded by the light: Oratorical clarity and poetic obscurity in Quintilian’, in S. Butler and A. Purves (eds), Synaesthesia and the Ancient Senses, Durham: Acumen, pp. 141–53. Dozois, D. J. A. and A. T. Beck (2011), ‘Cognitive therapy’, in J. D. Herbert and E. M. Forman (eds), Acceptance and Mindfulness in Cognitive Behaviour Therapy, Hoboken, NJ: Wiley, pp. 26–56. Draaisma, D. (2000), Metaphors of Memory: A History of Ideas About the Mind, Cambridge: Cambridge University Press. Drayson, Z. (2009), ‘Embodied cognitive science and its implications for psychopathology’, Philosophy, Psychiatry and Psychology 16.4: 329–40. Dunbar, N. (1995), Aristophanes’ Birds, Oxford: Oxford University Press. Dupont, F. (1992), Daily Life in Ancient Rome, Oxford: Oxford University Press. Düring, I. (1930), Die Harmonielehre des Klaudios Ptolemaios, Göteborgs Högskolas Arsskrift 36, Gothenburg: Elanders Boktryckeri Aktiebolag. Edghill, E. M. (trans.) (1928), Aristotle: Categories, in The Works of Aristotle, vol. 1, ed. W. D. Ross, Oxford: Oxford University Press. Edwards, C. (1996), Writing Rome, Cambridge: Cambridge University Press. Eichenbaum, H., T. Otto and N. Cohen (2010), ‘Two functional compo-



bibliogra p h y 253

nents of the hippocampal memory system’, Behavioral and Brain Sciences 17.3: 449–72. Eidinow, E. (2011), ‘Networks and narratives: A model for ancient Greek religions’, Kernos 24: 9–38. Eidinow, E. (2015a), ‘Ancient Greek religion: “Embedded” . . . and embodied’, in C. Taylor and K. Vlassopoulos (eds), Communities and Networks in the Ancient Greek World, Oxford: Oxford University Press, pp. 54–79. Eidinow, E. (2015b), ‘φανερὰν ποιήσει τὴν αὑτοῦ διάνοιαν τοῖς θεοῖς: Some ancient Greek theories of (divine and mortal) mind’, in C. Ando and J. Rüpke (eds), Public and Private in Ancient Mediterranean Law and Religion, Berlin: De Gruyter, pp. 53–74. Ekman, P. (1992), ‘An argument for basic emotions’, Cognition and Emotion 6: 169–200. Ekman, P. (1999), ‘Basic emotions’, in T. Dalgleish and T. Power (eds), The Handbook of Cognition and Emotion, Chichester: John Wiley, pp. 45–60. Ekman, P. and D. Cordaro (2011), ‘What is meant by calling emotions basic’, Emotion Review 3: 364–70. Ekman, P. and W. V. Friesen (1969), ‘The repertoire of non-verbal behaviour: Categories, origins, usage, and coding’, Semiotica 1: 49–98. Ekman, P., R. Levenson and W. Friesen (1983), ‘Autonomic nervous system activity distinguishing among emotions’, Science 221: 1208–10. Eldredge, S. (1996), Mask Improvisation for Actor Training and Performance, Evanston: Northwestern University Press. Ellsworth, P. C. and K. Scherer (2003), ‘Appraisal processes in emotion’, in R. Davidson, H. Goldsmith and K. Scherer (eds), Handbook of the Affective Sciences, Oxford: Oxford University Press, pp. 572–95. Enache, C. (2015), ‘The intelligence typology in Hippocrates’ De victu I 35’, Wiener Studien 128: 37–48. Engelmann, H. and D. Knibbe (1989), ‘Das Zollgesetz der Provinz Asia: Eine neue Inschrift aus Ephesos’, Epigraphica Anatolica 14: 1–206. Entralgo, P. (1970), La medicina Hipocrática, Madrid: Alianza. Epelboim, J. and P. Suppes (2001), ‘A model of eye movements and visual working memory during problem solving in geometry’, Vision Research 41.12: 1561–74. Esrock, E. (1994), The Reader’s Eye: Visual Imaging as Reader Response, Baltimore: Johns Hopkins University Press. Estany, A. and S. Martínez (2014), ‘ “Scaffolding” and “affordance” as integrative concepts in the cognitive sciences’, Philosophical Psychology 27.1: 1–14. Evans, J. (1999), ‘The material culture of Greek astronomy’, Journal for the History of Astronomy 30: 237–307. Evans, J. and C. C. Carman (2014), ‘Mechanical astronomy: A route to the ancient discovery of epicycles and eccentrics’, in N. Sidoli and G. Van Brummelen (eds), From Alexandria, through Baghdad: Surveys and Studies in the Ancient Greek and Medieval Islamic Mathematical Sciences in Honor of J. L. Berggren, Berlin: Springer, pp. 145–74.

254

bibliograp h y

Everson, S. (1991), ‘Introduction’, in S. Everson (ed.), Psychology: Companions to Ancient Thought, vol. 1, Cambridge: Cambridge University Press, pp. 1–12. Fagan, G. G. (2011), The Lure of the Arena: Social Psychology and the Crowd at the Roman Games, Cambridge: Cambridge University Press. Farenga, V. (2006), Citizen and Self in Ancient Greece: Individuals Performing Justice and the Law, Cambridge: Cambridge University Press. Fauconnier, G. and M. Turner (2002), The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities, New York: Basic Books. Fauconnier, G. and M. Turner (2008), ‘Conceptual projection and middle spaces’, UCSD: Department of Cognitive Science Technical Report 9401. Fay, N., C. Lister, T. Ellison and S. Goldin-Meadow (2014), ‘Creating a communication system from scratch: Gesture beats vocalization hands down’, Frontiers in Psychology 5: 354–9. Fedriani, C. (2014), Experiential Constructions in Latin, Leiden: Brill. Fejfer, J. (2008), Roman Portraits in Context, Berlin: De Gruyter. Feke, J. (2012), ‘Mathematizing the soul: The development of Ptolemy’s psychological theory from On the Kritêrion and Hêgemonikon to the Harmonics’, Studies in History and Philosophy of Science Part A, 43.4: 585–94. Fernandez, J. (1986), Persuasions and Performances: The Play of Tropes in Culture, Bloomington: Indiana University Press. Fernandez, J. (ed.) (1991), Beyond Metaphor: The Theory of Tropes in Anthropology, Stanford: Stanford University Press. Fink, R. (1971), Roman Military Records on Papyrus, Oxford: Oxford University Press. Finnegan, R. (1988), Literacy and Orality: Studies in the Technology of Communication, Oxford: Blackwell. Fitzgerald, W. (2000), Slavery and the Roman Literary Imagination, Cambridge: Cambridge University Press. Foley, H. (1980), ‘The masque of Dionysus’, Transactions of the American Philological Association 110: 107–33. Folkerts, M. (2002), ‘Ptolemaeus [65]’, in H. Cancik, H. Schneider, C. Salazar and D. Orton (eds), Brill’s New Pauly: Encyclopaedia of the Ancient World: Antiquity (English edn), Leiden and Boston: Brill; online at (last accessed 3 January 2018). Fortenbaugh, W. W. (1975), Aristotle on Emotion, London: Duckworth. Fortenbaugh, W. W. (2002), Aristotle on Emotion, 2nd edn, London: Duckworth. Frazer, J. G. (1906–15), The Golden Bough: A Study in Magic and Religion, 3rd edn, 12 vols, London: Macmillan. Frede, M. (1992), ‘On Aristotle’s conception of the soul’, in M. C. Nussbaum and A. Rorty (eds), Essays on Aristotle’s De Anima, Oxford: Oxford University Press, pp. 93–106. French, P. (1995), Corporate Ethics, Fort Worth: Harcourt Brace College Publishers. Fried, I., C. Wilson, K. MacDonald and E. Behnke (1998), ‘Electric current stimulates laughter’, Nature 391: 650.



bibliogra p h y 255

Friedenberg, J. and G. Silverman (2016), Cognitive Science: An Introduction to the Study of Mind, Thousand Oaks: Sage Publications. Frijda, N. (1986), The Emotions, Cambridge: Cambridge University Press. Frijda, N. (1993), ‘The place of appraisal in emotion’, Cognition and Emotion 7: 357–87. Frontisi-Ducroux, F. (1991), Le dieu-masque: une figure du Dionysos d’Athènes, Paris: La Découverte. Fuchs, T. (2005), ‘Corporealized and disembodied minds: A phenomenological view of the body in melancholia and schizophrenia’, Philosophy, Psychiatry and Psychology 12.2: 95–107. Fuchs, T. and H. De Jaegher (2009), ‘Enactive intersubjectivity: Participatory sense-making and mutual incorporation’, Phenomenology and the Cognitive Sciences 8: 465–86. Gagné, R. and M. Hopman (eds) (2013), Choral Mediations in Greek Tragedy, Cambridge: Cambridge University Press. Gallagher, S. (2001), ‘The practice of mind: Theory, simulation or primary interaction?’, Journal of Consciousness Studies 8.5–7: 83–108. Gallagher, S. (2004), ‘Neurocognitive models of schizophrenia: A neurophenomenological critique’, Psychopathology 37: 8–19. Gallagher, S. (2005), How the Body Shapes the Mind, Oxford: Oxford University Press. Gallagher, S. (2006), ‘The narrative alternative to theory of mind’, in R. Menary (ed.), Radical Enactivism: Intentionality, Phenomenology, and Narrative, Amsterdam: John Benjamins, pp. 223–9. Gallagher, S. (2008), ‘Direct perception in the intersubjective context’, Consciousness and Cognition 17.2: 535–43. Gallagher, S. (2009), ‘Philosophical antecedents of situated cognition’, in P. Robbins and M. Aydede (eds), The Cambridge Handbook of Situated Cognition, Cambridge: Cambridge University Press, pp. 35–51. Gallagher, S. (2012), ‘Empathy, simulation, and narrative’, Science in Context 25: 355–81. Gallagher, S. (2013), ‘The socially extended mind’, Cognitive Systems Research 25–6: 4–12. Gallagher, S. (2014), ‘Embodied cognition’, History of Distributed Cognition Project, (last accessed 6 February 2018). Gallagher, S. and D. D. Hutto (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: John Benjamins, pp. 17–38. García-Jurado, F. (2000), ‘Las metáforas de la vida cotidiana en latín’, Proceedings of the Congreso internacional de semántica 2, Madrid: Universidad de La Laguna, pp. 1571–84. Geertz, C. (1973), The Interpretation of Cultures, New York: Basic Books.

256

bibliograp h y

Gentner, D., M. Imai and L. Boroditsky (2002), ‘As time goes by: Evidence for two systems in processing space-time metaphors’, Language and Cognitive Processes 27.5: 537–65. Germany, R. (2016), Mimetic Contagion: Art and Artifice in Terence’s Eunuch, Oxford: Oxford University Press. Gernet, L. (1968), Anthropologie de la Grèce Antique, Paris: Maspero. Gibbs, R. (1994), The Poetics of Mind: Figurative Thought, Language, and Understanding, Cambridge: Cambridge University Press. Gibbs, R. (2005), Embodiment and Cognitive Science, Cambridge: Cambridge University Press. Gibbs, R. and H. Colston (1995), ‘The cognitive psychological reality of image schemas and their transformations’, Cognitive Linguistics 6: 347–78. Gierdien, F. (2009), ‘Musings on multiplication tables and associated mathematics and teaching practices’, Pythagoras 70: 16–31. Giere, R. (2002), ‘Models as parts of distributed cognitive systems’, in L. Magnani and N. Nersessian (eds), Model Based Reasoning: Science, Technology, Values, Dordrecht: Kluwer, pp. 227–41. Gill, C. (1996), Personality in Greek Epic, Tragedy, and Philosophy: The Self in Dialogue, Oxford: Oxford University Press. Gill, C. (2006), The Structured Self in Hellenistic and Roman Thought, Oxford: Oxford University Press. Gill, C. (2007), ‘Galen and the stoics: Mortal enemies or blood brothers?’, Phronesis: A Journal for Ancient Philosophy, 52.1: 88–120. Gill, C. (2010), Naturalistic Psychology in Galen and Stoicism, Oxford: Oxford University Press. Gish, D. (2012), ‘Defending dēmokratia: Athenian justice and the trial of the Arginusae generals in Xenophon’s Hellenica’, in F. Hobden and C. Tuplin (eds), Xenophon: Ethical Principles and Historical Enquiry, Leiden: Brill, pp. 161–212. Glautier, M. (1972), ‘A study in the development of accounting in Roman times’, Revue internationale des droits de l’antiquité 19: 311–43. Goatly, A. (2007), Washing the Brain: Metaphor and Hidden Ideology, Amsterdam: John Benjamins. Goldbeck, F. (2010), Salutationes: Die Morgenbegrüßungen in Rom in der Republik und der frühen Kaiserzeit, Berlin: De Gruyter. Goldhill, S. (2007), How to Stage Greek Tragedy Today, Chicago: University of Chicago Press. Goldin-Meadow, S. (2005), Hearing Gesture: How Our Hands Help Us Think, Cambridge, MA: Harvard University Press. Goldman, A. (2012), ‘A moderate approach to embodied cognitive science’, Review of Philosophy and Psychology 3.1: 71–88. Goldman, A. and F. de Vignement (2009), ‘Is social cognition embodied?’, Trends in Cognitive Sciences 15.10: 154–9. Gombrich, E. H. (1982), The Image and the Eye: Further Studies in the Psychology of Pictorial Representation, Ithaca: Cornell University Press.



bibliogra p h y 257

Goody, J. (1977), The Domestication of the Savage Mind, Cambridge: Cambridge University Press. Goody, J. (1988), The Interface between the Written and the Oral, Cambridge: Cambridge University Press. Gottlieb, P. (2009), The Virtue of Aristotle’s Ethics, Cambridge: Cambridge University Press. Gottschall, J. and D. S. Wilson (eds) (2005), The Literary Animal: Evolution and the Nature of Narrative, Evanston: Northwestern University Press. Gould, J. (1996), ‘Tragedy and collective experience’, in M. S. Silk (ed.), Tragedy and the Tragic: Greek Theatre and Beyond, Oxford: Oxford University Press, pp. 217–43. Grant, M. (trans.) (1996), Tacitus: The Annals of Imperial Rome, Harmondsworth: Penguin. Grasshoff, G. and A. Stückelberger (2006), Klaudios Ptolemaios Handbuch der Geographie, vols 1–2, Basel: Schwabe. Graver, M. (2007), Stoicism and Emotion, Chicago: University of Chicago Press. Green, C. M. C. (2007), Roman Religion and the Cult of Diana at Aricia, Cambridge: Cambridge University Press. Greenough, J. (1890), ‘Some Latin etymologies’, Harvard Studies in Classical Philology 1: 93–105. Greenspan, P. S. (1988), Emotions and Reasons: An Inquiry into Emotion Justification, New York and London: Routledge. Gregory, R. L. (1997), ‘Knowledge in perception and illusion’,  Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences 352.1358: 1121–7. Grethlein, J. (2015a), ‘Is narrative “the description of fictional mental functioning”? Heliodorus against Palmer, Zunshine & Co.’, Style 49: 257–84. Grethlein, J. (2015b), ‘Social minds and narrative time: Collective experience in Thucydides and Heliodorus’, Narrative 23: 123–39. Grethlein, J. and L. Huitink (2017), ‘Homer’s vividness: An enactive approach’, Journal of Hellenic Studies 137: 67–91. Grünbaum, T. (2007), ‘Action between plot and discourse’, Semiotica 165: 295–314. Gualandi, G. (1977), ‘Una testa di Adriano da Hierapolis (Frigia)’, Rivista di archeologia 1: 64–77. Gummere, R. M. (trans.) (1917), Epistles, in Seneca, vol. 4, Cambridge, MA: Harvard University Press. Gunderson, E. (2015), The Sublime Seneca, Cambridge: Cambridge University Press. Gundert, B. (1992), ‘Parts and their roles in Hippocratic medicine’, Isis 83: 453–65. Gundert, B. (2000), ‘Soma and psyche in Hippocratic medicine’, in J. P. Wright and P. Potter (eds), Psyche and Soma: Physicians and Metaphysicians on the Mind-Body Problem from Antiquity to Enlightenment, Oxford: Oxford University Press, pp. 13–35. Gurd, S. (2007), ‘Cicero and editorial revision’, Classical Antiquity 26: 49–80. Gurd, S. (2012), Work in Progress: Literary Revision as Social Performance in Ancient Rome, Oxford: Oxford University Press. Gutchess, A. and A. Indeck (2009), ‘Cultural influences on memory’, in J. Chiao

