Philosophical Perspectives on the Self (Lisbon Philosophical Studies – Uses of Languages in Interdisciplinary Fields) [New ed.] 9783034314022, 9783035107869, 3034314027

Table of contents :
Cover
Contents
Acknowledgments
Introduction (João Fonseca / Jorge Gonçalves)
I. The Background
II. The Essays
Part I. Metaphysics and Personal Identity
Part II. Epistemology and Phenomenology
Part III. Cognition, Psychology, Neuroscience
Part IV. Ontology and Taxonomy
References
Part I. Metaphysics and Personal Identity
Animalism and the Remnant-Person Problem (Eric T. Olson)
Introduction
1.
2.
3.
4.
5.
6.
7.
8.
9.
References
Will I ever be a Cyborg? (Rui Vieira da Cunha)
1. Introduction
2. A Cyborg Thought Experiment
3. Olson’s Animalist View
4. Back to the Cyborg
5. Conclusion
References
Part II. Epistemology and Phenomenology
How Consciousness explains the Self (Klaus Gärtner)
Justifying the Connection
What is Consciousness?
What is the ‘Self’?
The relation between a concept of Consciousness and a concept of the ‘Self’
Conclusion
References
Self-Knowledge, Introspection and Memory (António Marques)
References
Imagination as a Bodily Pattern: thinking about Sartrean’s account of Consciousness (Clara Morando)
Hypnagogic Images
References
Feelings and the Self (Dina Mendonça)
Part I: Situations and the Self
i. Situated approach to emotions
ii. Self and The Pattern of Emotional Activity
Part II: Some Emotions and the Implicated Self
i. Fear
ii. Love
iii. Pride
iv. Jealousy
Conclusions
References
Part III. Cognition, Psychology, Neuroscience
De Se Attitudes and Semiotic Aspects of Cognition (Erich Rast)
Overview
Puzzles of De Se Attitudes
Assessment of the Thought Experiments
Semiotic Aspects of Cognition
Summary and Conclusions
Figures
References
The Division of the Mind: Paradoxes and Puzzles (Vasco Correia)
1. Beyond divisionism
2. The paradoxes of the Freudian account
3. Fingarette: “ego” and “counter-ego”
4. Pears: the rival “centres of agency”
5. Davidson: the mind’s “compartments”
6. Conclusion: toward a unitary solution
References
Empirical and conceptual clarifications regarding the notion of ‘Core-Self’ from Gallagher’s and Merker’s Behavioural-Neuroscientific Proposals (João Fonseca)
I. Introduction: Conceptual Confusions and Methodological Fragmentation
II. Fundamentals of a Model-Theoretical Framework for Behavioural Neuroscience
III. Hierarchical taxonomy of psychological concepts: introducing ‘Nested Concepts’
IV. Redefining Merker’s and Gallagher’s proposals for Core-self
IV.1 Core-Self’ as a Theoretical Concept in Behavioural Neuroscience
IV.2 Merker’s upper brainstem proposal
IV.3 Gallagher’s cortical motor forward model proposal
V. Conciliating Merker and Gallagher’s proposals: ‘Core-Self’ as a Theoretical Nested Concept in Behavioural Neuroscience.
V.1 The ‘Explanatory Dilemma’
V.2 Core-Self as a ‘Psychological Domain’ of BN. Introducing the ‘Nested Concept Hypothesis’
V.3 Explanatory Unification and the Overcoming of the Dilemma
Conclusion
References
Part IV. Ontology and Taxonomy
Core Self and the Problem of the Self (Jorge Gonçalves)
References
The Reality of the Virtual Self as Interface to the Social World (Robert Clowes)
Introduction
Virtual Reality and the Experience of Presence
The Virtual Self and Pre-Reflective Self Experience
The Virtual Body and the Minimal Self
In What Sense is the Self Illusory?
Presence, the Minimal Self, and Ipseity Disturbance
Schizophrenia and Social Self Diminishment
Self Positions: An Extended Virtual Self  ?
Rethinking the Virtual Self: Actualizing Virtualities
The Virtuality of the Self and its Ontological Status
Actualizing Virtualities and the Conceptual Role of Self
Acknowledgements
References
Conceptual Personae of the “attentional self” (Alexander Gerner)
1. Introduction
1.1 Conceptual personae of the impossible attentional self: Monsieur Teste as the impossible “man” of attention
2. Heautoscopic “attentional self”
2.1 Italo Calvino’s Mister Palomar’s development from an aesthetic to an heautoscopic “attentional self”
From aesthetic to heautoscopic attention
2.2 Metzinger’s “attentional ghost”: the virtual self floating out of the body and the impossible full body illusion
2.2.1 The concept of disembodied attentional self in an out-of-body experience (OBE)
2.2.2 Virtual Out- of-body experiences and impossible full body illusions
3. Outlook
References
Notes on Contributors

Citation preview

For the last decade the topic of the Self has been under intense scrutiny from researchers of various areas spanning from philosophy, neurosciences, and psychology to anthropology and sociology. The present volume addresses the Self under different and influent philosophical perspectives: from phenomenology and psychoanalysis to metaphysics and neurophilosophy and discusses several and distinct problems such as personal identity, the core/narrative self-distinction, psychopathologies, the mind-body problem and the nature of the relations between self, consciousness and emotions. The book reflects these different philosophical problems and approaches and aims to provide a map of current philosophical perspectives on the topic of the Self. João Fonseca is a Post-Doc researcher at New University of Lisbon. His main research interest is philosophy of neuroscience. His area of competence is philosophy of neuroscience with particular interests in topics such as: the critical assessment to the mainstream neuro-behavior explanation of instrumental fear-conditioning, the quest for the evolutionary origins of self and consciousness and the links between phenomenology and neuroscience. Jorge de Almeida Gonçalves was born in Lisbon. He graduated in Psychology (1988) and in Philosophy (1997), both in Lisbon. He has a Master’s degree in Philosophy (2002) and a Ph.D. (2007) also in Philosophy. Between 1988 and 1999 he worked in Clinical Psychology. He currently works at the Institute of Philosophy of Language (New University of Lisbon) and his current research interests are consciousness and self studies, philosophy of psychology, philosophy of psychiatry, and philosophy of film.

Philosophical Perspectives on the Self

LISBON PHILOSOPHICAL STUDIES uses of language in interdisciplinary fields A P ub lic ation from the Ins t it ut e o f Philo s o phy o f L a ngu ag e at t h e N e w U n i v e r s i t y o f Li s b o n edited b y Antón io Marq ues (Ge ne ra l E dit o r ) Nu no Ventu rin ha (Ex e cut ive E dit o r ) E ditorial Board : Gab riele De A ng elis, Hum be r t o B r it o, J o ã o Fo ns e ca , Fra n c k Li h o r e au , A n t ó n i o M ar q u e s, Maria Filomen a Molde r, Dio go Pir e s Aur é lio, E r ich R a st , J o ão S àág u a, Nu n o Ve n t u r i n h a Advisory Board: Je an- P ierre Cometti ( Unive r sit é de Pr o ve nce ), Lynn Do b s o n ( U n i v e r s i t y o f Ed i n b u r g h ) , E rnest L epore (Ru tge r s Unive r s it y), R e na t o L e ssa ( IUPE- R i o d e Jan e i r o ) , A n d r e w Lu g g (Un iversity of Ottawa ), S t e f a n M a je t s cha k ( Unive r sit ä t K as s e l ) , J e s ú s Pad i l l a Gál v e z (Un iversidad de Cas t illa - L a M a ncha ), J o a chim S chult e ( U n i v e r s i t ät Zü r i c h )

PETER LANG Bern · Berlin · Bruxelles · Frankfurt am Main · New York · Oxford · Wien

João Fonseca & Jorge Gonçalves (eds)

Philosophical Perspectives on the Self

PETER LANG Bern · Berlin · Bruxelles · Frankfurt am Main · New York · Oxford · Wien

Bibliographic information published by die Deutsche Nationalbibliothek Die Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available on the Internet at ‹http://dnb.d-nb.de›. British Library Cataloguing-in-Publication Data: A catalogue record for this book is available from The British Library, Great Britain Library of Congress Cataloging-in-Publication Data Philosophical perspectives on the self / João Fonseca & Jorge Gonçalves (eds). – 1st ed. pages cm. – (Lisbon philosophical studies, 1663-7674 ; Vol. 5) Includes bibliographical references. ISBN 978-3-03-431402-2 1. Self (Philosophy) I. Fonseca, João, 1971- II. Gonçalves, Jorge, 1960BD450.P472345 2014 126–dc23 2014044012

Financial Support: Project Cognitive Foundations of the Self PTDC/FIL-FCI/110978/2009 (FCT – Fundação para a Ciência e Tecnologia) ISSN 1663-7674 pb. ISBN 978-3-0343-1402-2 pb.

ISSN 2235-641X eBook ISBN 978-3-0351-0786-9 eBook

This publication has been peer reviewed. © Peter Lang AG, International Academic Publishers, Bern 2015 Hochfeldstrasse 32, CH-3012 Bern, Switzerland [email protected], www.peterlang.com All rights reserved. All parts of this publication are protected by copyright. Any utilisation outside the strict limits of the copyright law, without the permission of the publisher, is forbidden and liable to prosecution. This applies in particular to reproductions, translations, microfilming, and storage and processing in electronic retrieval systems.

Contents

Acknowledgments..................................................................................7 João Fonseca, Jorge Gonçalves Introduction ...........................................................................................9

Part I. Metaphysics and Personal Identity Eric T. Olson Animalism and the Remnant-Person Problem.....................................21 Rui Vieira da Cunha Will I ever be a Cyborg?......................................................................41

Part II. Epistemology and Phenomenology Klaus Gärtner How Consciousness explains the Self..................................................63 António Marques Self-Knowledge, Introspection and Memory.......................................73 Clara Morando Imagination as a Bodily Pattern: thinking about Sartrean’s account of Consciousness...................................................85

6

Contents

Dina Mendonça Feelings and the Self..........................................................................101

Part III. Cognition, Psychology, Neuroscience Erich Rast De Se Attitudes and Semiotic Aspects of Cognition..........................121 Vasco Correia The Division of the Mind: Paradoxes and Puzzles............................147 João Fonseca Empirical and conceptual clarifications regarding the notion of ‘Core-Self’ from Gallagher’s and Merker’s Behavioural-Neuroscientific Proposals..............................................165

Part IV. Ontology and Taxonomy Jorge Gonçalves Core Self and the Problem of the Self...............................................207 Robert Clowes The Reality of the Virtual Self as Interface to the Social World........221 Alexander Gerner Conceptual Personae of the “attentional self ”...................................277 Notes on Contributors........................................................................325

Acknowledgments

We would like to thank, first and foremost the Portuguese Foundation for Science and Technology for their support in funding the Research project The Cognitive Foundations of The Self (PTDC/FIL-FCI/110978/2009). We also would like to thank Peter Lang for all the help during the process and the Instituto de Filosofia da Linguagem for all the support and facilities. Finally, we wish to thank the following people: António Marques, Nuno Venturinha, Inês Hipólito, Daniel Ramalho and all the collaborators in this volume, among whom Peter Olson for his patience and sheer professionalism. Lisbon, November 2013 João Fonseca and Jorge Gonçalves