258

bibliograp h y

(ed.), Cultural Neuroscience: Cultural Influences on Brain Function, Progress in Brain Research 178, New York: Elsevier, pp. 137–50. Habinek, T. (2005a), ‘Slavery and class’, in S. Harrison (ed.), A Companion to Latin Literature, Malden, MA: Blackwell, pp. 385–93. Habinek, T. (2005b), The World of Roman Song: From Ritualized Speech to Social Order, Baltimore and London: Johns Hopkins University Press. Habinek, T. (2009), ‘Situating literacy at Rome’, in W. Johnson and H. Parker (eds), Ancient Literacies, Oxford: Oxford University Press, pp. 114–40. Habinek, T. (2011), ‘Tentacular mind: Stoicism, neuroscience, and the configurations of physical reality’, in B. Stafford (ed.), Bridging the Gap between Neuroscience and the Humanities: A Field Guide to a New Meta-Field, Chicago: University of Chicago Press, pp. 64–83. Habinek, T. (2017), ‘Rhetoric, music, and the arts’, in M. MacDonald (ed.), The Oxford Handbook of Rhetorical Studies, Oxford: Oxford University Press, pp. 289–99. Häfner, R. (2011), Mysterien im Hain von Ariccia: Nicolas Poussins ‘Landschaft mit Numa Pompilius und der Nymphe Egeria’ im intellektuellen Kontext um 1630, Munich: Fink. Halbwachs, M. (1952), Les cadres sociaux de la memoire, Paris: Puf. Hall, J. D. (1994), ‘Persuasive design in Cicero’s De oratore’, Phoenix 48: 210–25. Halliwell, S. (1993), ‘The function and aesthetics of the Greek tragic mask’, in N. Slater and B. Zimmermann (eds), Intertextualität in der griechisch-römischen Komödie, vol. 2, Stuttgart: M & P, pp. 195–211. Halliwell, S. (2011), Between Ecstasy and Truth: Interpretations of Greek Poetics from Homer to Longinus, Oxford: Oxford University Press. Hamm, O., H. Schupp and A. Weike (2003), ‘Motivational organization of emotions: Autonomic changes, cortical responses, and reflex modulation’, in R. Davidson, K. Scherer and H. Goldsmith (eds), Handbook of Affective Sciences, Oxford: Oxford University Press, pp. 187–211. Hammer, D. (2002), The Iliad as Politics: The Performance of Political Thought, Norman: University of Oklahoma Press. Hampe, B. and J. Grady (eds) (2005), From Perception to Meaning: Image Schemas in Cognitive Linguistics, Berlin: De Gruyter. Han, S. and Y. Ma (2014), ‘Cultural differences in human brain activity: A quantitative meta-analysis’, NeuroImage 99: 293–300. Hansen, M. H. (1987), The Athenian Assembly in the Age of Demosthenes, Oxford: Oxford University Press. Hansen, M. H. (2004), ‘How did the Athenians vote?’, in P. J. Rhodes (ed.), Athenian Democracy, Edinburgh: Edinburgh University Press, pp. 40–61. Hardie, A. (2015), ‘The Camenae in cult, history, and song’, Classical Antiquity 34.2: 45–85. Harris, E. M. (2013), The Rule of Law in Action in Democratic Athens, Oxford: Oxford University Press. Harris, W. V. (1989), Ancient Literacy, Cambridge, MA: Harvard University Press. Hart, M. L. (2010), The Art of Ancient Greek Theater: Exhibition Catalogue, Los Angeles: J. Paul Getty Museum.



bibliogra p h y 259

Harwood, F. (1976), ‘Myth, memory and the oral tradition: Cicero in the Trobriands’, American Anthropologist 78.4: 783–96. Hasher, L. and R. Zacks (1979), ‘Automatic and effortful process in memory’, Journal of Experimental Psychology: General 108: 356–88. Haugeland, J. (1985), Artificial Intelligence: The Very Idea, Cambridge, MA: MIT Press. Haugeland, J. (1998), ‘Mind embodied and embedded’, in J. Haugeland, Having Thought: Essays in the Metaphysics of Mind, Cambridge, MA: Harvard University Press, pp. 207–37. Havelock, E. A. (1963), Preface to Plato, Cambridge, MA: Harvard University Press/ Oxford: Blackwell. Havelock, E. A. (1986), The Muse Learns to Write: Reflections on Orality and Literacy from Antiquity to the Present, New Haven, CT: Yale University Press. Hayles, N. K. (1999), How We Became Posthuman: Virtual Bodies in Cybernetics, Literature and Informatics, Chicago: University of Chicago Press. Heath, M. (2009), ‘Codifications of rhetoric’, in E. Gunderson (ed.), The Cambridge Companion to Ancient Rhetoric, Cambridge: Cambridge University Press, pp. 59–75. Heersmink, R. (2013), ‘A taxonomy of cognitive artifacts: Function, information, and categories’, Review of Philosophy and Psychology 4.3: 465–81. Heiberg, J. L. (1898–1903), Claudii Ptolemaei opera quae exstant omnia, vols 1.1–1.2, Leipzig: Teubner. Henderson, J. (2006), ‘Journey of a lifetime’, in K. Volk and G. Williams (eds), Seeing Seneca Whole, Leiden: Brill, pp. 123–46. Herman, G. (1987), Ritualised Friendship and the Greek City, Cambridge: Cambridge University Press. Hermans, H. and A. Hermans-Konopka (2010), Dialogical Self Theory: Positioning and Counter-Positioning in a Globalizing Society, Cambridge: Cambridge University Press. Hesk, J. (2011), ‘Euripidean euboulia and the problem of “tragic politics”’, in D. M. Carter (ed.), Why Athens? A Reappraisal of Tragic Politics, Oxford: Oxford University Press, pp. 119–43. Heyes, C. M. (2014), ‘Group-mindedness’, Current Biology 24.9: R344–R346. Hitz, Z. (2001), ‘Aristotle on self-knowledge and friendship’, Philosophers’ Imprint 11: 1–28. Hohwy, J. (2013), The Predictive Mind, Oxford: Oxford University Press. Holland, N. (1988), The Brain of Robert Frost: A Cognitive Approach to Literature, London and New York: Routledge. Holmes, B. (2010), The Symptom and the Subject: The Emergence of the Physical Body in Ancient Greece, Princeton: Princeton University Press. Honneth, A. (2005), Reification: A Recognition-Theoretical View (The Tanner Lectures on Human Values, Delivered at University of California, Berkeley, March 14–16, 2005), (last accessed 6 February 2018). Horn, F. (2015a), ‘Sleeping the brazen slumber: A cognitive approach to Hom. Il. 11.241’, Philologus 159: 197–206.

260

bibliograp h y

Horn, F. (2015b), ‘Visualising Hom. Il. 3.57: “Putting on the mantle of stone”’, Rheinisches Museum für Philologie 158: 1–7. Horn, F. (2016a), ‘ “Building in the deep”: Notes on a metaphor for mental activity and the metaphorical concept of mind in early Greek epic’, Greece & Rome 63: 163–74. Horn, F. (2016b), ‘ “Furious wasps” and “tuna in the frying-pan”: Animal metaphors, intertextuality, and cognition in Lycophron’s Alexandra’, Museum Helveticum 73: 154–66. Horn, F. and C. Breytenbach (eds) (2016), Spatial Metaphors: Ancient Texts and Transformations, Berlin: Topoi; available online at (last accessed 6 February 2018). Hornblower, S. (2011), ‘ “This was decided” (ἔδοξε ταῦτα): The army as polis in Xenophon’s Anabasis – and elsewhere’, in S. Hornblower, Thucydidean Themes, Oxford: Oxford University Press, pp. 226–49. Horwitz, A. V. and J. C. Wakefield (2007), The Loss of Sadness: How Psychiatry Transformed Normal Sorrow into Depressive Disorder, Oxford: Oxford University Press. Hu, J., G. Nagy, R. Kashi, G. Wilfong and D. Lopresti (2001), ‘Why table ground-truthing is hard’, in Proceedings of the Sixth International Conference on Document Analysis and Recognition, Los Alamitos: IEEE Computer Society, pp. 129–33. Huebner, B. (2014), Macrocognition: A Theory of Distributed Minds and Collective Intentionality, Oxford: Oxford University Press. Humphrey, J. H. (ed.) (1991), Literacy in the Roman World, Journal of Roman Archaeology Supplement 3, Ann Arbor: University of Michigan Press. Hurley, S. (1998), Consciousness in Action, Cambridge, MA: Harvard University Press. Hurley, S. (2010), ‘The varieties of externalism’, in R. Menary (ed.), The Extended Mind, Cambridge, MA: MIT Press, pp. 101–53. Hurley, S. and N. Chater (eds) (2005), Perspectives on Imitation: From Neuroscience to Social Science, 2 vols, Cambridge, MA: MIT Press. Hurley, S. and A. Noë (2003), ‘Neural plasticity and consciousness’, Biology and Philosophy 18: 131–68. Hurley, S. and L. Novick (2010), ‘Solving problems using matrix, network, and hierarchy diagrams: The consequences of violating construction conventions’, The Quarterly Journal of Experimental Psychology 63: 275–90. Hutchins, E. (1995), Cognition in the Wild, Cambridge, MA: MIT Press. Hutchins, E. (2001), ‘Distributed cognition’, in N. Smelser and P. Baltes (eds), The International Encyclopedia of the Social and Behavioral Sciences, Oxford: Pergamon, pp. 2068–72. Hutchins, E. (2005), ‘Material anchors for conceptual blends’, Journal of Pragmatics 37.10: 1555–77. Hutchins, E. (2006), ‘The distributed cognition perspective on human interaction’, in S. Levinson and N. Enfield (eds), Roots of Human Sociality: Culture, Cognition, and Interaction, London: Bloomsbury, pp. 375–98.



bibliogra p h y 261

Hutchins, E. (2010), ‘Enaction, imagination, and insight’, in J. R. Stewart, O. Gapenne and E. A. Di Paolo (eds), Enaction: Toward a New Paradigm for Cognitive Science, Cambridge, MA: MIT Press, pp. 425–50. Hutto, D. D. (2007), ‘Narrative and understanding persons’, in D. D. Hutto (ed.), Narrative and Understanding Persons, Cambridge: Cambridge University Press, pp. 1–16. Hutto, D. D. (2008a), Folk Psychological Narratives: The Sociocultural Basis of Understanding Reasons, Cambridge, MA: MIT Press. Hutto, D. D. (2008b), ‘The Narrative Practice Hypothesis: Clarifications and implications’, Philosophical Explorations 11.3: 175–92. Hutto, D. D. (ed.) (2009), Narrative and Folk Psychology, special issue of Journal of Consciousness Studies, 16.6–8. Hutto, D. and E. Myin (2012), Radicalizing Enactivism: Basic Minds without Content, Cambridge, MA: MIT Press. Innocenti, B. (1994), ‘Towards a theory of vivid description as practiced in Cicero’s Verrine Orations’, Rhetorica 12.4: 355–81. Inwood, B. (1985), Ethics and Human Action in Early Stoicism, Oxford: Oxford University Press. Inwood, B. and L. P. Gerson (1997), Hellenistic Philosophy: Introductory Readings, 2nd edn, Indianapolis: Hackett. Inwood, B. and R. Woolf (trans.) (2013), Aristotle: Eudemian Ethics, Cambridge: Cambridge University Press. Iser, W. (1978), The Act of Reading: A Theory of Aesthetic Response, Baltimore: Johns Hopkins University Press. Ishii, M. (1994), ‘The Noh theater: Mirror, mask, and madness’, Comparative Drama 28.1: 43–66. Jackson, L. (2014), ‘The Athenian dramatic chorus in the fourth century bc’, unpublished DPhil dissertation, University of Oxford. Jacob, C. (2006), The Sovereign Map: Theoretical Approaches in Cartography throughout History, Chicago: University of Chicago Press. Jaeger, M. (1997), Livy’s Written Rome, Ann Arbor: University of Michigan Press. Jajdelska, E., C. Butler, S. Kelly, A. McNeill and K. Overy (2010), ‘Crying, moving and keeping it whole: What makes literary description vivid?’, Poetics Today 31.3: 433–63. Jäkel, O. (1995), ‘The metaphorical concept of mind: “Mental activity is manipulation”’, in J. Taylor and R. MacLaury (eds), Language and the Cognitive Construal of the World, Berlin: De Gruyter, pp. 197–229. James, W. (1884), ‘What is an emotion?’, Mind 9: 188–205. James, W. [1890] (1981), The Principles of Psychology, ed. F. Bowers and F. Burkhardt, Cambridge, MA, and London: Harvard University Press. Johansen, T. K. (2004), Plato’s Natural Philosophy: A Study of the Timaeus-Critias, Cambridge: Cambridge University Press. Johansen, T. K. (2012), The Powers of Aristotle’s Soul, Oxford: Oxford University Press.

262

bibliograp h y

Johansson, G. (1973), ‘Visual perception of biological motion and a model for its analysis’, Perception and Psychophysics 14: 201–11. Johansson, R., J. Holsanova and K. Holmqvist (2006), ‘Pictures and spoken descriptions elicit similar eye movement during mental imagery, both in light and in complete darkness’, Cognitive Science 30: 1053–79. Johnson, M. (1987), The Body in the Mind: The Bodily Basis of Meaning, Imagination and Reason, Chicago: University of Chicago Press. Johnson, W. A. and H. N. Parker (2009), Ancient Literacies: The Culture of Reading in Greece and Rome, Oxford: Oxford University Press. Johnstone, S. (2011), A History of Trust in Ancient Greece, Chicago: University of Chicago Press. Jola, C. and M. Reason (2016), ‘Audiences’ experience of proximity and co-presence in live dance performance’, in C. Falletti, G. Sofia and V. Jacono (eds), Theatre and Cognitive Neuroscience, London: Bloomsbury, pp. 75–92. Jones, A. (2008), ‘Ptolemy’, in P. T. Keyser and G. Irby-Massie (eds), The Encyclopedia of Ancient Natural Scientists: The Greek Tradition and its Many Heirs, London and New York: Routledge, pp. 706–7. Jones, J. (1962), On Aristotle and Greek Tragedy, Oxford: Oxford University Press. Jones, W. H. S. (ed.) (1923a), Hippocrates: Vol. I, Cambridge, MA; Harvard University Press (Loeb Classical Library). Jones, W. H. S. (ed.) (1923b), Hippocrates: Vol. II, Cambridge, MA: Harvard University Press (Loeb Classical Library). Jones, W. H. S. (ed.) (1931), Hippocrates: Vol. IV, Cambridge, MA: Harvard University Press (Loeb Classical Library). Joost-Gaugier, C. (2002), Raphael’s Stanza della Segnatura: Meaning and Invention, Cambridge: Cambridge University Press. Jouanna, J. (1990), De l’ancienne médecine, Paris: Les Belles Lettres. Jouanna, J. (2007), ‘La theorie de la sensation, de la pensée et de l’âme dans le traité hippocratique du Régime: Ses rapports avec Empedocle et le Timée de Platon’, AION: Annali dell’Istituto universitario orientale di Napoli: Sezione filologico-let​ teraria 29: 9–38. Jouanna, J. (2013), ‘The typology and aetiology of madness in ancient Greek medical and philosophical writings’, in W. V. Harris (ed.), Mental Disorders in the Classical World, Leiden: Brill, pp. 97–118. Kadioĝlu, M. (2006), Die Scaenae frons des Theaters von Nysa am Mäander, Mainz: Philipp von Zabern. Kahn, C. (1992), ‘Aristotle on thinking’, in A. O. Rorty and M. C. Nussbaum (eds), Essays on Aristotle’s De Anima, Oxford: Oxford University Press, pp. 359–80. Kaimio, M. (1970), The Chorus of Greek Drama within the Light of the Person and Number Used, Helsinki: Societas Scientiarum Fennica. Kazantzidis, G. (2017), ‘Empathy and the limits of disgust in the Hippocratic Corpus’, in D. Lateiner and D. Spatharas (eds), The Ancient Emotion of Disgust, Oxford: Oxford University Press, pp. 45–68.