Introduction

I.  The Background The quest for the nature and scope of the human Self has been one of the most important intellectual tasks in western thought. Nevertheless, It was not until Descartes and the rise of modern philosophy, that the cluster of problems we now associate to the notion of ‘Self ’ were identified as such (eg.: self-identity, the nature of self-reflection, the epistemological status of self-evidence, the unity of conscious experience, among others). What was more, this set of problems were taken to be among the most crucial philosophical tasks to be addressed in the upcoming centuries. The work of such diverse authors as Hume, Lock, Kant, Nietzsche, William James, Husserl, Wittgenstein, or Sartre, to name just a few, testifies this importance. In the last 10 to 15 years the topic of the Self has strongly re-emerged. This renewed interest is illustrated by the number of recent collections of essays and anthologies (Gallagher, 1998; Kircher, 2003; Gallagher, 2010). One of the main factors holding behind such interest has to do with the recent burst of different methodologies and approaches adopted to face the set of problems related to the Self. These methodologies include but go beyond the more traditional philosophical approaches (like phenomenology or linguistic analysis) (Dan Zahavi, 2005; Perry, 2002), into empirical researches in the areas of cognitive psychology (Gallagher, 2005, 2008; Hofstadter 2007) several branches of the neurosciences (Damasio, 1999; LeDoux, 2002; Kircher, 2003), analysis of psychiatric pathologies (such as schizophrenia) (Parnas, 2010) and other disciplines and methodologies related to the interdisciplinary field of current cognitive sciences and even social theorists and cultural analysis (Elliot, 2007). These are, thus, exciting times in what the studies regarding the notion the ‘Self’ are concerned: neuroscientists and cognitive psychologists are accessing this notion by providing empirical methods and scientific tools (redefining and revolutionizing the way the western

10 

Introduction

tradition approaches the problem) philosophers and cognitive scientists from various traditions are rediscovering old methodologies like phenomenology and psychoanalysis and, psychiatric approaches to cases like schizophrenia are establishing links with philosophical proposals concerning the nature of the self and its ontological status. Succinctly, the last decade has seen the reemergence of the interest on the notion of ‘Self’ under a new interdisciplinary umbrella. As a drawback the adoption of such different approaches and methodologies seems to result in a proliferation of unarticulated and, most of the time, incommensurable concepts and results. Gallagher and Zahavi (2008, pp. 197–198) identify this problem: [T]his disparity, which is both problematic and productive, is directly related to the variety of methodological approaches taken within philosophy and in related interdisciplinary studies of the self. They include introspection, phenomenological analysis, the use of thought experiments, empirical research in cognitive and brain sciences, and studies of exceptional and pathological behaviour. One problem to be posed in this light is whether different characterizations of self signify diverse aspects of a unitary concept of selfhood, or whether they pick out different and unrelated concepts.

Therefore, this new transdisplinary approach comes with a cost: a taxonomical confusion and fragmentation inherent to the proliferation of so distinct methodological approaches. In itself, this consequence constitutes an unfortunate obstacle to the very progress in the field of the studies about the nature of the ‘Self’. Where it should be expected conceptual unity there is, instead, incommensurability and lack of communication. We believe that Philosophy, given its general, far reaching, synoptic and conceptual approach is specially suited to overcome this unfortunate scenario by providing conceptual clarification that facilitates the establishment of links between disciplines. The present volume is an attempt to a first approximation to different sub-topics and methodologies about the Self from a Philosophical standpoint. It should be stressed that, and as stated in the quote above, philosophical approaches to the problems of the self are, by themselves, everything but unified and monolithic. This diversification can be understood at two different levels: the level of the different methodological tools used, and the level concerning the diversity of topics and problems. Within the first level; the methodological one, a further division is useful: the distinction

Introduction

11

between, on one hand philosophical methodologies proper (Phenomenology, Conceptual/Linguistic analysis, third-person accounts) and, on the other, the relation established between Philosophy and other scientific approaches (Neuroscience, Cognitive Psychology, Artificial Intelligence, psychiatry). At the level of the different philosophical topics related to the Self they include: the problem of personal identity, discussions regarding the ontological status of the Self, the topic of self-knowledge and Immunity to Error trough Misidentification, modern assessments to the Mind-Body Problem and the nature of the relations between Self and consciousness and emotions. This collection of essays aims to provide a non-exhaustive map of this diversity within contemporary philosophy on the Self at both identified levels. It reflects the pluralism of philosophical perspectives associated with the problem (or set of problems) of the Self. Besides providing a general view on such diversity, we aim, at the end of the day and more implicitly, to suggest possible bridges unifying and relating apparent protracted and unrelated data and methodologies.

II.  The Essays Part I.  Metaphysics and Personal Identity Eric Olson explores his original proposal according to which persons are animals, i.e., complete organisms. In ‘Animalism and the Remnant-Person Problem’, Olson discusses a reading of the transplanted brains thought experiment. Animalism clashes with the conviction that we should go with our transplanted brains. A good reply is that if animalism were true, we could explain easily enough both why the conviction is false and why it seems compelling. But another objection cannot be answered so easily. Animalism seems to imply that the detached brain would be a person who comes into being when the brain is removed and ceases to exist when the brain gets into a new head. And this seems absurd. The article argues that, although

12 

Introduction

this is equally problematic for many views besides animalism, it has no obvious solution. In ‘Will I ever be a Cyborg’ Rui Vieira da Cunha criticises Olson’s views. Vieira da Cunha states that although very attractive, in no doubt because of its appeal to a scientific worldview, animalism is not without its problems. In Olson’s own brand of animalism, the Organism View, one of those problems is the answer to be given to situations of inorganic replacement, which this paper explores in a very specific thought experiment. If indeed Olson is right in saying that animal or human animal or organism is what best serves as a substance concept in the case of beings like you and I, then it seems a hard task for the animalist to account for the intuitions arising from the thought experiment in this paper, at least without changing substantially the concept of organism. Part II.  Epistemology and Phenomenology In ‘How Consciousness explains the Self’, Klaus Gärtner asks for an epistemological access to the Self in order to answer some ontological questions regarding its own existence. When we talk about the metaphysics of the ‘Self’ we want to know something about the ‘Self ’s’ nature. Since in Philosophy of Mind it is less than clear if the ‘Self ’ exists or not, it seems that we need a way to analyze it. A natural suggestion is that Consciousness might give us the access we are looking for. The article suggests a way of how a concept of the ‘Self ’ can be tight to a concept of Consciousness. The key to establish such a conceptual connection is Self-Consciousness. Such a relation has the advantage that the ‘Self’ is not isolated, it is rather connected to a phenomenon that is interdisciplinary studied. After testing this idea in a case where the ‘Self’ is compromised (schizophrenia), the article suggests that a concept of the ‘Self’ depends necessarily (but not sufficiently) on a concept of Consciousness. In ‘Self-Knowledge, Introspection and Memory’António Marques claims that self-knowledge by introspection (s-ki) leads to the question of the status of the content of retrodictions, wich are memory dependent statements. These are in specie different of retrodictions that are

Introduction

13

not memory dependent and therefore s-ki expresses itself in statements that have not truth value. Furthermore the fact that s-ki has a dual time structure (the representation at time 2 of an event or experience of a past time 1) doesn’t mean that any s-ki retrodiction is a kind a of a metarepresentation (a representation of representations). Finally, Marques states three other claims, namely: 1. s-ki contains a dual time structure based on memory, 2. it must be direct (non mediated by any exterior observer) and 3. it must have first-person authority. Clara Morando’s article aims to identify and analyze the phenomenon of imagination in Sartre’s philosophy, intending at the same time to clarify some possible connections between imaging skills as a specific activity of consciousness and the way physical bodies essentially incorporate those kinds of data. A synthetic explanation about how Sartre sees his theory of imagination is outlined. The article questions in what way this theory can contribute to a more accurate idea of psychophysical relations, contending that it makes no sense to employ the expression ‘psychophysical relations’ simply because the mind is the body and the body is the mind. Metaphysically, “imaginative consciousness”, seems to be the very core of the “transcendental consciousness”, since it can be compared to a simple intentional movement towards ‘objects’, also characterized by an essential “nothingness” and a great proximity to the “phenomenon of quasi-observation”, which is not genuine observation. In ‘Feelings and the Self’ Dina Mendonça starts by noticing that philosophers of emotion agree that emotions always implicate a self. However, it is not at all clear within the literature what kind of self, nor what kind of implication, philosophers have in mind. The article argues for a situational approach to the nature of emotions by, first, showing how a situational approach brings to the surface the interesting connections between the self and emotions, and second, by showing how this approach allows us to understand how emotions contribute to the constitution of the self. While the first part lays down the situational approach to emotion and the many ways in which the implicated self can be understood, the second part looks at some emotions (fear, love, pride and jealousy) in order to illustrate and elaborate on the conceptual map constructed in the first part. The article concludes indicating the open-ended character of both emotions and self.

14 

Introduction

Part III.  Cognition, Psychology, Neuroscience In ‘De Se Attitudes and Semiotic Aspects of Cognition’ Erich Rast explores the connection between so-called de se puzzles that are wellknown in the Philosophy of Language to Jackson’s Knowledge Argument. De se puzzles attempt to show that references of an agent to herself formulated in a 3rd-person perspective do not necessarily have the same explanatory power for the agent’s behavior than corresponding 1st-person self-ascriptions, while the Knowledge Argument attempts to refute physicalism by showing that the way a certain color feels, its phenomenal character, cannot be explained or emulated by mere physical knowledge no matter how exhaustive it may be. According to Rast, both puzzles need to be addressed from the perspective of the semiotics of cognition. Drawing from a computational metaphor he suggests a trivializing interpretation of the puzzles according to which actual thought tokens of one sort cannot be replaced by tokens of another sort because these play different roles in cognition. From this perspective the irreducibility asserted by both kinds of puzzles is ultimately the result of the (trivial) difference between actually cognating and explaining cognition. In ‘The Division of the Mind: Paradoxes and Puzzles’ Vasco Correia, starts by the claim according to which divisionist models of the mind argue that irrational phenomena such as akrasia and self-deception can only be understood if one assumes that the mind is somewhat differentiated in relatively autonomous sub-systems. Yet, the divisionist postulate seems to be intrinsically paradoxical in many regards. The article reviews some of the most influential divisionist models and argues that each of them leads to specific inconsistencies. It is not to suggest, however that the mind cannot possibly suffer any sort of partitioning, and even acknowledges this possibility in pathological cases of mental dissociation. Instead, Correia claims that the divisionist hypothesis is not necessary to account for ordinary cases of irrationality. This analysis relies on a unitary account of the mind which maintains that irrational attitudes typically derive from conflicts that take place between individual mental states (e.g., a desire and a belief), and not between differentiated parts of the mind. In ‘Empirical and conceptual clarifications regarding the notion of ‘Core-Self’ from Gallagher’s and Merker’s Behavioural-Neuroscientific

Introduction

15

Proposals’ João Fonseca addresses the problem of conceptual fragmentation brought about by the methodological and disciplinary diversity concerning current scientific studies on the Self. Fonseca focuses specifically on two different proposals regarding the nature of Core (or Minimal)-Self: Merker’s evolutionary perspective and Gallagher’s phenomenological one. Each one corresponding to very different proposals for the neural implementation of Core-Self: Merker’s brainstem/ sub-cortical and Gallagher’s pre-motor cortex suggestion. At a first approximation these two proposals seem mutually incompatible. By deploying a model-theoretic framework for theoretical concepts of behavioral neuroscience, Fonseca tries to uncover some of the underlying principles sustaining both proposals. Using the instrumental and fundamental notion of ‘Nested Concept’ within such framework, he shows how the two proposals can relate to each other. Fundamentally he shows that Merker’s and Gallagher’s suggestions of Core-Self (including their different proposals for its neuronal implementation) are related in a common nested conceptual relation formally defined. Such nested relation provides conceptual clarity and empirical unification into a hitherto fragmented and confused scenario. By bringing both proposals under the same formal conceptual framework Fonseca shows how to gain conceptual and taxonomic clarification, explanatory richness and bridges, both conceptual and empirical, between different disciplines and practices regarding current scientific studies on the Self. Part IV.  Ontology and Taxonomy The concepts of consciousness and self have been central in contemporary philosophy of the mind. Inevitably, this lead to the recuperation of a few conceptions from classical Phenomenology, starting with Husserl. It is the case of the concept of “pre-reflective self-consciousness”. The approach of these philosophers is not existential, but what could be called “biological” in the sense that they considered consciousness and self as natural phenomena, explained scientifically. One of the problems that these philosophers intended to resolve is the renowned problem of the self that was initially formulated by David Hume and more recently by Metzinger, among others. In his article Jorge Gonçalves, departing