bibliogra p h y 263

Kazantzidis, G. (forthcoming), ‘Medical and scientific understandings’, in D. Cairns (ed.), A Cultural History of Emotions, vol. 1, London: Bloomsbury. Kelly, J. R. (2001), ‘Mood and emotion in groups’, in M. A. Hogg and R. S. Tindale (eds), Blackwell Handbook of Social Psychology: Group Process, Malden, MA: Blackwell, pp. 164–81. Kemmerer, D. (2006), ‘Action verbs, argument structure constructions, and the mirror neuron system’, in M. A. Arib (ed.), Action to Language via the Mirror Neuron System, Cambridge: Cambridge University Press, pp. 347–73. Kennedy, J. (1981), ‘Arithmetic with Roman numerals’, American Mathematical Monthly 88.1: 29–32. Kent Sprague, R. (1985), ‘Aristotle on red mirrors (On Dreams II 459b24–460a23)’, Phronesis 30: 323–6. Kenty, J. (2017), ‘Cicero’s representation of an oral community in De Oratore’, in N. Slater (ed.), Voice and Voices in Antiquity, Leiden: Brill, pp. 351–76. Kesner, R. and J. Martinez (eds) (2007), Neurobiology of Learning and Memory, Boston: Academic Press. Keyser, P. T. and G. Irby-Massie (2008), The Encyclopedia of Ancient Natural Scientists: The Greek Tradition and its Many Heirs, London and New York: Routledge. King, H. (2013a), ‘Fear of the flute girls, fear of falling’, in W. V. Harris (ed.), Mental Disorders in the Classical World, Leiden: Brill, pp. 265–82. King, H. (2013b), ‘Female fluids in the Hippocratic Corpus: How solid was the humoral body?’, in P. Horden and E. Hsu (eds), The Body in Balance: Humoral Medicines in Practice, Oxford: Berghahn Books, pp. 25–49. Kirsh, D. (1995), ‘The intelligent use of space’, Artificial Intelligence 73.1: 31–68. Kirsh, D. (2008), ‘Problem solving and situated cognition’, in P. Robbins and M. Aydede (eds), The Cambridge Handbook of Situated Cognition, Cambridge: Cambridge University Press, pp. 264–306. Kirsh, D. (2009), ‘Projections, problem space, and anchoring’, in N. A. Taatgen and H. van Rijn (eds), Proceedings of the 31st Annual Conference of the Cognitive Science Society, Austin: Cognitive Science Academy, pp. 2310–15. Kirsh, D. (2010), ‘Thinking with external representations’, AI & Society 25: 441–54. Kirsh, D. (2013a), ‘Embodied cognition and the magical future of interaction design’, ACM Transactions on Computer-Human Interaction (TOCHI) 20.1: 2–32. Kirsh, D. (2013b), ‘Thinking with external representations’, in S. Cowley and F. Vallée-Tourangeau (eds), Cognition Beyond the Brain: Computation, Interactivity and Human Artifice, London: Springer, pp. 171–94. Kirsh, D. and P. Maglio (1994), ‘On distinguishing epistemic from pragmatic action’, Cognitive Science 18: 513–49. Kleinsmith, A. and N. Bianchi-Berthouze (2013), ‘Affective body expression perception and recognition: A survey’, IEEE Transactions on Affective Computing 4.1: 15–33. Knappet, C. (2002), ‘Photographs, skeumorphs and marionettes: Some thoughts on mind, agency and object’, Journal of Material Culture 7.1: 97–117.

264

bibliograp h y

Knuuttila, S. (2004), Emotions in Ancient and Medieval Philosophy, Oxford: Oxford University Press. Konstan, D. (1997), Friendship in the Classical World, Cambridge: Cambridge University Press. Konstan, D. (2006), The Emotions of the Ancient Greeks: Studies in Aristotle and Classical Literature, Toronto: University of Toronto Press. Konstan, D. (2017), ‘Reason vs emotion in Seneca’, in D. Cairns and D. Nelis (eds), Emotions in the Classical World: Methods, Approaches, and Directions, Stuttgart: Steiner, pp. 231–43. Korshak, Y. (1987), Frontal Faces in Attic Vase Painting of the Archaic Period, Chicago: Ares. Kosslyn, S. (2006), ‘On the evolution of human motivation: The role of social prosthetic systems’, in S. Platek and J. Keenan (eds), Evolutionary Cognitive Neuroscience, Cambridge, MA: MIT Press, pp. 541–54. Kovacs, D. (ed.) (2003), Euripides, vol. 6, Cambridge, MA: Harvard University Press. Kövecses, Z. (2002), Metaphor, Oxford: Oxford University Press. Kövecses, Z. (2005), Metaphor in Culture: Universality and Variation, Cambridge: Cambridge University Press. Kövecses, Z. (2006), Language, Mind and Culture, Oxford: Oxford University Press. Kraus, C. (1994), Livy: Ab Urbe Condita, Book VI, Cambridge: Cambridge University Press. Krebs, A. (2009), ‘ “Wie ein Bogenstrich, der aus zwei Saiten eine Stimme zieht”: Eine dialogische Philosophie der Liebe’, Deutsche Zeitschrift für Philosophie 57: 729–43. Kreibig, S. (2010), ‘Autonomic nervous system activity in emotion: A review’, Biological Psychology 84: 394–421. Krueger, J. and T. Szanto (2016), ‘Extended emotions’, Philosophy Compass 11: 863–78. Krumeich, R., N. Pechstein and B. Seidensticker (1998), Das griechische Satyrspiel, Darmstadt: Wissenschaftliche Buchgesellschaft. Kuijpers, M. M. and D. S. Miall (2011), ‘Bodily involvement in literary reading: An experimental study of readers’ bodily experiences during reading’, in F. Hakemulder (ed.), De stralende lezer: Wetenschappelijk onderzoek naar de invloed van het lezen, Delft: Eburon, pp. 160–74. Kukkonen, K. (2014), ‘Presence and prediction: The embodied reader’s cascades of cognition’, Style 48.3: 367–84. Kukkonen, K. and M. Caracciolo (2014), ‘Introduction: What is the second generation?’, Style 48.3: 261–74. Kuzmičová, A. (2012a), ‘Fidelity without mimesis: On mental imagery from visual description’, in G. Currie, P. Kot’átko and M. Pokorný (eds), Mimesis: Metaphysics, Cognition, Pragmatics, London: College Publications, pp. 273–315. Kuzmičová, A. (2012b), ‘Presence in the reading of literary narrative: A case for motor enactment’, Semiotica 189: 23–48.



bibliogra p h y 265

Kuzmičová, A. (2013), ‘The words and worlds of literary narrative: The tradeoff between verbal presence and direct presence in the activity of reading’, in L. Bernaerts, D. de Geest, L. Herman and B. Vervaeck (eds), Stories and Minds: Cognitive Approaches to Literary Narrative, Lincoln and London: University of Nebraska Press, pp. 107–28. Lacan, J. (1981), The Language of the Self, trans. A. Wilden, Baltimore: John Hopkins University Press. Lakoff, G. (1987), Women, Fire and Dangerous Things: What Categories Reveal about the Mind, Chicago: University of Chicago Press. Lakoff, G. (1993), ‘The contemporary theory of metaphor’, in A. Ortony (ed.), Metaphor and Thought, Cambridge: Cambridge University Press, pp. 202–51. Lakoff, G. and M. Johnson (1980), Metaphors We Live By, Chicago: University of Chicago Press. Lakoff, G. and M. Johnson (1999), Philosophy in the Flesh: The Embodied Mind and its Challenges to Western Thought, New York: Basic Books. Lami, A. (2007), ‘Lo scritto ippocratico sui disturbi virginali’, Galenos 1: 15–59. Lammers, J. (1931), Die Doppel- und Halbchöre in der antiken Tragödie, Münster and Paderborn: Schöningh. Lammert, F. (1961), Claudii Ptolemaei opera quae exstant omnia, 2nd edn, vol. 3.2, Leipzig: Teubner. Lang, P. J. (1995), ‘The emotion probe: Studies of motivation and attention’, American Psychologist 50: 372–85. Lang, P. J., M. Bradley, J. Fitzsimmons, B. Cuthbert, J. Scott, B. Moulder and V. Nangia (1998), ‘Emotional arousal and activation of the visual cortex: An fMRI analysis’, Psychophysiology 35: 199–210. Langholf, V. (1990), Medical Theories in Hippocrates: Early Texts and the ‘Epidemics’, Berlin: De Gruyter. Larmour, D. and D. Spencer (eds) (2007), The Sites of Rome: Time, Space, Memory, Oxford: Oxford University Press. Lausberg, H. (1998), Handbook of Literary Rhetoric: A Foundation for Literary Study, trans. D. E. Norton and R. Dean Anderson, Leiden: Brill. Lazarus, R. (1966), Psychological Stress and the Coping Process, New York: McGraw-Hill. Lazarus, R. (1982), ‘Thoughts on the relation between emotion and cognition’, American Psychologist 37.9: 1019–24. Lazarus, R. (1991), Emotion and Adaptation, Oxford: Oxford University Press. Le Bon, G. [1895] (1960), The Crowd: A Study of the Popular Mind, Harmondsworth: Penguin. Le Goff, J. (1992), History and Memory, New York: Columbia University Press. Leach, E. (1988), The Rhetoric of Space: Literary and Artistic Representations of Landscape in Republican and Augustan Rome, Princeton: Princeton University Press. Leavis, F. R. (1962), Two Cultures? The Significance of C. P. Snow, London: Chatto and Windus. LeDoux, J. E. (1996), The Emotional Brain: The Mysterious Underpinnings of Emotional Life, New York: Simon & Schuster.

266

bibliograp h y

Leeman, A., H. Pinkster and E. Rabbie (eds) (1989), M. Tullius Cicero: De oratore libri III, 3. Band: Buch II, Heidelberg: Carl Winter. Lehoux, D. (2007), ‘Observers, objects, and the embedded eye; Or, seeing and knowing in Ptolemy and Galen’, Isis 98.3: 447–67. Leja, M. (2015), ‘Issues in early mass visual culture’, in J. Davis, J. Greenhill and J. La Fountain (eds), A Companion to American Art, Chichester: Wiley-Blackwell, pp. 507–24. Leunissen, M. (2010), Explanation and Teleology in Aristotle’s Science of Nature, Cambridge: Cambridge University Press. Lévi-Strauss, C. (1962), La Pensée sauvage, Paris: Plon. Liapis, V. (2012), A Commentary on the Rhesus Attributed to Euripides, Oxford: Oxford University Press. Liddell, H. G., R. Scott and H. Stuart Jones (1996), A Greek-English Lexicon, with revised supplement edited by P. G. W. Glare and A. A. Thomson, Oxford: Oxford University Press. Lindner, R. (1994), Mythos und Identität, Studien zur Selbstdarstellung kleinasiatischer Städte in der römischen Kaiserzeit, Stuttgart: Franz Steiner. Lintott, A. (1999), The Constitution of the Roman Republic, Oxford: Oxford University Press. List, C. and P. Pettit (2011), Group Agency: The Possibility, Design, and Status of Corporate Agents, Oxford: Oxford University Press. Liu, Z., N. J. Nersessian and J. T. Stasko (2008), ‘Distributed cognition as a theoretical framework for information visualization’, IEEE Transactions on Visualization and Computer Graphics (Proceedings Vis/InfoVis ’08) 14: 1173–80; available at (last accessed 7 March 2018). Llewellyn-Jones, L. (2005), ‘Body language and the female role player in Greek tragedy and Japanese kabuki theatre’, in D. Cairns (ed.), Body Language in the Greek and Roman Worlds, Swansea: Classical Press of Wales, pp. 73–105. Lo Presti, R. (2007), ‘The ambiguous role of perception: Empiricist views and biological perspectives among the Hippocratics’, Acta Classica 50: 129–46. Lo Presti, R. (2016), ‘Perceiving the coherence of the perceiving body: Is there such a thing as a “Hippocratic” view on sense perception and cognition?’, in L. Dean-Jones and R. M. Rosen (eds), Ancient Concepts of the Hippocratic, Leiden: Brill, pp. 163–94. Loftus, E., M. Banaji, J. Schooler and R. Foster (1987), ‘Who remembers what? Gender differences in memory’, Michigan Quarterly Review 26: 64–85. Long, A. A. (1996), Stoic Studies, Cambridge: Cambridge University Press. Long, A. A. (1999), ‘Stoic psychology’, in K. Algra, J. Barnes, J. Mansfeld and M. Schofield (eds), The Cambridge History of Hellenistic Philosophy, Cambridge: Cambridge University Press, pp. 560–84. Long, A. A. and D. N. Sedley (1987), The Hellenistic Philosophers, Cambridge: Cambridge University Press.



bibliogra p h y 267

Lotman, Y. (1990), Universe of the Mind: A Semiotic Theory of Culture, Bloomington: Indiana University Press. Loucas, I. (1989), ‘Ritual surprise and terror in ancient Greek possession-dromena’, Kernos: Revue internationale et pluridisciplinaire de religion grecque antique 2: 97–104. Lowe, N. J. (2000), The Classical Plot and the Invention of Western Narrative, Cambridge: Cambridge University Press. Lucchese, K. (2009), ‘Landscape synchesis: A Demeter temple in Latium’, in G. Casadio and P. Johnston (eds), Mystic Cults in Magna Graecia, Austin: University of Texas Press, pp. 161–89. Lulić, J. (2014), ‘Dalmatian Silvanus: A cognitive approach to re-interpretation of the reliefs representing Silvanus from Roman Dalmatia’, in H. Platts, J. Pearce, C. Barron, J. Lundock and J. Yoo (eds), TRAC 2013: Proceedings of the TwentyThird Theoretical Roman Archaeology Conference, Oxford: Oxbow, pp. 37–51. Luria, A. (1979), The Making of Mind, Cambridge, MA: Harvard University Press. McClary, S. (1995), ‘Music, the Pythagoreans, and the body’, in S. Foster (ed.), Choreographing History, Bloomington: Indiana University Press, pp. 82–104. Mackay, E. A. (2008), Orality, Literacy, Memory in the Ancient Greek and Roman World, Leiden: Brill. McNeill, D. (2015), Why We Gesture: The Surprising Role of Hand Movements in Communication, Cambridge: Cambridge University Press. Maher, D. and J. Makowski (2001), ‘Literary evidence for Roman arithmetic with fractions’, Classical Philology 96.4: 376–99. Malafouris, L. (2013), How Things Shape the Mind: A Theory of Material Engagement, Cambridge, MA: MIT Press. Manieri, A. (1998), L’immagine poetica nella teoria degli antichi: phantasia ed enargeia, Pisa and Rome: Istituti editoriali e poligrafici internazionali. Mankin, D. (ed.) (2011), Cicero: De Oratore Book III, Cambridge: Cambridge University Press. Maravati, A. and A. Iriki (2004), ‘Tools for the body (schema)’, Trends in Cognitive Science 8.2: 79–86. Marchese, F. (2013), ‘Tables and early information visualization’, in F. Marchese and E. Banissi (eds), Knowledge Visualization Currents: From Text to Art to Culture, London: Springer-Verlag, pp. 35–61. Marr, J. L. and P. J. Rhodes (2008), The ‘Old Oligarch’: The Constitution of the Athenians Attributed to Xenophon, Oxford: Aris & Phillips. Marshall, G. D. and P. G. Zimbardo (1979), ‘Affective consequences of inadequately explained physiological arousal’, Journal of Personality and Social Psychology 37: 970–88. Martinez, M. (2010), Learning and Cognition: The Design of the Mind, Upper Saddle River, NJ: Merrill. Meijering, R. (1987), Literary and Rhetorical Theories in Greek Scholia, Groningen: Forsten. Meineck, P. (2011), ‘The neuroscience of the tragic mask’, Arion 19: 113–58.

268

bibliograp h y

Meineck, P. (2012), ‘The embodied space: Performance and visual cognition at the fifth-century Athenian theatre’, New England Classical Journal 39: 3–46. Meineck, P. (2016), ‘Cognitive theory and Aeschylus: Translating beyond the lexicon’, in S. Constantinidis (ed.), The Reception of Aeschylus’ Plays through Shifting Models and Frontiers, Leiden: Brill, pp. 147–75. Meineck, P. (2017),  Theatrocracy: Greek Drama, Cognition and the Imperative for Theatre, Abingdon: Taylor and Francis. Menary, R. (2006), ‘Attacking the bounds of cognition’, Philosophical Psychology 19: 329–44. Menary, R. (2007), Cognitive Integration: Mind and Cognition Unbounded, Basingstoke: Palgrave Macmillan. Menary, R. (ed.) (2010), The Extended Mind, Cambridge, MA: MIT Press. Menninger, K. (1977), Number Words and Number Symbols: A Cultural History of Numbers, Cambridge, MA: Harvard University Press. Michaelian, K. (2012), ‘Is external memory memory? Biological memory and extended mind’, Consciousness and Cognition 21.3: 1154–65. Miller, F. D. (2012), ‘Aristotle on the separability of the mind’, in C. Shields (ed.), The Oxford Handbook of Aristotle, Oxford: Oxford University Press, pp. 306–39. Miller, G. (1956), ‘The magical number seven, plus or minus two: Some limits on our capacity for processing information’, The Psychological Review 63: 81–97. Miller, K. and D. Parades (1996), ‘On the shoulders of giants: Cultural tools and mathematical development’, in R. Sternberg and T. Ben-Zeev (eds), The Nature of Mathematical Thinking, Psychology Press, Mahwah, NJ: Erlbaum, pp. 83–117. Miller, W. I. (1997), The Anatomy of Disgust, Cambridge, MA: Harvard University Press. Minchin, E. (1992), ‘Scripts and themes: Cognitive research and the Homeric epic’, Classical Antiquity 11: 229–41. Minchin, E. (1996), ‘The performance of lists and catalogues in the Homeric epics’, in I. Worthington (ed.), Voice into Text: Orality and Literacy in Ancient Greece, Leiden: Brill, pp. 3–20. Minchin, E. (2001a), Homer and the Resources of Memory: Some Applications of Cognitive Theory to the Iliad and the Odyssey, Oxford: Oxford University Press. Minchin, E. (2001b), ‘How Homeric is hysteron proteron?’, Mnemosyne 54: 635–45. Minchin, E. (2001c), ‘Similes in Homer: Image, mind’s eye, and memory’, in J. Watson (ed.), Speaking Volumes: Orality and Literacy in the Greek and Roman World, Leiden: Brill, pp. 25–52. Minchin, E. (2002), ‘Speech acts in the everyday world and in Homer: The rebuke as a case study’, in I. Worthington and J. M. Foley (eds), Epea and Grammata: Oral and Written Communication in Ancient Greece, Leiden: Brill, pp. 71–97. Minchin, E. (2005), ‘Homer on autobiographical memory: The case of Nestor’, in R. Rabel (ed.), Approaches to Homer: Ancient and Modern, Swansea: Classical Press of Wales, pp. 55–72. Minchin, E. (2007), Homeric Voices: Discourse, Memory, Gender, Oxford: Oxford University Press.



bibliogra p h y 269

Minchin, E. (2008), ‘Spatial memory and the composition of the Iliad’, in E. A. Mackay (ed.), Orality, Literacy, Memory in the Ancient Greek and Roman World, Leiden: Brill, pp. 9–34. 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. (2014), ‘Poet, audience, time, and text: Reflections on medium and mode in Homer and Vergil’, in R. Scodel (ed.), Between Orality and Literacy: Communication and Adaptation in Antiquity, Leiden: Brill, pp. 267–88. Minchin, E. (2016a), ‘Heritage in the landscape: The “heroic tumuli” in the Troad region’, in J. McInerney and I. Sluiter (eds), Valuing Landscape in Classical Antiquity: Natural Environment and Cultural Imagination, Leiden: Brill, pp. 255–75. Minchin, E. (2016b), ‘Voice and voices: Homer and the stewardship of memory’, in N. W. Slater (ed.), Voice and Voices in Antiquity: Orality and Literacy in the Ancient World, Leiden: Brill, pp. 11–30. Mingucci, G. (2015), La fisiologia del pensiero in Aristotele, Bologna: Il Mulino. Minsky, M. (1986), Society of Mind, New York: Simon & Schuster. Mistry, J. and B. Rogoff (1994), ‘Remembering in cultural context’, in W. Lonner and R. Malpass (eds), Psychology and Culture, Boston: Allyn & Bacon, pp. 139–44. Mitchell, S. (1990), ‘Festivals, games, and civic life in Roman Asia Minor’, Journal of Roman Studies 80: 183–93. Mitchell, S. (1993), Anatolia: Land, Men, and Gods in Asia Minor, vol. 1, Oxford: Oxford University Press. Mittenhuber, F. (2009), ‘The tradition of texts and maps in Ptolemy’s Geography’, in A. Jones (ed.), Ptolemy in Perspective: Use and Criticism of his Work from Antiquity to the Nineteenth Century, Dordrecht and New York: Springer, pp. 95–120. Moretti, F. (2000), ‘Conjectures on world literature’, New Left Review 1: 54–68. Moretti, F. (2013), Distant Reading, London: Verso. Morgan, T. (2015), Roman Faith and Christian Faith: Pistis and Fides in the Early Roman Empire and Early Churches, Oxford: Oxford University Press. Mori, M. (1970), ‘The uncanny valley’, Energy 7.4: 33–5. Most, G. W. (1996), ‘Reading Raphael: “The School of Athens” and its pre-text’, Critical Inquiry 23: 145–82. Murray, A. (1978), Reason and Society in the Middle Ages, Oxford: Oxford University Press. Nails, D. (2002), The People of Plato: A Prosopography of Plato and Other Socratics, Indianapolis: Hackett. Napier, A. D. (1986), Masks, Transformation, and Paradox, Irvine: University of California Press. Nathan, M. (2008), ‘An embodied cognition perspective on symbols, gesture, and grounding instruction’, in M. de Vega, A. Glenberg and A. Graesser (eds), Symbols and Embodiment: Debates on Meaning and Cognition, Oxford: Oxford University Press, pp. 375–96.

270

bibliograp h y

Netz, R. (2002), ‘Counter culture: Towards a history of Greek numeracy’, History of Science 40.3: 321–52. Netz, R. and M. Squire (2016), ‘Sight and the perspectives of mathematics: The limits of ancient optics’, in M. Squire (ed.), Sight and the Ancient Senses, Abingdon: Routledge, pp. 68–84. Newby, Z. (2003), ‘Art and identity in Asia Minor’, in S. Scott and J. Webster (eds), Roman Imperialism and Provincial Art, Cambridge: Cambridge University Press, pp. 192–213. Newby, Z. (2005), Greek Athletics in the Roman World: Victory and Virtue, Oxford: Oxford University Press. Ng, D. Y. (2015), ‘Commemoration and élite benefaction of buildings and spectacles in the Roman world’, Journal of Roman Studies 105: 101–23. Ng, D. Y. (2016), ‘Making an artful case: Public sculptural programs as instruments of civic rivalry in imperial Perge and Pamphylia’, Istanbuler Mitteilungen 66: 225–55. Ng, D. Y. (2018), ‘The Salutaris foundation: Monumentality through periodic rehearsal’, in B. Longfellow and E. Perry (eds), Roman Artists, Patrons and Public Consumption: Familiar Works Reconsidered, Ann Arbor: University of Michigan Press, pp. 63–87. Ng, D. Y. (forthcoming), ‘Thinking with statues: The Roman public portrait and the cognition of commemoration’, in P. Meineck, W. Short and J. Devereaux (eds), The Routledge Handbook of Classics and Cognitive Theory, New York: Routledge. Noë, A. (2004), Action in Perception, Cambridge, MA: MIT Press. Noë, A. (2009), Out of Our Heads: Why You Are Not Your Brain, and Other Lessons from the Biology of Consciousness, New York: Macmillan. Noë, A. (2015), Strange Tools: Art and Human Nature, New York: Hill and Wang. Novack, M. A. and S. Goldin-Meadow (2016), ‘Gesture as representational action: A paper about function’, Psychonomic Bulletin and Review 24.3: 652–65; available at (last accessed 7 February 2018). Novick, L. and S. Hurley (2001), ‘To matrix, network, or hierarchy: That is the question’, Cognitive Psychology 42: 158–216. Nünlist, R. (2009), The Ancient Critic at Work: Terms and Concepts of Literary Criticism in Greek Scholia, Cambridge: Cambridge University Press. Nussbaum, M. C. (1994), The Therapy of Desire: Theory and Practice in Hellenistic Ethics, Princeton: Princeton University Press. Nussbaum, M. C. (1996), ‘Aristotle on emotions and rational persuasion’, in A. O. Rorty (ed.), Essays on Aristotle’s Rhetoric, Berkeley: University of California Press, pp. 303–23. Nussbaum, M. C. (2001), Upheavals of Thought: The Intelligence of Emotions, Cambridge: Cambridge University Press. Nussbaum, M. C. (2010), Not for Profit: Why Democracy Needs the Humanities, Princeton: Princeton University Press.



bibliogra p h y 271

Nusselder, A. (2009), Interface Fantasy: A Lacanian Cyborg Ontology, Cambridge, MA: MIT Press. Nutting, H. (1926), ‘Cicero and the younger Pliny’, Classical Journal 21.6: 420–30. O’Brien Moore, A. (1935), ‘Senatus’, Paulys Realencyclopädie der classischen Altertumswissenschaft, Suppl. VI, Stuttgart: Metzler, coll. 660–800. O’Regan, J. K. and A. Noë (2001a), ‘A sensorimotor account of vision and visual consciousness’, Behavioral and Brain Sciences 24: 939–1031. O’Regan, J. K. and A. Noë (2001b), ‘What it is like to see: A sensorimotor theory of perceptual experience’, Synthese 129: 79–103. Ober, J. (1989), Mass and Elite in Democratic Athens: Rhetoric, Ideology, and the Power of the People, Princeton: Princeton University Press. Ober, J. (1999), Political Dissent in Democratic Athens: Intellectual Critics of Popular Rule, Princeton: Princeton University Press. Ober, J. (2008), Democracy and Knowledge: Innovation and Learning in Classical Athens, Princeton: Princeton University Press. Öhman, A. and A. Flykt (2000), ‘Unconscious emotion: Evolutionary perspectives, psychophysiological data, and neuropsychological mechanisms’, in R. D. Lane and L. Nadel (eds), Cognitive Neuroscience of Emotion, Oxford: Oxford University Press, pp. 296–327. Öhman, A., A. Flykt and F. Esteves (2001), ‘Emotion drives attention: Detecting the snake in the grass’, Journal of Experimental Psychology: General 130: 466–78. Olick, J. (2008), ‘From collective memory to the sociology of mnemonic practices and products’, in A. Erll and A. Nünning (eds), Cultural Memory Studies, Berlin: De Gruyter, pp. 151–61. Olson, D. (1988), ‘Mind, media, and memory: The archival and epistemic functions of written text’, in D. de Kerckhove and C. Lumsden (eds), The Alphabet and the Brain, Berlin: Springer, pp. 422–41. Olson, D. (1996), The World on Paper: The Conceptual and Cognitive Implications of Writing, Cambridge: Cambridge University Press. Olson, D. and M. Cole (eds) (2006), Technology, Literacy, and the Evolution of Society, Mahwah, NJ: Erlbaum. Ong, W. (1982), Orality and Literacy: The Technologizing of the Word, New York: Methuen. Onians, J. (1999), Classical Art and the Cultures of Greece and Rome, New Haven and London: Yale University Press. Ortolano, G. (2005), ‘F. R. Leavis, science, and the abiding crisis of modern civilisation’, History of Science 43: 161–85. Osbeck, L. and N. Nersessian (2014), ‘Situating distributed cognition’, Philosophical Psychology 27.1: 82–97. Osborne, C. (2009), ‘Selves and other selves in Aristotle’s Eudemian Ethics vii 12’, Ancient Philosophy 29: 1–23. Osborne, R. (1999), ‘Inscribing performance’, in S. Goldhill and R. Osborne (eds), Performance Culture and Athenian Democracy, Cambridge: Cambridge University Press, pp. 341–58.

272

bibliograp h y

Osborne, R. and P. J. Rhodes (2003),  Greek Historical Inscriptions 404–323 bc, Oxford:  Oxford University Press. Otto, N. (2009), Enargeia: Untersuchung zur Charakteristik alexandrinischer Dichtung, Stuttgart: Steiner. Overwien, O. (2014), Hippokrates: Über die Säfte, Berlin: De Gruyter. Pace, C. (2008), ‘ “Peace and tranquillity of mind”: The theme of retreat and Poussin’s landscapes’, in P. Rosenberg and K. Christiansen (eds), Poussin and Nature: Arcadian Visions, New York: Metropolitan Museum of Art, pp. 73–90. Pagán Cánovas, C. (2010), ‘Erotic emissions in Greek poetry: A generic integration network’, Cognitive Semiotics 6: 7–32. Pagán Cánovas, C. (2011), ‘The genesis of the arrows of love: Diachronic conceptual integration in Greek mythology’, American Journal of Philology 132: 553–79. Pagán Cánovas, C. (2014), ‘Cognitive patterns in Greek poetic metaphors of emotion: A diachronic approach’, in J. E. Díaz Vera (ed.), Metaphor and Metonymy through Time and Cultures, Berlin: De Gruyter, pp. 295–18. Pagán Cánovas, C. (2016), ‘Rethinking image schemas: Containment and emotion in Greek poetry’, Journal of Literary Semantics 45: 117–39. Pagán Cánovas, C. and M. Antović (2016), ‘Formulaic creativity: Oral poetics and cognitive grammar’, Language and Communication 47: 66–74. Palmer, A. (2004), Fictional Minds, Lincoln: University of Nebraska Press. Palmer, A. (2010), Social Minds in the Novel, Columbus: Ohio State University Press. Palmer, G. (1996), Toward a Theory of Cultural Linguistics, Austin: University of Texas Press. Panksepp, J. (1998), Affective Neuroscience, Oxford: Oxford University Press. Panksepp, J. (2005), ‘Affective consciousness: Core emotional feelings in animals and humans’, Consciousness and Cognition 14: 30–80. Papastamati-von Moock, C. (2014), ‘The theatre of Dionysus Eleuthereus in Athens: New data and observations on its “Lycurgan” phase’, in E. Csapo, H. R. Goette and J. R. Green (eds), Greek Theatre in the Fourth Century bc, Berlin and Boston: De Gruyter, pp. 15–76. Parker, H. (1999), ‘Does Caesar mime?’, in B. Bergmann and C. Kondoleon (eds), The Art of Ancient Spectacle, New Haven: Yale University Press, pp. 163–80. Parker, R. C. T. (1996), Athenian Religion: A History, Oxford: Oxford University Press. Parvez, K., O. Stewart, S. Sangha and K. Lukowiak (2005), ‘Boosting intermediate-term into long-term memory’, Journal of Experimental Biology 208.8: 1525–36. Parvizi, J., S. Anderson, C. Martin, H. Damasio and A. Damasio (2001), ‘Pathological laughter and crying: A link to the cerebellum’, Brain 124: 1708–19. Pearson, G. (2014), ‘Aristotle and the cognitive component of emotions’, Oxford Studies in Ancient Philosophy 46: 165–212. Pébarthe, C. (2006), Cité, démocratie et écriture. Histoire de l’alphabétisation d’Athènes à l’époque classique, Paris: De Boccard. Peelen, M. V. and P. E. Downing (2005), ‘Selectivity for the human body in the fusiform gyrus’, Journal of Neurophysiology 93.1: 603–8.



bibliogra p h y 273

Peters, C. H. F. and E. B. Knobel (1915), Ptolemy’s Catalogue of Stars: A Revision of the Almagest, Washington, DC: Carnegie Institution of Washington. Peterson, L. and M. Peterson (1959), ‘Short-term retention of individual verbal items’, Journal of Experimental Psychology 58: 193–8. Pettit, P. (2003), ‘Groups with minds of their own’, in F. F. Schmitt (ed.), Socializing Metaphysics: The Nature of Social Reality, Lanham, MD: Rowman & Littlefield, pp. 167–93. Phillipson, T. (2017), ‘Calculators: Latest gadget to obsolete relic’, History of Distributed Cognition Project, (last accessed 10 January 2018). Pickard-Cambridge, A. W. (1968), The Dramatic Festivals of Athens, revised by J. Gould and D. M. Lewis, Oxford: Oxford University Press. Pickard-Cambridge, A. W., J. Gould and D. M. Lewis (1968), The Dramatic Festivals of Athens, rev. edn, Oxford: Oxford University Press. Pigeaud, J. (1981), La Maladie de l’âme: étude sur la relation de l’âme et du corps dans la tradition médico-philosophique antique, Paris: Les Belles Lettres. Pigeaud, J. (1991), ‘Qu’est-ce que tranquilliser?’, Psychiatrie Française 1: 15–24. Piggin, J.-P. (2015), Notes on the History of the Stemma, (last accessed 29 May 2016). Plamper, J. (2015), The History of Emotions: An Introduction, Oxford: Oxford University Press. Pomeroy, A. J. (ed.) (1999), Arius Didymus: Epitome of Stoic Ethics, Atlanta: Society of Biblical Literature. Potter D. and D. Mattingly (1999), Life, Death, and Entertainment in the Roman Empire, Ann Arbor: University of Michigan Press. Preus, A. (1968), ‘On Dreams 2, 459b24–460a33, and Aristotle’s ὄψις’, Phronesis 13: 175– 82. Previc, F. H. (1998), ‘The neuropsychology of 3-D space’, Psychological Bulletin 124.2: 123–32. Previc, F. H. (2006), ‘The role of the extrapersonal brain systems in religious activity’, Consciousness and Cognition 15: 510–28. Price, A. (1989), Love and Friendship in Plato and Aristotle, Oxford: Oxford University Press. Price, A. (2010), ‘Emotions in Plato and Aristotle’, in P. Goldie (ed.), The Oxford Handbook of Philosophy of Emotion, Oxford: Oxford University Press, pp. 121–42. Price, S. (1999), Religions of the Ancient Greeks, Cambridge: Cambridge University Press. Price, S. (2005), ‘Local mythologies in the Greek East’, in C. Howgego and A. Burnett (eds), Coinage and Identity in the Roman Provinces, Oxford: Oxford University Press, pp. 115–24. Prinz, J. (2002), ‘Consciousness, computation and emotion’, in S. C. Moore and M. Oaksford (eds), Emotional Brain: From Brain to Behaviour, Amsterdam and Philadelphia: John Benjamins, pp. 137–55.