16 

Introduction

from scientific data regarding the developmental origins of the self, argues that the concept of pre-reflective self-consciousness does not solve the problem. In spite of the facts not being conclusive, Gonçalves states that there are good reasons to reject the idea that the entire form of phenomenal consciousness assumes a feeling, no matter how small, of self. In ‘The Reality of the Virtual Self as Interface to the Social World’ Robert Clowes explores the idea and implications of the virtual self. This idea has so, up until now, been most associated with the philosopher Thomas Metzinger phenomenal self model hypothesis (Metzinger, 2004, 2009). Metzinger takes the idea to imply a “no self ” thesis (Metzinger, 2011). Clowes’ paper puts the idea against a background of virtualist representation (Clowes & Chrisley, 2012) and from here questions the metaphysical implications which are often drawn from the idea of the virtual self. Clowes’ paper starts with a review of the perplexing place of virtuality more generally in theorizing about mind. His paper then focuses in on the arguments Metzinger employs to argue for a virtual self, asking whether Metzinger’s approach can be cast into a broader virtualist framework. Arguing that indeed it can, Clowes then attempts to show that while Metzinger gives us an interesting way to think about the self that the conclusion he draws from them  about the non existence of self are unwarranted. In this context, the final part of the paper attempts to show that the virtual self may in fact be a useful way of making the self theoretically tractable for further scientific investigation including in the context of psychopathology. The paper concludes that the concept of the virtual self may not only be a useful theoretical tool but it may be real enough to supply the conceptual roles required in much theorizing around the self. Concluding that the virtual self does not need imply a “no self ”, Clowes concludes it may in fact be a useful way of unifying several current ideas about self. The virtualist view of self may be best thought of as a fruitful scientific reduction rather than an elimination as Metzinger argues. In his paper “Conceptual Personae of the ‘attentional self’” Alexander Gerner explores conceptual personae of the self in relation to the phenomenon of attention.  Philosophical concepts, as the concept of the self, can be personae that we think with and that emerge from

Introduction

17

a constitutive plane of philosophy in which they create concepts on this plane  that are more complex than a specific model  -of the selfor a specific mode of symbolic representation.  While Gerner takes up Thomas Metzinger’s claim that the strong first person perspective or “subjective self ”, defined as the centre of awareness, is the possibility of being able to manipulate the focus of attention in order to stabilize subjective experience, he proposes attention as a constitutive ground of the  self  that exceeds the  self-model theory. For Gerner the  self  as attentional self is less about stabilization of subjective experience but rather responsible for the constitutive imbalance of the self. Hence Gerner puts Metzinger’s thesis of the “control of the focus of attention” and the resulting complex notion of an attentional self into a new perspective through an approximation of  two possible conceptual personae of the  self: a) the impossible  attentional  self  in Paul Valéry’s ‘Monsieur Teste’ and b) the ‘heautoscopic’  attentional  self  in Italo Calvino’s ‘ Mister Palomar’. João Fonseca Jorge Gonçalves

References Clowes, R. W., & Chrisley, R. (2012). Virtualist Representation. International Journal of Machine Consciousness, 04(02), 503–522. Damásio, A. (1999). The Feeling of What Happens: Body and Emotion in the Making of Consciousness. San Diego: Harcourt Brace. Elliott, A. (2007). Concepts of the Self. Malden: Polity Press. Gallagher, S. (ed.) (2010). Oxford Handbook of the Self. Oxford: Oxford University Press. Gallagher, S. & Shear, J. (eds.) (1999). Models of the Self. Exeter: Imprint Academic. Gallagher, S. & Zahavi, D. (2008). The Phenomenological Mind: An Introduction to Philosophy of Mind and Cognitive Science. New York: Routledge.

18 

Introduction

Hofstadter, D. (2007). I Am a Strange Loop. New York: Basic Books. Kircher, T. (ed.) (2003). The Self in Neuroscience and Psychiatry. Cambridge: Cambridge University Press. LeDoux, J. (2002). Synaptic Self: How Our Brains Become Who We Are. New York: Viking Adult. Metzinger, T. (2004). Being No One: The Self-Model Theory of Subjectivity. Cambridge, MA: Bradford Book. Metzinger, T. (2009). The Ego Tunnel: The Science of the Mind and the Myth of the Self. New York: Basic Books. Metzinger, T. (2010). The no-self alternative. In Gallagher, S. (ed.), Oxford Handbook of the Self. Oxford: Oxford University Press. 279–296. Parnas, J. & Sass, L. (2010). The Structure of Self-Consciousness in Schizophrenia. In Gallagher, S. (ed.), Oxford Handbook of the Self, Oxford: Oxford University Press. 521–546. Perry, J. (2002). Identity, Personal Identity, and the Self. Cambridge: Hackett Publishing Company. Zahavi, D. (2005). Subjectivity and Selfhood: Investigating the First-Person Perspective. Cambridge: MIT Press.

Part I Metaphysics and Personal Identity

Animalism and the Remnant-Person Problem Eric T. Olson

Introduction Animalism clashes with the conviction that we should go with our transplanted brains. A good reply is that if animalism were true, we could explain easily enough both why the conviction is false and why it seems compelling. But another objection cannot be answered so easily. Animalism seems to imply that the detached brain would be a person who comes into being when the brain is removed and ceases to exist when the brain goes into a new head. And that seems absurd. The paper argues that, although this is equally problem for many views besides animalism, it has no obvious solution.

1. Animalism is the view that you and I are animals. That is, we are animals in the straightforward sense of having the property of being an animal, or in that each of us is identical to an animal – not merely in the derivative sense of having animal bodies, or of being “constituted by” animals. And by ‘animal’ I mean an organism of the animal kingdom.1 Sensible though it may appear, animalism is highly contentious. The most common objection is that it conflicts with widespread and deep 1

Many philosophers say that we “are” animals, but mean only that we are nonanimals constituted by animals. And some deny that human animals are organisms. For an example of both views, see Johnston (2007, pp. 49, 56).

22 

Eric T. Olson

beliefs about our identity over time. These beliefs are brought out in reactions to fictional cases. Suppose, for instance, that your brain is transplanted into my head. The being who ends up with that organ, everyone assumes, will remember your life and not mine. More generally, he will have your beliefs, preferences, plans, and other mental properties, for the most part at least. Who would he be – you, me, or someone else? Animalism implies that he would be me. That’s because the operation does not move a biological organism from one head to another. It simply moves an organ from one animal to another, just as a liver transplant does. One organism loses its brain and remains behind as an empty-headed vegetable; another has its brain removed and replaced with yours. (Or perhaps, as van Inwagen (1990, pp. 172–181) proposes, the naked brain would itself be an organism, and the empty-headed thing left over would be a mere hunk of living tissue, like a severed arm, owing to the brainstem’s role in directing a human organism’s life-sustaining functions. In that case, the operation would pare down an animal to the size of a brain and move it to another head, and the being who ended up with your brain and the rest of me would be you, even according to animalism. If this is right, the objectors must replace ‘brain’ with ‘cerebrum’. No one thinks an organism could be pared down to a naked cerebrum. I will ignore this complication in the sequel.) So if you and I are animals, I could swap my brain for yours. In that case I should suddenly acquire your knowledge, skills, and interests, and lose my own. I should lose all the memories of my past. In their place I should acquire memories of your past: of holidays I never took, people I never met, experiences I never had. My head would be filled with false beliefs: I should be convinced that I lived in your house, worked at your job, and was married to your spouse. I should think I was you. I should be systematically mistaken about who I am and how I fit into the world. As for you: if the operation didn’t kill you outright, it would cause you to lose all your memories, knowledge, plans, abilities – everything that matters. Unless, that is, you too got a new brain. If you got my brain, your lost memories would be replaced by memories of my life. You would be convinced, mistakenly, that you lived in my house, worked at my job, and were married to my spouse. That is what animalism implies about the transplant story. And it is easy to be unhappy with this description. In my experience, those

Animalism and the Remnant-Person Problem

23

presented with the thought experiment have an almost irresistible tendency to say that the one who got your brain would be you, not me. The operation would not give me your brain, but would give you my body. Transplanting your brain amounts to transplanting you. A brain transplant is not at all like a liver transplant. Call this conviction the transplant intuition. (What about me? Well, the intuition implies when the surgeons remove my brain to make way for yours, they remove me from my own head, just as they remove you from yours.) The objection, then, is this: Even if our brains are never actually transplanted, each of us has the capacity to go with one’s transplanted brain: to be pared down to a naked brain and moved to another head. But no animal has that capacity. Transplant an animal’s brain and the animal stays behind. It follows (by Leibniz’s Law) that we are not animals. To put the point another way: if your brain were transplanted, then according to the transplant intuition you would go one way, while the animal we call your body – the animal you would be if you were any animal at all – would go another. But it is impossible for a thing and itself to go their separate ways. If it is possible for you and your animal body to go your separate ways, it follows that you are one thing and the animal is another thing, numerically different from you. And there is no other animal or organism that you could be. It follows that you are not an animal, or any other a biological organism. You may still relate in some intimate way to an animal: you might have an animal as your body, or be constituted by an animal. But you are not yourself an animal.

2. Is this a strong objection? Well, here is a way of defending animalism against it. Suppose for the sake of argument that we were animals (hardly a wild conjecture). That would not only entail that the transplant intution was false, but it would explain in an unmysterious way why it was false. If we were animals, we should not go with our transplanted brains because no animal would go with its transplanted brain. An animal, even a human organism, simply loses an organ and gets an empty head

24 

Eric T. Olson

when its brain is removed. (At any rate that is a premise of the transplant objection.) And if we were animals, that is what would happen to us. Animalists can also explain why we should nonetheless find the transplant intuition attractive, even compelling, if it were false for this reason. Why might the transplant intuition seem true if we were animals? Why does it seem so obvious that the person who ended up with your brain would be you? How could generations of philosophers and their students have got it so badly wrong? Well, there would be strong grounds for accepting the transplant intuition even if it were false because we are animals. In fact, human animals would have the same reasons for supposing that they would go with their transplanted cerebrums as we have to believe that we should. There are two main grounds supporting the transplant intuition. First, the psychological and behavioural evidence that supports judgments about personal identity in familiar cases – judgments that are compatible with animalism – also supports the transplant intuition. Why do I suppose that the man who will wake up in my bed tomorrow (in normal circumstances) is me, and not a new person who came into being during the night? One reason is that he will have my memories, beliefs, preferences, and plans, or at least memories, beliefs, and plans that are causally dependent in a special way on my current mental states. In other words, he will be psychologically continuous, then, with me as I am now. What’s more, this psychological continuity will be (as Unger puts it) continuously physically realized in my brain, which will remain intact overnight. And no one in any real situation is ever psychologically continuous with someone without being that person: no one is ever psychologically continuous with someone else. In real life, someone’s being psychologically continuous with you at some time in the past or future is powerful evidence – probably conclusive evidence – for his being you. The person who ended up with your brain in the transplant story would likewise be psychologically continuous with you, and this continuity would be continuously physically realized. And there would be no psychological continuity between that person and me. In real life, this would be conclusive evidence for his being you. So it would hardly be surprising if this led us to believe strongly that the brain recipient would be you rather than me. And we should be no less inclined to think so