274

bibliograp h y

Prinz, J. (2004a), ‘Embodied emotions’, in R. C. Solomon (ed.), Thinking About Feeling, Oxford: Oxford University Press, pp. 44–60. Prinz, J. (2004b), Gut Reactions: A Perceptual Theory of Emotion, Oxford: Oxford University Press. Pulvermüller, F. (2005), ‘Brain mechanisms linking language and action’, Nature Reviews Neuroscience 6: 576–82. Rabinowitz, M. and J. Mandler (1983), ‘Organisation and information retrieval’, Journal of Experimental Psychology: Learning, Memory and Cognition 9: 430–9. Rackham, H. (trans.) (1991), Eudemian Ethics, in Aristotle, vol. 20, Cambridge, MA: Harvard University Press (Loeb Classical Library). Radden, J. (2009), Moody Minds Distempered: Essays on Melancholy and Depression, Oxford and New York: Oxford University Press. Ramachandran, V. S. and D. R. Ramachandran (1996), ‘Synaesthesia in phantom limbs induced with mirrors’, Proceedings: Biological Sciences 263: 377–86. Rauff, J. (2003), ‘Counting on your body in Papua New Guinea’, Mathematical Connections 2.2: 20–41. Rawson, E. (1987), ‘Discrimina ordinum: The lex Julia theatralis’, Papers of the British School in Rome 55: 83–114. Reeve, M. (1996), ‘Classical scholarship’, in J. Kraye (ed.), Cambridge Companion to Renaissance Humanism, Cambridge: Cambridge University Press, pp. 20–46. Reicher, S. (2001), ‘The psychology of crowd dynamics’, in M. A. Hogg and R. S. Tindale (eds), Blackwell Handbook of Social Psychology: Group Process, Malden, MA: Blackwell, pp. 182–208. Reicher, S. and S. A. Haslam (2006), ‘Rethinking the psychology of tyranny: The BBC prison study’, British Journal of Social Psychology 45: 1–40. Reinach, T. (1888), ‘L’inscription de Lygdamis’, REG 1: 27–48. Revermann, M. (2006a), Comic Business: Theatricality, Dramatic Technique, and Performance Contexts of Aristophanic Comedy, Oxford: Oxford University Press. Revermann, M. (2006b), ‘The competence of theatre audiences in fifth- and fourth-century Athens’, The Journal of Hellenic Studies 126: 99–124. Reydams-Schils, G. (2005), The Roman Stoics: Self, Responsibility, and Affection, Chicago: University of Chicago Press. Reyes, H. (2009), ‘The rhetorical frame of Poussin’s theory of the modes’, Intellectual History Review 19.3: 287–302. Reynolds, D. and M. Reason (2012), Kinesthetic Empathy in Creative and Cultural Practices, Bristol and Chicago: Intellect Books. Reynolds, J. (1982), Aphrodisias and Rome, Documents from the Excavation of the Theatre at Aphrodisias Conducted by Professor Kenan T. Erim, Together with Some Related Texts, London: Society for the Promotion of Roman Studies. Reynolds. J. (1991), ‘Epigraphic evidence for the construction of the theatre: 1st c. bc to mid 3rd c. ad’, in R. R. R. Smith and K. T. Erim (eds), Aphrodisias Papers 2, Journal of Roman Archaeology Supplement 2, Ann Arbor: University of Michigan Press, pp. 15–28.



bibliogra p h y 275

Rhodes, P. J. (1972), The Athenian Boule, Oxford: Oxford University Press. Richardson, D. C., M. J. Spivey, L. W. Barsalou and K. McRae (2003), ‘Spatial representations activated during real-time comprehension of verbs’, Cognitive Science 27: 767–80. Riggsby, A. (forthcoming), Mosaics of Knowledge: Representing Information in the Roman World, Oxford: Oxford University Press. Ritti, T. (1983), ‘Epigrafi dedicatorie imperiali di Hierapolis di Frigia’, Rendiconti atii della Academia Nazionale die Lincei 38: 171–82. Ritti, T. (1985), Hierapolis I: Fonti letterarie ed ephigrafiche, Rome: Bretschneider Giorgio. Rizzolatti, G. and L. Craighero (2004), ‘The mirror-neuron system’, Annual Review of Neuroscience 27: 169–92. Rizzolatti, G. and C. Sinigaglia (2008), Mirrors in the Brain: How our Minds Share Actions and Emotions, trans. Frances Anderson, Oxford: Oxford University Press. Robinson, J. (1995), ‘Startle’, Journal of Philosophy 92: 53–74. Robinson, J. (2010), ‘Emotional responses to music: What are they? How do they work? And are they relevant to aesthetic appreciation?’, in P. Goldie (ed.), The Oxford Handbook of Philosophy of Emotion, Oxford: Oxford University Press, pp. 651–80. Robson, E. (2003), ‘Tables and tabular formatting in Sumer, Babylonia, and Assyria, 2500 bce–50 ce’, in M. Campbell-Kelly, M. Croarken, R. G. Flood and E. Robson (eds), The History of Mathematical Tables from Sumer to Spreadsheets, Oxford: Oxford University Press, pp. 18–47. Robson, E. (2004), ‘Accounting for change: The development of tabular book-keeping in early Mesopotamia’, in M. Hudson and C. Wunsch (eds), Creating Economic Order: Record-keeping, Standardization, and the Development of Accounting in the Ancient Near East, Bethesda: CDL Press, pp. 107–44. Rocca, J. (2003), Galen on the Brain: Anatomical Knowledge and Physiological Speculation in the Second Century ad, Leiden and Boston: Brill. Roger, G. (1991), The Sacred Identity of Ephesus, London: Routledge. Rogers, Y. (1997), ‘A brief introduction to distributed cognition’, (last accessed 7 February 2018). Rogoff, B. and J. Mistry (1985), ‘Memory development in cultural context’, in M. Pressley and C. Brainerd (eds), The Cognitive Side of Memory Development, New York: Springer, pp. 117–42. Rogoff, B. and K. Waddell (1982), ‘Memory for information organized in a scene by children from two cultures’, Child Development 53: 1224–8. Rood, T. (2004), ‘Xenophon and Diodorus: Continuing Thucydides’, in C. Tuplin (ed.), Xenophon and his World, Stuttgart: Steiner, pp. 341–95. Rosenzweig, M., E. Bennett, P. Colombo, D. Lee and P. Serrano (1993), ‘Shortterm, intermediate-term, and long-term memories’, Behavioural Brain Research 57.2: 193–8.

276

bibliograp h y

Ross, W. D. (trans.) (1925), Aristotle: Nicomachean Ethics, in The Works of Aristotle, vol. 9, ed. W. D. Ross, Oxford: Oxford University Press. Rowlands, M. (1999), The Body in Mind, Cambridge: Cambridge University Press. Rowlands, M. (2003), Externalism: Putting Mind and World Back Together Again, London: Acumen. Rowlands, M. (2009), ‘Enactivism and the extended mind’, Topoi 28.1: 53–62. Rowlands, M. (2010), The New Science of the Mind: From Extended Mind to Embodied Phenomenology, Cambridge, MA: MIT Press. Rubin, D. (1995), Memory in Oral Traditions, Oxford: Oxford University Press. Rubin, D. and A. Wenzel (1996), ‘One hundred years of forgetting: A quantitative description of retention’, Psychological Review 103.4: 743–60. Rudder Baker, L. (2011), ‘Beyond the Cartesian self’, Phenomenology and Mind 1: 48–57. Rudolph, K. (2016), ‘Sight and the Presocratics: Approaches to visual perception in early Greek philosophy’, in M. Squire (ed.), Sight and the Ancient Senses, Abingdon: Routledge, pp. 36–53. Rumelhart, D. E., P. Smolensky, J. L. McClelland and G. E. Hinton (1986), ‘Schemata and sequential thought processes in PDP models’, in J. L. McClelland, D. E. Rumelhart and PDP Group (eds), Parallel Distributed Processing: Explorations in the Microstructure of Cognition, vol. 2, Cambridge, MA: MIT Press, pp. 7–57. Rupert, R. (2009), Cognitive Systems and the Extended Mind, Oxford: Oxford University Press. Rupert, R. (2014), ‘Against group cognitive states’, in S. Chant, F. Hindriks and G. Preyer (eds), From Individual to Collective Intentionality: New Essays, Oxford: Oxford University Press, pp. 97–111. Rüpke, J. (2001), Die Religion der Römer: Eine Einführung, Munich: Beck. Rüpke, J. (2011), The Roman Calendar from Numa to Constantine: Time, History, and the Fasti, Oxford: Wiley-Blackwell. Russell, D. A. (ed.) (1964), ‘Longinus’: On the Sublime, Oxford: Oxford University Press. Russell, D. A. (ed.) (2001), Quintilian: Institutio Oratoria, 4 vols, Cambridge, MA: Harvard University Press (Loeb Classical Library). Ryan, F. (1998), Rank and Participation in the Roman Senate, Stuttgart: Steiner. Ryan, M.-L. (2001), Narrative as Virtual Reality: Immersion and Interactivity in Literature and Electronic Media, Baltimore: Johns Hopkins University Press. Ryle, G. (1949), The Concept of Mind, New York: Barnes and Noble. Şahin, S. (2000), ‘Bauhistorisch-epigraphische Zeugnisse aus dem Theater von Perge’, Olba: Publications of the Research Center of Cilician Archaeology 3: 199–209. Sanford, A. J. and C. Emmott (2012), Mind, Brain and Narrative, Cambridge: Cambridge University Press. Sarbin, T. (1986), Narrative Psychology, Westport, CT: Praeger. Saxe, G. and I. Esmonde (2012), Cultural Development of Mathematical Ideas, Cambridge: Cambridge University Press.



bibliogra p h y 277

Scarry, E. (1995), ‘On vivacity: The difference between daydreaming and imagining-under-authorial-instruction’, Representations 52: 1–26. Scarry, E. (1999), Dreaming by the Book, New York: Farrar, Straus and Giroux. Schacter, D. (1996), Searching for Memory: The Brain, the Mind, and the Past, New York: Basic Books. Schank, R. C. and R. P. Abelson (1977), Scripts, Plans, Goals, and Understanding: An Inquiry into Human Knowledge Structures, Hillsdale, NJ: Erlbaum. Schärlig, A. (2001), Compter avec des cailloux, Lausanne: Presses polytechniques et universitaires romandes. Scherer, K. R. (1993), ‘Studying the emotion-antecedent appraisal process: An expert system approach’, Cognition and Emotion 7: 325–55. Scherer, K. R. and P. C. Ellsworth (2009), ‘Appraisal theories’, in D. Sander and K. R. Scherer (eds), The Oxford Companion to Emotion and the Affective Sciences, Oxford: Oxford University Press, pp. 45–9. Schiefsky, M. (2012), ‘Galen and the tripartite soul’, in R. Barney, T. Brennan and C. Brittain (eds), Plato and the Divided Self, Cambridge: Cambridge University Press, pp. 331–49. Schlimm, D. and H. Neth (2008), ‘Modeling ancient and modern arithmetic practices’, in V. Sloutsky, B. Love and K. McRae (eds), Proceedings of the 30th Annual Meeting of the Cognitive Science Society, Austin: Cognitive Science Society, pp. 2097–102. Schneider, W. and M. Detweiler (1987), ‘A connectionist/control architecture for working memory’, in G. Bower (ed.), The Psychology of Learning and Motivation, vol. 21, New York: Academic Press, pp. 53–119. Schütrumpf, E. (2015), ‘Black bile as the cause of human accomplishments and behaviors in Pr. 30.1: Is the concept Aristotelian?’, in R. Mayhew (ed.), The Aristotelian Problemata Physica, Leiden: Brill, pp. 357–79. Scodel, R. (2009), ‘Ignorant narrators in Greek tragedy’, in J. Grethlein and A. Rengakos (eds), Narratology and Interpretation, Berlin: De Gruyter, pp. 421–47. Scodel, R. (2012), ‘ἦ and Homeric Theory of Mind’, in M. Meier-Brügger (ed.), Homer, gedeutet durch ein großes Lexikon, Berlin: De Gruyter, pp. 319–34. Scodel, R. (ed.) (2014a), Between Orality and Literacy: Communication and Adaptation in Antiquity, Orality and Literacy in the Ancient World 10, Leiden: Brill. Scodel, R. (2014b), ‘Narrative focus and elusive thought in Homer’, in D. Cairns and R. Scodel (eds), Defining Greek Narrative, Edinburgh: Edinburgh University Press, pp. 55–74. Scodel, R. (2015), ‘Sunt lacrimae rerum’, Classical Journal 111: 211–39. Scullard, H. (1973), Roman Politics, Oxford: Oxford University Press. Seal, C. B. (2009), ‘Philosophy and Community in Seneca’s Prose’, unpublished PhD dissertation, Princeton University. Sear, F. (2006), Roman Theatres: An Architectural Study, Oxford: Oxford University Press. Sedley, D. N. (1999), ‘The ideal of godlikeness’, in G. Fine (ed.), Plato, vol. 2, Ethics, Politics, Religion, and the Soul, Oxford: Oxford University Press, pp. 309–28.

278

bibliograp h y

Sedley, D. N. (2007), Creationism and its Critics in Antiquity, Berkeley and Los Angeles: University of California Press. Semin, G., M. Garrido and T. Palma (2013), ‘Interfacing body, mind, the physical, and the social world: Socially situated cognition’, in D. Carlston (ed.), The Oxford Handbook of Social Cognition, Oxford: Oxford University Press, pp. 637–55. Shammi, P. and D. T. Stuss (1999), ‘Humour appreciation: A role of the right frontal lobe’, Brain 122: 657–66. Shapiro, L. (2010), Embodied Cognition, London and New York: Routledge. Shapiro, L. (ed.) (2014), The Routledge Handbook of Embodied Cognition, London and New York: Routledge. Sharples, R. W. (2010), Peripatetic Philosophy 200 bc to ad 200, Cambridge: Cambridge University Press. Sheppard, A. (2014), The Poetics of Phantasia: Imagination in Ancient Aesthetics, London: Bloomsbury. Shore, B. (1996), Culture in Mind, Oxford: Oxford University Press. Short, W. M. (2008), ‘Thinking places, placing thoughts: Spatial metaphors of mental activity in Roman culture’, Quaderni del ramo d’oro 1: 106–29. Short, W. M. (2012), ‘A Roman folk model of the mind’, Arethusa 45: 109–47. Short, W. M. (2013a), ‘Getting to the truth: Metaphors of mistakenness in Greek and Latin’, Arion 21: 139–68. Short, W. M. (2013b), ‘Latin de: A view from cognitive semantics’, Classical Antiquity 32: 378–405. Short, W. M. (2013c), ‘ “Transmission” accomplished? Latin’s alimentary metaphors of communication’, American Journal of Philology 134: 247–75. Short, W. M. (2014), ‘Metafora’, in M. Bettini and W. M. Short (eds), Con i romani: un’antropologia della cultura antica, Bologna: Il Mulino, pp. 329–52. Short, W. M. (ed.) (2016a), Embodiment in Latin Semantics, Amsterdam: John Benjamins. Short, W. M. (2016b), ‘Spatial metaphors of time in Roman culture’, Classical World 109: 381–412. Singer, P. (1992), ‘Some Hippocratic mind-body problems’, in J. A. López Férez (ed.), Tratados hipocráticos: Actas del VIIe colloque international hippocratique, Madrid: Universidad Nacional de Educación a Distancia, pp. 131–43. Slaby, J. (2014), ‘Emotions and the extended mind’, in C. von Scheve and M. Salmela (eds), Collective Emotions, Oxford: Oxford University Press, pp. 32–46. Slater, N. W. (ed.) (2016), Voice and Voices in Antiquity, Orality and Literacy in the Ancient World 11, Leiden: Brill. Small, H. (2013), The Value of the Humanities, Oxford: Oxford University Press. Small, J. P. (1997), Wax Tablets of the Mind: Cognitive Studies of Memory and Literacy in Classical Antiquity, London and New York: Routledge. Smith, A. M. (1988), ‘The psychology of visual perception in Ptolemy’s Optics’, Isis 79.2: 188–207. Smith, A. M. (1999), Ptolemy and the Foundations of Ancient Mathematical Optics: A Source Based Guided Study, Philadelphia: American Philosophical Society.



bibliogra p h y 279

Smith, C. and P. C. Ellsworth (1985), ‘Patterns of cognitive appraisal in emotion’, Journal of Personality and Social Psychology 48: 813–38. Smith, C. F. (ed.) (1923), Thucydides: History of the Peloponnesian War, vol. 4, Cambridge, MA: Harvard University Press. Smith, P. and O. R. Jones (1986), The Philosophy of Mind: An Introduction, Cambridge: Cambridge University Press. Smith, R. R. R. and K. T. Erim (1991), ‘Sculpture from the theater: A preliminary report’, in R. R. R. Smith and K. T. Erim (eds), Aphrodisias Papers 2, Journal of Roman Archaeology Supplement 2, Ann Arbor: University of Michigan Press, pp. 67–98. Smith, R. R. R., with S. Dillon, C. H. Hallett, J. Lenaghan and J. Van Voorhis (2006), Aphrodisias II: Roman Portrait Statuary from Aphrodisias, Mainz: Philipp von Zabern. Smith, W. D. (ed.) (1994), Hippocrates: Vol. VII, Cambridge, MA: Harvard University Press (Loeb Classical Library). Snow, C. P. [1959] (1993), The Two Cultures, ed. S. Collini, Cambridge: Cambridge University Press. Solomon, R. C. (1976), The Passions: Emotions and the Meaning of Life, Garden City, NY: Anchor Books. Sommerstein, A. (ed.) (2009), Aeschylus, vol. 2, Cambridge, MA: Harvard University Press. Sorabji, R. (1992), ‘Intentionality and physiological processes: Aristotle’s theory of sense-perception’, in M. C. Nussbaum and A. O. Rorty (eds), Essays on Aristotle’s De Anima, Oxford: Oxford University Press, pp. 192–226. Sorabji, R. (1993), Animal Minds and Human Morals: The Origins of the Western Debate, London: Duckworth. Speck, O., T. Ernst, J. Braun, C. Koch, E. Miller and L. Chang (2000), ‘Gender differences in the functional organization of the brain for working memory’, NeuroReport 11.11: 2581–5. Spivey, M. J. (2007), The Continuity of Mind, Oxford: Oxford University Press. Spivey, M. J. and J. J. Geng (2001), ‘Oculomotor mechanisms activated by imagery and memory: Eye movements to absent objects’, Psychological Research 65: 235–41. Spolsky, E. (1993), Gaps in Nature: Literary Interpretation and the Modular Mind, Albany: State University of New York Press. Spolsky, E. (2015), The Contracts of Fiction: Cognition, Culture, Community, Oxford and New York: Oxford University Press. Sprevak, M. (2009), ‘Extended cognition and functionalism’, The Journal of Philosophy 106.9: 503–27. Sprevak, M. (2011), ‘Neural sufficiency, reductionism, and cognitive neuropsychiatry’, Philosophy, Psychiatry and Psychology 18.4: 339–44. Squire, M. (2013), ‘Ekphrasis at the forge and the forging of ekphrasis: The shield of Achilles in Graeco-Roman word and image’, Word and Image 29: 157–91. Squire, M. (2016), ‘Introductory reflections: Making sense of ancient sight’, in

280

bibliograp h y

M. Squire (ed.), Sight and the Ancient Senses, London and New York: Routledge, pp. 1–35. Stanfield, R. A. and R. A. Zwaan (2001), ‘The effect of implied orientation derived from verbal context and picture recognition’, Psychological Science 12: 153–6. Steel, C. (2001), ‘The moral purpose of the human body: A reading of Timaeus 69–72’, Phronesis 46: 105–28. Sterelny, K. (2010), ‘Minds: Extended or scaffolded?’, Phenomenology and the Cognitive Sciences 9: 465–81. Stern-Gillet, S. (1995), Aristotle’s Philosophy of Friendship, Albany: State University of New York Press. Stewart, J., O. Gapenne and E. A. Di Paolo (eds) (2010), Enaction: Toward a New Paradigm for Cognitive Science, Cambridge, MA: MIT Press. Stewart, P. (2003), Statues in Roman Society, Oxford: Oxford University Press. Stock, St. G. (trans.) (1925), Aristotle: Magna moralia, in The Works of Aristotle, vol. 9, ed. W. D. Ross, Oxford: Oxford University Press. Straehley, I. C. and J. L. Loebach (2014), ‘The influence of mode and musical experience on the attribution of emotions to melodic sequences’, Psychomusicology: Music, Mind, and Brain 24.1: 21–34. Striker, G. (1996), ‘Emotions in context: Aristotle’s treatment of the passions in the Rhetoric and his moral psychology’, in A. O. Rorty (ed.), Essays on Aristotle’s Rhetoric, Berkeley: University of California Press, pp. 286–302. Susi, T. (2011), ‘Embodied interaction, coordination and reasoning in computer gameplay’, in L. Shapiro (ed.), The Routledge Handbook of Embodied Cognition, London and New York: Routledge, pp. 184–93. Sutton, D. F. (1985), ‘Named choreuts in satyr plays’, The American Journal of Philology 106: 107–10. Sutton, J. (2007a), ‘Material agency, skills, and history: Distributed cognition and the archaeology of memory’, in K. Knappett and L. Malafouris (eds), Material Agency: Towards a Non-Anthropocentric Approach, Berlin: Springer, pp. 37–55. Sutton, J. (2007b), ‘Spongy brains and material memories’, in M. Floyd-Wilson and G. A. Sullivan Jr (eds), Environment and Embodiment in Early Modern England, New York: Palgrave Macmillan, pp. 14–34. Sutton, J. (2010), ‘Exograms and interdisciplinarity: History, the extended mind, and the civilizing process’, in R. Menary (ed.), The Extended Mind, Cambridge, MA: MIT Press, pp. 189–225. Sutton, J., C. Harris, P. Keil and A. Barnier (2010), ‘The psychology of memory, extended cognition, and socially distributed remembering’, Phenomenology and the Cognitive Sciences 9.4: 521–60. Taisbak, C. (1965), ‘Roman numerals and the abacus’, Classica et Mediaevalia 26: 153–60. Talbert, R. J. A. (2008), ‘Greek and Roman mapping: Twenty-first century perspectives’, in R.  J.  A. Talbert and R.  W. Unger (eds), Cartography in Antiquity and the Middle Ages: Fresh Perspectives, New Methods, Leiden and Boston: Brill, pp. 9–27.



bibliogra p h y 281

Talbert, R. J. A. and K. Brodersen (eds) (2004), Space in the Roman World: Its Perception and Presentation, Antike Kultur und Geschichte 5, Münster and Piscataway, NJ: Transaction. Tallis, R. (2011), Aping Mankind: Neuromania, Darwinitis and the Misrepresentation of Humanity, Durham: Acumen. Tanner, J. (2000), ‘Portraits, power, and patronage in the late Roman republic’, Journal of Roman Studies 90: 18–50. Taplin, O. (2001), ‘Masks in Greek drama and in Tantalus’, Didaskalia 5.2, (last accessed 7 February 2018). Taplin, O. (2007), Pots and Plays: Interactions between Tragedy and Greek Vase-Painting of the Fourth Century bc, Los Angeles: Getty Publications. Taplin, O. and R. Wyles (eds) (2010), The Pronomos Vase and its Context, Oxford: Oxford University Press. Taub, L. C. (1993), Ptolemy’s Universe: The Natural Philosophical and Ethical Foundations of Ptolemy’s Astronomy, Chicago: Open Court. Taub, L. C. (2002), ‘Instruments of Alexandrian astronomy: The uses of the equinoctial rings’, in C. Tuplin and T. E. Rihll (eds), Science and Mathematics in Ancient Greek Culture, Oxford: Oxford University Press, pp. 133–49. Taub, L. C. (2008), ‘Eratosthenes sends greetings to King Ptolemy’, in J. W. Dauben (ed.), Mathematics Celestial and Terrestrial: Festschrift für Menso Folkerts zum 65. Geburtstag, Halle and Stuttgart: Deutsche Akademie der Naturforscher Leopoldina/Wissenschaftliche Verlagsgesellschaft, pp. 285–302. Taylor, L. (1951), ‘New indications of Augustan editing in the Capitoline Fasti’, Classical Philology 46: 73–80. Taylor, L. and R. Scott (1969), ‘Seating space in the Roman senate and the Senatores Pedarii’, Transactions and Proceedings of the American Philological Association 100: 529–82. Tegmark, M. (2014), Our Mathematical Universe: My Quest for the Ultimate Nature of Reality, New York: Knopf. Thagard, P. and J. Nerb (2002), ‘Emotional gestalts: Appraisal, change, and the dynamics of affect’, Personality and Social Psychology Review 6: 274–82. Theiner, G. (2011), ‘Varieties of group cognition’, in L. Shapiro (ed.), The Routledge Handbook of Embodied Cognition, London and New York: Routledge, pp. 347–57. Theodorescu, D. (1996), ‘La frons scaenae du théâtre: innovations et particularités à l’époque de Zoilos’, in C. Roueché and R. R. R. Smith (eds), Aphrodisias Papers 3, Journal of Roman Archaeology Supplement 20, Ann Arbor: University of Michigan Press, pp. 127–48. Thomas, N. J. T. (1999), ‘Are theories of imagery theories of imagination? An active perception approach to conscious mental content’, Cognitive Science 23: 207–45. Thomas, R. (1992), Literacy and Orality in Ancient Greece, Cambridge: Cambridge University Press. Thomas, R. (1996), ‘Literacy and the city-state in archaic and classical Greece’,

282

bibliograp h y

in A. K. Bowman and G. Woolf (eds), Literacy and Power in the Ancient World, Cambridge: Cambridge University Press, pp. 33–50. Thomas, R. (2011), ‘ “And you, the demos, made an uproar”: Performance, mass audiences and text in Athenian democracy’, in A. P. M. H. Lardinois, J. H. Blok and M. G. M. van der Poel (eds), Sacred Words: Orality, Literacy and Religion, Leiden: Brill, pp. 161–87. Thompson, E. (2007a), ‘Look again: Phenomenology and mental imagery’, Phenomenology and the Cognitive Sciences 6: 137–70. Thompson, E. (2007b), Mind in Life: Biology, Phenomenology, and the Sciences of Mind, Cambridge, MA: Harvard University Press. Thompson, E. and M. Stapleton (2009), ‘Making sense of sense-making: Reflections on enactive and extended mind theories’, Topoi 28.1: 23–30. Thumiger, C. (2015), ‘Patient function and physician function in the Epidemics cases’, in G. Petridou and C. Thumiger (eds), Homo Patiens: Approaches to the Patient in the Ancient World, Leiden: Brill, pp. 107–37. Thumiger, C. (2016), ‘Grief and cheerfulness in early Greek medical writings’, in P. R. Bosman (ed.), Ancient Routes to Happiness, Pretoria: Classical Association of South Africa, pp. 95–116. Tieleman, T. (1996), Galen and Chrysippus on the Soul: Argument and Refutation in the De Placitis, Books II–III, Leiden and New York: Brill. Tieleman, T. (2003), Chrysippus’ On Affections: Reconstruction and Interpretation, Leiden: Brill. Tollefsen, D. P. (2006), ‘From extended mind to collective mind’, Cognitive Systems Research 7.2: 140–50. Tollefsen, D. P. (2015), Groups as Agents, Cambridge: Polity Press. Tomasello, M. (2014), A Natural History of Human Thinking, Cambridge, MA: Harvard University Press. Tomkins, S. S. (1962), Affect, Imagery, Consciousness, New York: Springer. Tribble, E. (2005), ‘Distributing cognition in the Globe’, Shakespeare Quarterly 56.2: 135–55. Tribble, E. (2011), Cognition in the Globe: Attention and Memory in Shakespeare’s Theatre, New York: Palgrave Macmillan. Tribble, E. and N. Keene (2011), Cognitive Ecologies and the History of Remembering: Religion, Education and Memory in Early Modern England, New York: Palgrave Macmillan. Tribble, E. and J. Sutton (2011), ‘Cognitive ecology as a framework for Shakespearean studies’, Shakespearean Studies 39: 94–103. Troscianko, E. T. (2013), ‘Reading imaginatively: The imagination in cognitive science and cognitive literary studies’, Journal of Literary Semantics 42: 181–98. Troscianko, E. T. (2014), Kafka’s Cognitive Realism, New York and London: Routledge. Tsur, R. (1992), Towards a Theory of Cognitive Poetics, Amsterdam: North-Holland. Tudeau-Clayton, M. (2008), Jonson, Shakespeare and Early Modern Virgil, Cambridge: Cambridge University Press.



bibliogra p h y 283

Tulving, E. and W. Donaldson (eds) (1972), Organisation of Memory, New York: Academic Press. Tümpel, K. (1894), ‘Mnemon’, in W. Roscher (ed.), Ausfürliches Lexicon der griechischen und römischen Mythologie, vol. 2, Leipzig: Teubner, coll. 3075–6. Turner, J. (1951), ‘Roman elementary mathematics’, Classical Journal 47.2: 63–74, 106–8. Turner, M. (1991), Reading Minds: The Study of English in the Age of Cognitive Science, Princeton: Princeton University Press. Turner, M. (1996), The Literary Mind: The Origins of Thought and Language, Oxford: Oxford University Press. Tylén, K., R. Fusaroli, J. Bjørndahl, J. Raczaszek-Leonardi, S. Østergaard and F. Stjernfelt (2015), ‘Diagrammatic reasoning: Abstraction, interaction, and insight’, Pragmatics and Cognition 22: 263–81. Van den Stock, J. and B. de Gelder (2014), ‘Face identity matching is influenced by emotions conveyed by face and body’, Frontiers in Human Neuroscience 8, (last accessed 7 February 2018). Van den Stock, J., R. Righart and B. de Gelder (2007), ‘Body expressions influence recognition of emotions in the face and voice’, Emotion 7.3: 487–99. Van der Eijk, P. J. (ed.) (1994), Aristoteles: De Insomniis, De Divinatione per Somnum, Berlin: Akademie Verlag. Van der Eijk, P. J. (1997), ‘The matter of mind: Aristotle on the biology of “psychic” processes and the bodily aspects of thinking’, in W. Kullmann and S. Föllinger (eds), Aristotelische Biologie: Intentionen, Methoden, Ergebnisse, Stuttgart: Steiner, pp. 231–58. Van der Eijk, P. J. (2000), ‘Aristotle’s psycho-physiological account of the soulbody relationship’, in J. P. Wright and P. Potter (eds), Psyche and Soma: Physicians and Metaphysicians on the Mind-Body Problem from Antiquity to Enlightenment, Oxford: Oxford University Press, pp. 57–77. Van der Eijk, P. J. (2005), Medicine and Philosophy in Classical Antiquity: Doctors and Philosophers on Nature, Soul, Health and Disease, Cambridge: Cambridge University Press. Van Duijn, M. J. (2016), ‘The lazy mindreader: A humanities perspective on mindreading and multiple-order intentionality’, PhD thesis, Leiden University. Varakis, A. (2010), ‘Body and mask in Aristophanic performance’, Bulletin of the Institute of Classical Studies 53.1, (last accessed 7 February 2018). Varela, F., E. Thompson and E. Rosch (1991), The Embodied Mind: Cognitive Science and Human Experience, Cambridge, MA: MIT Press. Vasaly, A. (1993), Representations: Images of the World in Ciceronian Oratory, Berkeley: University of California Press. Vatri, A. (2015), ‘Ancient Greek writing for memory’, Mnemosyne 68: 750–73. Vernant, J.-P. (1965), ‘Aspects mythiques de la mémoire’, in Mythe et pensée chez les Grecs, Paris: Maspero, pp. 49–78.