Animalism and the Remnant-Person Problem

25

if in fact he would not be you because (unbeknownst to us) you are an animal. The second sort of evidence supporting the transplant intuition is practical. The person who ends up with your brain may have “what matters in identity” for you. Before the operation you might have the same reason to care about the welfare of that person as you normally have to care about your own welfare. He might be morally responsible for the things you did before the operation. Everyone might be morally obliged to treat him just as if he were you; and he might be obliged to treat your children, spouse, and friends as if they were his own. And maybe the empty-headed being left behind would not have what matters in identity for you. Even if that being were subsequently given a new brain from some third party, you might have no more reason, before the operation, to care about his or her welfare then than you have to care about the welfare of any other stranger. In all actual circumstances, a person bears these relations of practical importance only to herself: it’s never the case that someone else has “what matters in identity” for you. So again, it would hardly be surprising if the fact that the person who got your transplanted brain would have what matters in identity for you inclined us to suppose that he would be you. And this inclination would be in no way diminished if the supposition were false because we are animals.2 These are all grounds for supposing that you and I and every other human person would go with his transplanted cerebrum. And they are grounds that we should have even if we were animals and it were false. In that case our mistake would be entirely understandable. Nor would there be the slightest mystery about why the transplant intuition was false. So it looks as if we have little right to any great confidence in the transplant intuition unless we already have a good reason to suppose that we are not animals.

2

“Fission” is another imaginary case where what is actually conclusive evidence for someone’s being you might be consistent with her not being you. If each half of your brain were transplanted into a different head, resulting in two beings psychologically continuous with you, it is easy to suppose that both would be you. Yet even many opponents of animalism concede that this cannot be, as there is only one of you, and one thing cannot be numerically identical to two things.

26 

Eric T. Olson

3. Even if this is a good defence of animalism, however, the brain-transplant story raises another objection that cannot be answered so easily. Think about your brain in mid-transplant, detached from your skull but kept alive. It is of course not alive in the way that an organism is alive. But it can be alive in the sense that its cells remain alive, like a liver awaiting transplant. It seems possible for the brain, in this condition, to support thought and consciousness. At any rate this is almost universally assumed, and I am not going to challenge it here. “Brain-in-a-vat” thought experiments are a staple in the diet of philosophy students everywhere, and it is part of the story that a detached brain could produce thought. (It may be that mental life is possible only if the cerebrum is stimulated in some special way by the brainstem, so that a naked cerebrum merely kept alive in a vat would have little or no mental life. In that case, if we are transplanting only the cerebrum, we must imagine it provided artificially with whatever stimulation it needs to support mental life.) But if the brain supports thought and consciousness, then there is a thinking, conscious being there – a being psychologically more or less like you. In other words, your brain, kept alive in a vat or whatever, would be a person. Or if it would not be a person itself, it would at least “realize” or “constitute” a person. Mark Johnston, to whom I owe this objection, calls it a remnant person.3 The possibility of remnant people is trouble for animalism. Animalism seems to imply that this remnant person would not be you. Otherwise it would be possible to pare down a human organism until it was nothing but a brain. But if an organism could be reduced to a detached brain (or cerebrum), then it could be transplanted from one head to another, and we have already ruled that out. (I will revisit the ruling in the next section.) The problem is not that we are convinced that the remnant person would have to be you, contrary to animalism. That would be the transplant intuition again, or at any rate an intuition epistemically 3

We might say that something is a remnant person at a time t just if she is a wholly organic person at t, she is not herself an organism or a thing constituted by an organism at t, and this condition is a result of cutting away a large portion of a normal human person at some time before t.

Animalism and the Remnant-Person Problem

27

indistinguishable from it, and animalism could be defended against it in the same way. Nor is the problem that the remnant person would not be an organism. That is perfectly compatible with animalism, which does not say that necessarily all people are organisms, but only that we are – we normal human people. (For all animalism says, there might be entirely inorganic beings who count as people in the sense of being rational, self-conscious, and so on: angels, for instance.) The trouble comes when we ask where the remnant person could have come from, if he or she could not be you. Surely he did not exist before the operation. Otherwise there would have been two people within your skin – the organism, who became an empty-headed vegetable, and the remnant person, who became a naked brain; and that is absurd. It looks as if animalists must say that the operation brings the remnant person into being. But that looks absurd. For one thing, it is evident that there are just two people in the transplant story, you and I, even if there is dispute about what happens to us. If the remnant person were someone new, there would be three. Or rather four: if removing your brain from your head creates one new remnant person, then removing my brain to make room for yours creates another. There are you, I, and the two remnant people created when our brains are removed. That’s two too many. More seriously, it is impossible to believe that removing someone’s brain from her head could create a new person. As Johnston puts it: You can’t bring a person into being simply by removing tissue from something…, unless that tissue was functioning to suppress mental life or the capacity for mental life. A developing fetus might have a massive tumor in its developing brain, which suppresses its mental life, and perhaps even its capacity for mental life. Given that, we can understand how removing the tumor could allow a person in Locke’s sense to be present for the first time. But how could removing a sustaining [head and] torso bring this about? (2007, p. 47).