284

bibliograp h y

Vernant, J.-P. and F. Frontisi-Ducroux (1988), ‘Features of the mask in ancient Greece’, in J.-P. Vernant and P. Vidal-Naquet (eds), Myth and Tragedy in Ancient Greece, New York: Zone Books, pp. 189–206. Verzone, P. (1978), ‘Hierapolis di Frigia nei lavori della missione archeologica italiana’, Quaderni di ricerca scientifica 100: 391–475. Visvardi, E. (2015), Emotion in Action: Thucydides and the Tragic Chorus, Leiden: Brill. Von Staden, H. (2000), ‘Body, soul, and nerves: Epicurus, Herophilus, Erasistratus, the Stoics and Galen’, in J. P. Wright and P. Potter (eds), Psyche and Soma: Physicians and Metaphysicians on the Mind-Body Problem from Antiquity to Enlightenment, Oxford: Oxford University Press, pp. 79–116. Wainer, H. (1997), ‘Improving tabular displays, with NAEP tables as examples and inspirations’, Journal of Educational and Behavioral Statistics 22: 1–30. Waldron, J. (1995), ‘The wisdom of the multitude: Some reflections on book 3, chapter 11, of Aristotle’s Politics’, Political Theory 23: 563–84. Walker, A. D. (1993), ‘Enargeia and the spectator in Greek historiography’, Transactions of the American Philological Association 123: 353–77. Wallace, R. W. (2004), ‘The power to speak – and not to listen – in ancient Athens’, in I. Sluiter and R. M. Rosen (eds), Free Speech in Classical Antiquity, Leiden: Brill, pp. 221–32. Wankel, H. (1979), Die Inschriften von Ephesos, Bonn: Rudolf Habelt. Ward, D. (2012), ‘Enjoying the spread: Conscious externalism reconsidered’, Mind 121.483: 731–51. Ward, D. (2014), ‘Enactivism’, History of Distributed Cognition Project, (last accessed 7 February 2018). Watson, J. (ed.) (2001), Speaking Volumes: Orality and Literacy in the Greek and Roman World, Leiden: Brill. Webb, R. (2009), Ekphrasis, Imagination and Persuasion in Ancient Rhetorical Theory and Practice, Farnham: Ashgate. Webb, R. (2016), ‘Sight and insight: Theorizing vision, emotion and imagination in ancient rhetoric’, in M. Squire (ed.), Sight and the Ancient Senses, London and New York: Routledge, pp. 205–19. Weber, A. and F. Varela (2002), ‘Life after Kant: Natural purposes and the autopoietic foundations of biological individuality’, Phenomenology and the Cognitive Sciences 1: 97–125. Webster’s New International Dictionary of the English Language (1959), 2nd edn, Springfield: G. and C. Merriam Company. Welles, C. B. (1934), Royal Correspondence in the Hellenistic Period, New Haven: Yale University Press. West, M. (1989), ‘The early chronology of Attic tragedy’, Classical Quarterly 39: 251–4. Wheeler, J. A. (1990a), ‘Information, physics, quantum: The search for links’, in W. H. Zurek (ed.), Complexity, Entropy, and the Physics of Information, Redwood City, CA: Addison-Wesley. Wheeler, J. A. (1990b), ‘It from bit’, in S. Kobayashi, H. Ezawan, Y. Murayama and



bibliogra p h y 285

S. Nomura (eds), Proceedings of the Third International Symposium of Quantum Mechanics in the Light of New Technology, Tokyo: Physical Society of Japan, pp. 354–68. Wheeler, M. (2005), Reconstructing the Cognitive World: The Next Step, Cambridge, MA: MIT Press. Wheeler, M. (2010), ‘In defense of extended functionalism’, in R. Menary (ed.), The Extended Mind, Cambridge, MA: MIT Press, pp. 245–70. Wheeler, M. (2013), ‘Is cognition embedded or extended? The case of gestures’, in Z. Radman (ed.), The Hand, an Organ of the Mind: What the Manual Tells the Mental, Cambridge, MA: MIT Press, pp. 269–302. Wheeler, M. (2014), ‘Distributed cognition in the analytic and continental traditions’, History of Distributed Cognition Project, (last accessed 7 February 2018). Wheeler, M. (2015), ‘Extended consciousness: An interim report’, Southern Journal of Philosophy 53: 155–75. Wheeler, M. and A. Clark (2008), ‘Culture, embodiment and genes: Unravelling the triple helix’, Philosophical Transactions of the Royal Society 363: 3563–75. Wild, B., F. A. Rodden, W. Grodd and W. Ruch (2003), ‘Neural correlates of laughter and humour’, Brain 126: 1–18. Wiles, D. (2007), Mask and Performance in Greek Tragedy: From Ancient Festival to Modern Experimentation, Cambridge: Cambridge University Press. Williams, B. and R. Williams (1995), ‘Finger numbers in the Greco-Roman world and the early Middle Ages’, Isis 86.4: 587–608. Williams, R. F. (2013), ‘Distributed cognition and gesture’, in C. Müller, A. Cienki, E. Fricke, S. Ladewig, D. McNeill and S. Tessendorf (eds), Body – Language – Communication: An International Handbook on Multimodality in Human Interaction, Berlin: De Gruyter, pp. 240–58. Wilson, A. M. (1977), ‘The individualized chorus in old comedy’, The Classical Quarterly 27: 278–83. Wilson, R. A. (2005), ‘Collective memory, group minds, and the extended mind thesis’, Cognitive Processing 6.4: 227–36. Wilson, R. A. and G. Theiner (2013), ‘Group mind’, in B. Kaldis (ed.), Encyclopedia of Philosophy and the Social Sciences, vol. 1, Los Angeles: Sage, pp. 401–4. Winkielman, P., K. C. Berridge and S. Sher (2011), ‘Emotion, consciousness and social behavior’, in J. Decety and J. Cacioppo (eds), The Oxford Handbook of Social Neuroscience, Oxford: Oxford University Press, pp. 195–211. Wolf, W. (2001), ‘The emergence of experiential iconicity and spatial perspective in landscape descriptions in English fiction’, in O. Fischer and M. Nänny (eds), The Motivated Sign, Amsterdam and Philadelphia: John Benjamins, pp. 323–50. Wörrle, M. (1988), Stadt und Fest im kaiserzeitlichen Kleinasien, Studien zu einer agonistischen Stiftung aus Oinoanda, Munich: C. H. Beck. Wright, C. (2007), Poussin Paintings: A Catalogue Raisonné, London: Chaucer Press. Yu, N. (2003), ‘Chinese metaphors of thinking’, Cognitive Linguistics 14.2–3: 141–65.

286

bibliograp h y

Zahavi, D. (2014), Self and Other: Exploring Subjectivity, Empathy, and Shame, Oxford: Oxford University Press. Zanibbi, R., D. Blostein and J. Cordy (2004), ‘A survey of table recognition’, Document Analysis and Recognition 7: 1–16. Zanker, G. (1981), ‘Enargeia in the ancient criticism of poetry’, Rheinisches Museum 124: 297–311. Zedda, S. (2000), ‘How to build a world soul: A practical guide’, in M. R. Wright (ed.), Reason and Necessity: Essays on Plato’s Timaeus, London: Duckworth and Classical Press of Wales, pp. 23–41. Zhang, L. and W. Lin (2013), Selective Visual Attention: Computational Models and Applications, Oxford: Wiley. Zorzetti, N. (1991), ‘Poetry and the ancient city: The case of Rome’, Classical Journal 86: 311–29. Zuiderhoek, A. (2009), The Politics of Munificence in the Roman Empire, Cambridge: Cambridge University Press. Zunshine, L. (2006), Why We Read Fiction: Theory of Mind and the Novel, Columbus: Ohio State University Press. Zwaan, R. A. (2004), ‘The immersed experiencer: Towards an embodied theory of language comprehension’, in B. H. Ross (ed.), The Psychology of Learning and Motivation, San Diego and London: Elsevier Academic Press, pp. 35–62. Zwaan, R. A. (2008), ‘Experiential traces and mental simulations in language comprehension’, in M. de Vega, A. M. Glenberg and A. C. Graesser (eds), Symbols and Embodiment: Debates on Meaning and Cognition, Oxford: Oxford University Press, pp. 165–80. Zwaan, R. A. (2009), ‘Mental simulation in language comprehension and social cognition’, European Journal of Social Psychology 39: 1142–50. Zwaan, R. A. (2014), ‘Embodiment and language comprehension: Reframing the discussion’, Trends in Cognitive Sciences 18.5: 229–34. Zwaan, R. A. and C. J. Madden (2005), ‘Embodied sentence comprehension’, in D. Percher and R. A. Zwaan (eds), Grounding Cognition: The Role of Perception and Action in Memory, Language and Thinking, Cambridge: Cambridge University Press, pp. 224–45.



in dex 287

I n de x

Note: Page numbers in italics indicate illustrations. Page numbers followed by ‘n’ refer to footnotes. 4E cognition, viii, 3–9; see also embedded cognition; embodied cognition; enactive cognition; extended cognition action, 155 adaptivity, 154 Aeschylus Libation Bearers, 80 Theoroi, 86–7 affective cognition, 19–20 affective space, 138 affectivity, 7 cognitive, 19–20 Agamemnon (Aeschylus), 196–8 agency see collective agency; material agency Almagest (Ptolemy) see Syntaxis mathematica (Ptolemy) Anabasis (Xenophon), 170 Ancient Greek friendship see Greek friendship anger, 139, 146 animals, 154, 155–6, 158, 166 Annales (Tacitus), 122 Antović, Mihailo, 23 Aphrodisias, theatre of, 126, 130–1 Appadurai, A., 89 appraisal, 132 embodied, 140 Aristophanes, 75 Aristotle democracy, 200–1 emotions, 133, 144, 145–6 friendship, 34, 209, 212–13, 214–24 group minds, 34 hylomorphic theory, 19 performance, 90 Poetics, 89 in School of Athens, Plate 3, 226, 227 arithmetical calculation, 102–6

armillary sphere, 49–50 artefacts, 16–17 cognitive, 120 religious, 86–7 see also paintings artificial intelligence, 11 artificial neural networks, 68–9 Asia Minor, 121, 123–4, 125, 126, 128–31 astronomical instruments, 49–51 astronomy, 48–9 Athenian tragedy, 25 Athens divided groups, 196–200 group agency, 204–6 group minds, 206–8 group minds and individual minds, 200–4 unitary and unanimous groups, 192–6 see also entries starting with ‘Greek’ Augustine of Hippo, 103 Ausonius, Decimius Magnus, 212 autopoiesis, 154 autopoietic enactivism, 6–7 Avatar (movie), 85 Bacchae (Euripides), 84, 87 Barthes, R., 73 basic affects / emotions, 135n Bettini, Maurizio, 96 body Greek dramatic masks, 76–7, 78–9 Hippocratic Corpus, 134–49 perceiving body, 148 Ptolemy, Claudius, 40 Stoicism, 152–3 see also embodied cognition body movements, 81, 82, 182–9; see also gestures Bommas, Martin, 95 brain, 41, 77, 134–6, 137

288

in dex

Brennan, S.E., 83 ‘brief introduction to distributed cognition, A’ (Rogers), 209 Budelmann, F., 22 Bunraku puppets, 85 Butler, Judith, 12 Cairns, D. L., 24–5 calendars, 67 Cameron, James, 85 Capitoline Fasti (Roman lists of magistrates and military victors), 61 caricatures, 82–3 Cartesian dualism, 156–7; see also mind-body dualism; substance dualism cartography, 44–8, 51 Categories (Aristotle), 217 Chalmers, David, 92 Chaniotis, Angelos, 125 choruses, 34, 79 collective agency, 205, 206 combined group and individual minds, 202–4 as divided groups, 196–8 gestures, 80 group minds, 206–8 Theatre of Dionysus, 78 as unitary groups, 192, 194–6 Chrysippus of Soli, 161 churches, 118–19 Cicero, Marcus Tullius De oratore, 35–6, 99–100, 114–15, 233–8 enargeia, 175, 179 friendship, 209 nomenclator, 97–8 Clark, Andy, 3–4, 12–13, 86, 92, 149, 222 class see social class Classical Plot and the Invention of Western Narrative, The (Lowe), 22 Classics and cognitive sciences, 20–4 and distributed cognition, 18–20, 24–7 clothing, 124–6 cognition, 1–2, 150–1 and emotions, 135 Cognition in the Wild (Hutchins), 118 cognitive affectivity, 19–20 cognitive artefacts, 120 cognitive ecology, 117–18 cognitive humanities, 10–17 cognitive linguistics, 23–4 cognitive metaphors see metaphors cognitive narratology, 22 cognitive niche construction, 3–4, 14, 15 cognitive sciences, 11 and Classics, 20–4 cognitive turn, 10–17

collective agency, 204–6 collective cognition see group minds collective experience, 23 Colombetti, Giovanna, 132, 137, 146–7, 161–2 comic masks, 72 communication see gestures; language complementarity, 119 conceptual metaphor theory, 109–10 Connerton, Paul, 95 costume see clothing; dramatic masks Csapo, Eric, 79 cultural neuroscience, 208n cultural studies, 11 culture, 15, 17 cybernetics, 11 cyborgs, natural born, 4 Damasio, A., 7, 77, 135–6, 137, 138 dance drama, 79 Darius Vase, 105, 105 data tables (kanones), 43–4, 47–9, 51, 56 De anima (Aristotle), 133, 144, 145 De Gelder, B., 76, 77, 78 De humoribus (Hippocratic Corpus), 140–1, 148 De Jaegher, H., 178, 212 De morbo sacro (Hippocratic Corpus), 134, 137, 142, 144–5, 148 De oratore (Cicero), 35–6, 99–100, 114–15, 233–8 De victu (Hippocratic Corpus), 136 DeLoache, J.S., 86 democracy, 200–1 dēmos, 33–4, 192–4 collective agency, 205–6 combined group and individual minds, 200–2 as divided groups, 198–200 group minds, 206–8 Demosthenes, 125, 182 Dennett, Daniel, 11, 12 development, 158 Dewey, John, 223 diagrams, 68–9; see also lists; multiply-nested lists; nested lists; tables dialecticians, 144 Dinarchus, 194 Dio Chrysostom, 124, 125, 127–8 Dionysus (Greek god of the theatre), 76, 84, 86 distant reading, 12n distributed cognition, viii, 1–2, 209–10, 215n 4E cognition, viii, 3–9; see also embedded cognition; embodied cognition; enactive cognition; extended cognition and Classics, 18–20, 24–7 and the cognitive turn, 10–17 history of, 225–6 Donald, M., 120



in dex 289

Dow, Jamie, 146 dramatic masks, 29, 71–2 body and emotions, 76–7 changes over time, 91 comic masks, 72 emotional, 77–9 emotional intimacy, 90–1 enactive, 81–3 as extended mind tool, 73–4 as fetish, 89–90 as frame, 74 gaze direction, 74–6 gestures, 79–81 as material agent, 87–9; see also masks: as fetish as material anchor, 84–6 Pronomos Vase, Plate 1, 72, 75, 76, 81, 82, 88 as representations or surrogates, 83–4 satyr masks, Plate 2, 72, 86–7, 88 as surrogate face, 86–7 dress see clothing dualism see mind-body dualism; substance dualism Dupont, Florence, 115 Easterling, P., 22 eidetic memory, 94 Ekman, P., 77–8 ekplēxis, 187 Eldredge, Sears, 88 elementary automatisms, 140 elite portrait statuary, 126–8 elpis, 25 emanationist theory of vision, 46n embedded cognition, 3–4, 5, 8, 77 embodied appraisal, 140 embodied cognition, 3, 5, 12, 14, 77, 238 metaphors, 24–5 reasoning, 92 and Stoicism, 152, 156–62, 166 Timaeus (Plato), 162–4 Embodiment in Latin Semantics (Short), 24 emissionist theory of vision, 46 emotion metaphors, 24–5 emotional contagion, 190 emotional intimacy, 91 emotional masks, 77–9 emotionality, 90 emotions, 214–15 Aristotle, 133, 144, 145–6 Athenian tragedy, 25 basic emotions, 135n enactivism, 7 Greek dramatic masks, 76–9 Hippocratic Corpus, 32, 132–3, 133–49 Quintilian, 188

sociality of, 19–20 Stoicism, 160–2 see also love enacted representations, 54 enactive cognition, 6–7, 77 Hippocratic Corpus, 144–5, 146–7 Homeric poems, 27 and Stoicism, 151, 152–6, 161 enactive intersubjectivity, 212 enactivist theory of visual experience and imagination, 169–70, 176–9 bodily movement, 182–9 narrative space, 179–82 enargeia, 33, 169–70 enactivist account, 179–89 pictorialist account, 171–6 Ephesus, theatre of, 123–4, 125, 126 Epidemics (Hippocratic Corpus), 143 Eudemian Ethics (Aristotle), 218, 219, 220, 222 eunoia, 216 Euripides Bacchae, 84, 87 Iphigenia in Tauris, 185 Medea, 198 Phaethon, 185–7 Rhesus, 202–4, 205 exograms, 120 extended cognition, 4–5, 73 and friendship, 221 in memorisation strategies, 96–102 social dimension, 8 and Stoicism, 167 see also externalisation strategies extended mind hypothesis / theory, 26–7, 92, 119–20, 149 external storage devices, 120 externalisation strategies, 103–6 face recognition, 83 Fagan, Garrett, 26 fear, 141, 142, 188 Fedriani, Chiara, 113 feelings see emotions; sensations fetishism, methodological, 89–90 Fitzgerald, William, 211 Foley, H., 84 French, P., 204 friendship, 34, 209–10, 212–14 as collaborative activity, 219–20 and love, 215–17, 222 and self-knowledge, 220–1 and trust, 222–3 and virtue, 218–19 Frontisi-Ducroux, F., 84 Fuchs, T., 178, 212

290

in dex

Galen of Pergamon, 143–4 Gallagher, Shaun, 152, 157, 177 García-Jurado, Francisco, 113 gaze direction, 74–6 gender, 95 Generation of Animals (Aristotle), 201 Geography (Ptolemy), 44–5, 45–6, 47, 54 gestures, 79–81; see also body movements Gibbs, Raymond, 138, 149 Giere, R., 60, 68–9 Globe theatre, 118 god-pneuma, 164, 165 Goldin-Meadow, S., 80 Gombrich, E.H., 83 Goody, J., 60 Greek dramatic masks, 29, 71–2 body and emotions, 76–7 changes over time, 91 comic masks, 72 emotional, 77–9 emotional intimacy, 90–1 enactive, 81–3 as extended mind tool, 73–4 as fetish, 89–90 as frame, 74 gaze direction, 74–6 gestures, 79–81 as material agent, 87–9; see also masks: as fetish as material anchor, 84–6 Pronomos Vase, Plate 1, 72, 75, 76, 81, 82, 88 as representations or surrogates, 83–4 satyr masks, Plate 2, 72, 86–7, 88 as surrogate face, 86–7 Greek friendship, 34, 209–10, 212–14 as collaborative activity, 219–20 and love, 215–17, 222 and self-knowledge, 220–1 and trust, 222–3 and virtue, 218–19 Greek memorisation techniques, 96–7, 99–100 Greek metaphors, 24–5, 113–14 Greek-style theatres, 123–4, 125–6, 128–31 Gregory, Richard, 83 Grethlein, Jonas, 23, 27 group agency, 204–6 group mind hypothesis, 8 group minds, 33–4, 190–208 divided groups, 196–200 and individual minds, 200–4 unitary and unanimous groups, 192–6 see also social cognition Gundert, Beate, 147–8 Habinek, Thomas, 26–7 Halbwachs, Maurice, 95

happiness, 139 harmonic instruments, 51–4 Harmonics (Ptolemy), 41–2, 55 Harris, W.V., 62 hearing, 40, 148 hēgemonikon, 41, 41n helikōn, 52 Herodotus, 213 Hierapolis, theatre of, 129–30 Hierocles, 158–60, 161 Hippias Maior (Plato), 95 Hippocratic Corpus, 32, 132–49 emotions and body, 132–3, 133–44 enactive cognition, 144–5, 146–7 historical narratives, 128–31 history, 225–6 Hitz, Zena, 219–20 Holland, Norman, 14 hollow mask experiment, 83 Homer, 23, 170 Odyssey, 90 Homer and the Resources of Memory (Minchin), 21 Homeric Hymn to Apollo, 90 Homeric poems, 21, 27 Homeric Voices (Minchin), 21 Honneth, Axel, 223 hormē, 155 Huitink, L., 27 humanities see cognitive humanities humans, 14, 153, 154–5, 166 Hurley, S., 59–60, 62 Hutchins, Edwin cognitive ecology, 118 distributed cognition, 1 enacted representations, 54 material anchors, 52 navigation work, 53, 200 navigational charts, 44, 45 hylomorphic theory, 19 identity, 126–8 Iliad (Homer), 170 imagination see readerly imagination individual identity, 126–8 information, 167 information technology, 57, 58; see also diagrams; lists; multiply-nested lists; nested lists; tables inlustratio, 175; see also enargeia Institutio oratoria (Quintilian), 100–1 instruments, 28, 37–8 astronomical, 49–51 harmonic, 51–4 textual, 42–9 intellect, 39 intimacy, 91



in dex 291

Ion (Plato), 95, 96 Iphigenia in Tauris (Euripides), 185 Iser, W., 171 Jacob, C., 51 James, William, 139 Japan Bunraku puppets, 85 Noh drama, 80 Johansson, G., 82 Johnson, M., 3, 24 Jones, J., 84 kanōn, 43, 52–3, 54 kanones (data tables), 43–4, 47–9, 51, 56 kata, 80 Keene, N., 118–19 Kirsh, D., 73, 84–5, 120 Knappett, C., 87, 88 Konstan, David, 139–40 Krebs, Angelika, 223 Lacan, J., 11 Lakoff, G., 3, 24, 109–10 Landscape with Numa Pompilius and the Nymph Egeria (Poussin), Plate 4, 35, 230–2 land-surveyors, 29, 63, 64–5 language, 11 Le Bon, Gustav, 190 LeDoux, Joseph, 77 Letters to Lucilius (Seneca the Younger), 210–12 Libation Bearers (Aeschylus), 80 ‘Liebeslied’ (Rilke), 223–4 linguistic signs, 83 linguistic turn, 11 linguistics see cognitive linguistics List, C., 204 lists, 60 in Ancient Rome, 61–2, 65–7 see also diagrams; multiply-nested lists; nested lists; tables literacy, 62, 65 literary criticism, 33 Llewellyn-Jones, Lloyd, 80 long-term memory, 93, 94 looking, 180–1; see also seeing; sight; vision love, 215–17, 222, 223 Lowe, N. J., 22 Lure of the Arena (Fagan), 26 Magna moralia (Aristotle), 220 Malafouris, Lambros, 72, 73, 77, 81, 83–4, 87, 89 maps, 44–8, 51 Marchese, F., 59, 60–1 marionettes see puppets

masks, 29, 71–2 body and emotions, 76–7 changes over time, 91 comic masks, 72 emotional, 77–9 emotional intimacy, 90–1 enactive, 81–3 as extended mind tool, 73–4 as fetish, 89–90 as frame, 74 gaze direction, 74–6 gestures, 79–81 as material agent, 87–9; see also masks: as fetish as material anchor, 84–6 Pronomos Vase, Plate 1, 72, 75, 76, 81, 82, 88 as representations or surrogates, 83–4 satyr masks, Plate 2, 72, 86–7, 88 as surrogate face, 86–7 material agency, 87–9; see also methodological fetishism material anchors, 52 masks as, 84–6 Material Engagement Theory (MET), 71–2, 73, 83–4, 87 material signs, 83–4 mathematics, 42; see also arithmetical calculation Medea (Euripides), 198 Meineck, Peter, 25 melting, 82 memes, 11 memorisation strategies, 30, 95–6 in Ancient Greece, 96–7, 99–100 in Ancient Rome, 96, 97–8, 100–2 see also metaphors memory gender differences, 95 Homeric poems, 21 types of, 93–4 see also mnemotechnics mental images see readerly imagination meshing, 149 metaphors, 93 in Ancient Greece, 24–5, 113–14 in Ancient Rome, 30, 107–12 methodological collectivists, 191 methodological fetishism, 89–90 methodological individualists, 191 military duty rosters, 63, 65, 66–7, 69 Minchin, Elizabeth, 21, 23, 26 mind, 133, 164, 165 received (non-distributed) view of, 2 see also group minds mind-body dualism, 40, 156–7, 162–3, 163–4; see also substance dualism minimal embodiment, 157

292

in dex

mnēmōn, 96–7, 98–9 mnemotechnics, 30, 99–102 monitor, 98–9 monochords see kanōn moods, 7; see also emotions Mooney Faces, 83 Mori, Masahiro, 85 motives, 155–6 movement, 81, 82, 138–9, 182–9 multiply-nested lists, 28, 59, 62 museum artefacts, 16–17 musical instruments see harmonic instruments mythological friezes, 128–31 narrative Athenian drama, 90 historical, 128–31 see also cognitive narratology narrative space, 179–82 natural born cyborgs, 4 natural theology, 152–3, 166 navigational charts, 44 nested lists, 28–9 in Ancient Rome, 64, 66, 67–8, 69–70 definition, 58–9 multiply-nested lists, 28, 59, 62 relationship between tables and, 70 Neth, H., 103 Netz, Reviel, 106 neuroscience, 10, 25, 76, 77, 93 cultural, 208n Nicomachean Ethics (Aristotle), 214, 216, 218 Noë, A., 17, 46, 77, 176 Noh drama, 80 nomenclator, 97–8, 98–9, 112, 210–11 non-distributed view of mind, 2 Novack, M. A., 80 Novick, L., 59–60, 62 Nussbaum, Martha, 217 Nysa, theatre of, 128–9 Odyssey (Homer), 90 Oedipus vase, 75n On Friendship (Cicero), 209 On the Criterion of Truth (Ptolemy), 38–41, 54, 55 On the Latin Language (Varro), 63 ontological metaphors, 113 open-air performances, 73–4 oral poetics, 21, 26 O’Regan, J. K., 46, 176 Osborne, Catherine, 221 Pagán Cánovas, Cristóbal, 23–4 paintings Landscape with Numa Pompilius and the Nymph Egeria (Poussin), Plate 4, 35, 230–2

School of Athens (Raphael), Plate 3, 35, 226–8, 229, 232 parasites, 211 participatory sense-making, 9 passions see emotions perceiving body, 148 perception self-perception, 158–9, 161 sense perception, 148 personal identity, 126–8 perspective aérienne, 232 Pettit, Philip, 200, 204 Phaethon (Euripides), 185–7 phantasia, 39n, 46, 133, 155, 171–6, 184 phenomenology, 6, 10–11 philēsis, 216 philia, 216–17, 219 Philoctetes (Sophocles), 82 philosophy, 227 phrikē, 25 physicists, 144 Pickard-Cambridge, A., 82, 84 pictorialist account of enargeia and phantasia, 171–6 pistis, 222–3 pity, 215 Plato Hippias Maior, 95 Ion, 95, 96 Protagoras, 35, 228–9 in School of Athens, Plate 3, 226, 227 Timaeus, 31–2, 151: and embodiment, 162–4; universe as rational, 164–8 Plautus, Titus Macchius, 113 Pliny the Elder, 103 Pliny the Younger, 106 Plutarch, 170, 175 pneumatic theory, 19 Poetics (Aristotle), 89 poetry see Homeric poems; oral poetics politics see dēmos; Roman senate Politics (Aristotle), 200–1 Polybius, 180–1 portrait statuary, 126–8 postmodern deconstructionism, 11–13 Poussin, Nicolas Landscape with Numa Pompilius and the Nymph Egeria, Plate 4, 35, 230–2 pre-emotions, 161 primary intersubjectivity, 177–8 Prinz, Jesse, 140, 141–2 proedria, 124 projection, 84–6; see also surrogacy Pronomos Vase, Plate 1, 72 enactive masks, 82 gaze direction, 75, 76



in dex 293

gestures, 81 material agency, 88 prosōpon, 89 Protagoras (Plato), 35, 228–9 Protestant churches, 118–19 Ps.-Longinus, 175, 180, 181, 184–7 psychē, 19, 25 Claudius Ptolemy, 38–42, 55 see also world-psyche Psychologie des foules (Le Bon), 190 psychology, 150, 163 psychotherapy, 146 Ptolemy, Claudius, 27–8, 37–55 astronomical instruments, 49–51 digitalised manuscripts, 56 harmonic instruments, 51–4 psychē, 38–42, 55 textual instruments, 42–9 puppets, 85, 87 Pythagoras, Plate 3, 226, 230–1 Quintilian enargeia, 170, 172, 174, 175, 179, 180, 188 mnemotechnics, 100–1 place, 114 Raphael, School of Athens, Plate 3, 35, 226–8, 229, 232 rationality, 151, 164–8 readerly imagination, 33, 169–70 enactivist account, 176–9: bodily movement, 182–9; narrative space, 179–82 pictorialist account, 171–6 reason, 41–2 received view of mind, 2 Reformation era, 118–19 religious artefacts, 86–7 representations, 83–4, 86 Rhesus (Euripides), 202–4, 205 Rhetoric (Aristotle), 133, 145–6, 216 Rhetorica ad Herennium, 100, 101, 102 Rilke, Rainer Maria, 223 Rogers, Yvonne, 209 Roman information technology see lists; multiply-nested lists; nested lists; tables Roman senate, 30, 105–6, 113 Roman technical practices, 30, 93 arithmetical calculation, 102–6 memorisation strategies, 96–102, 112–16 metaphors, 30, 107–12 Roman-period theatres, 31, 117, 119–20 clothing, 124–6 mythological friezes, 128–31 portrait statuary, 126–8 seating arrangements, 120–3 Rupert, Robert, 206

Sabinus, Calvisius, 210–11, 212 Salutaris in Ephesus, 126 satyr masks, Plate 2, 72, 86–7, 88 scaffolding, 63–4, 237–8 schēmata, 80 Schlimm, D., 103 School of Athens (Raphael), Plate 3, 35, 226–8, 229, 232 scientific investigation, 28 Scodel, Ruth, 22–3 seating arrangements, 120–3 seeing, 46, 148, 173, 178; see also looking; sight; vision self-knowledge, 220–1 self-perception, 158–9, 161 Seneca the Younger, 25, 97, 98, 210–12 sensations, 138, 140, 163, 164 sense organs, 39, 42–3 sense perception, 148 sense-making, 154–5, 155–6 participatory, 9 senses, 40; see also hearing; touch; vision sensorimotor enactivism, 6 sensory memory, 93 Serling, Rod, 89 Shakespearean theatre, 117–18 shame, 215 shiori, 80 Short, William, 21, 24 short-term memory, 93, 94 sight, 40; see also looking; seeing; vision Simonides of Ceos, 99 singing, 53–4 slaves, 97–9, 112, 210–12, 215, 234 Small, Jocelyn Penny, 21–2, 101, 102 social class and clothing, 124–6 and portrait statuary, 126–8 and seating arrangements, 121–3 social cognition, 7–9, 26; see also friendship; group minds social manifestation hypothesis, 8 Socrates, 228, 229 ‘Soma and psyche in Hippocratic medicine’ (Gundert), 147–8 somatic markers, 7 Sophocles, 82 spatial metaphors, 30, 93, 107, 110–11 split focus, 75 Spolsky, Ellen, 14 star catalogues (Ptolemy), 48–9 Stoicism, 31, 32 and embodiment, 26–7, 152, 156–62 and enactivism, 151, 152–6 pneumatic theory, 19 universe as rational, 164–8

294

in dex

storytelling, 96 substance dualism, 2; see also mind-body dualism Suetonius (Gaius Suetonius Tranquillus), 104 surrogacy, 83–4, 86–7 Sutton, John, 118, 119 symbolic representations, 86 Syntaxis mathematica (Ptolemy), 37, 48, 51, 54

touch, 40, 42 tragedy, 25 Tribble, Evelyn, 117–19 trust, 222–3 Twilight Zone, The (Serling), 89

tables, 28–9 in Ancient Rome, 60–1, 62–3, 64–7 benefits, 60, 69 definition, 58 modern study of, 59–60 relationship between nested lists and, 70 see also data tables (kanones); diagrams; lists Tacitus, Publius Cornelius, 122 technical practices see Roman technical practices tentacular mind, 26 textual instruments, 42–9 Theatre of Dionysus, 73, 78 theatres, 67–8 Roman-period, 31, 117, 119–20: clothing, 124–6; mythological friezes, 128–31; portrait statuary, 126–8; seating arrangements, 120–3 Shakespearean, 117–18 see also Greek dramatic masks theatrical frames, 74 Theoroi (Aeschylus), 86–7 Theory of Mind, 22–3 Thespis, 91 Thompson, E., 7 thymos, 25 Timaeus (Plato), 31–2, 151 and embodiment, 162–4 universe as rational, 164–8 tools see instruments

Van der Eijk, Philip, 145 Varro, Marcus Terentius, 63 Vasaly, A., 173 Vatri, Alessandro, 96 Vernant, J.-P., 84 virtue, 218–19 vision, 33, 46–7 emanationist theory of, 46n emissionist theory of, 46 see also looking; seeing; sight visual experience see enactivist theory of visual experience and imagination visualisation see pictorialist account of enargeia and phantasia vividness see enargeia vote counting, 105–6, 113 voting with the feet, 30, 106

uncanny valley, 85 universe, 151, 164–8

Ward, Dave, 152 Wax Tablets of the Mind (Small 1997), 21–2 Webb, R., 173–4, 188 Wiles, D., 84, 88 Wilson, R.A., 8 Wilson, Robert A., 191 working memory, 93 world-psyche, 164, 165 writing, 21, 30, 95–6, 237 Xanthias, 211–12 Xenophon, 170, 175, 198–200