If animalism implies that the transplant operation would bring a remnant person into being, it violates what we might call the creation principle: that you cannot bring a person into being merely by cutting away sustaining tissues. Animalism would violate a second and equally attractive principle as well, namely that you cannot destroy a person merely by surrounding

28 

Eric T. Olson

him with sustaining tissues: the destruction principle. Suppose we implant our remnant person into a new head – or, for that matter, into your own head again – hooking up the utilities in such a way as to make the resulting person more or less normal. According to animalism, this normally embodied person would be an organism. And no organism was ever a detached brain. The result of implanting a brain into an empty head is not that the brain comes to be an organism; rather, the organism simply acquires a new organ, just as it might acquire a new liver or kidney. Yet the remnant person does not cease to be a person when he or she is implanted. Nor does it come to be the case that there are two people within the same skin. It follows that the remnant person must cease to exist when he is put into a head. But providing a maimed person with the parts he was missing is a funny way of destroying him! So the new objection to animalism is that it is incompatible with the creation and destruction principles. You cannot create a person just by cutting away sustaining tissues, or destroy one just by providing them. Because animalism implies that you can create and destroy a person in these ways, it must be false. This is not just the original transplant objection put differently, and it demands a different response on the part of the animalist. Our being animals cannot explain why the creation and destruction principles are false. It may explain why none of us can come to be a remnant person, and thus why the remnant person who results from removing your brain from your head would not be you: the explanation would be that you are an animal, and no animal can become a remnant person. But our being animals cannot explain how the operation could have brought a person into being. The remnant person’s not being you is one thing; his coming into existence when your brain is removed is another. Likewise, our being animals may explain why the remnant person is not the normally embodied person who would result from putting your brain into my head: the explanation is that no animal could once have been a remnant person. But it cannot explain how implanting the remnant person into a new head could destroy him. His ceasing to exist is different from his not being identical to any normally embodied person. To explain how removing your brain from your head could bring someone into existence, and how replacing that organ could destroy him, we need an account of the metaphysical nature of remnant people. We need

Animalism and the Remnant-Person Problem

29

an account that would tell us, among other things, why remnant people come into being and pass away in such a surprising way – why they have such funny persistence conditions. But animalism offers only an account of our metaphysical nature: that of normally embodied human people. So even if animalism entails that the creation and destruction principles must be false, it cannot explain why they are false. The strange behavior of remnant people would remain a mystery. Call this the remnant-person problem.

4. I can offer no account of why removing someone’s brain from her head would bring a remnant person into being, or why that person would cease to exist if the brain were put back. If animalism really did imply that this would be so, it would be a serious objection. I think animalists should accept the creation and destruction principles, and deny that a brain transplant would create and then destroy a remnant person. How could that be? Who would the remnant person be, and how would he relate to the animal people, you and me? One suggestion is that despite appearances, the remnant person is you, the donor organism. Removing your brain or cerebrum from your head would not remove an organ from a human animal, leaving that animal with an empty head. And putting your brain or cerebrum into my head would not supply this animal – my body – with a new organ. Rather, the operation would pare an organism down to a naked brain and later supply it with new peripheral parts to replace the ones cut away. Of course, the brain or cerebrum in mid-transplant is not an organism, at least not then. But perhaps an organism is not an organism essentially, and can exist for a while as a nonorganism, just as a student can take a leave of absence and exist for a while as a nonstudent. This is logically consistent with animalism. Animalism is the view that we are organisms, not the view that we are organisms essentially. So maybe a human animal really would go with its transplanted brain. In that case, no remnant person would be created or destroyed in the operation. This would answer not only the remnant-person objection, but the transplant

30 

Eric T. Olson

objection as well. Both objections would be based on a false assumption about what it takes for a human animal to persist. Because this proposal implies that we are animals accidentally and not essentially, we might call it accidental animalism. Accidental animalism raises a number of worries. Perhaps the most obvious is that the empty-headed thing left behind after your brain is removed may still be alive – especially if the operation transplants only the cerebrum. That is, it might be a living organism. And if it were, it would seem to have the same life, as Locke would say, as the original organism had: the original organism’s life-sustaining functions would have continued uninterrupted throughout the operation in the large thing left behind, just as they would had the surgeons removed the liver rather than the cerebrum. And if an organism’s biological life carries on, how could it not continue to be the life of that same organism? Even if an organism could continue existing after its life comes to an end, how could an organism be outlived by its own life?4 Yet accidental animalism implies that this living, empty-headed organism would not be the original animal. That looks obviously false. Surely it is possible for a human animal to have an organ not essential to the maintenance of its life-sustaining functions removed, and to continue existing without that organ. If a human animal could not possibly exist even for a moment without its cerebrum, then we ought seriously to wonder whether any animal could exist even for a moment without its liver or kidney or appendix. It would be a real epistemic possibility that an appendectomy reduces an animal to the size of an appendix, while the living organism left behind is not the survivor of the operation, but something new. For that matter, accidental animalism creates a new version of the problem it was meant to avoid: a “remnant-animal problem”. It implies that removing your brain would create a new animal – the one with an empty skull. And replacing your brain with a new one, or even rehousing it in its original head, would cause this remnant animal to cease to exist, as the resulting “whole” animal would be the former brain rather than the former brainless animal. But surely you cannot bring an organism into being simply by removing tissue from something, unless (as Johnston 4

For a defence of Locke’s view that the continuation of an organism’s life is both necessary and sufficient for the organism to persist, see van Inwagen (1990, pp. 142–158).

Animalism and the Remnant-Person Problem

31

would say) that tissue was functioning to suppress life or the capacity for life. Perhaps there could be tissues that were capable of developing into or coming to make up a living organism but for the presence of a tumour or other foreign body. In that case removing the tumour might bring an organism into being. But removing the cerebrum of a normal, healthy human being cannot do this. Nor can you destroy an animal living without a cerebrum or other nonvital organ simply by providing it with such an organ. These principles, surely, have the same force as the original creation and destruction principles. Accidental animalism avoids the objection that remnant people would have absurd persistence conditions by proposing that organisms have them. Nothing is gained.

5. Here is a better proposal: The remnant person who would result from removing your brain from your head would be simply your brain – that is, the thing that is now your brain. The remnant person is never an organism, and the operation doesn’t make him any bigger or smaller. Removing your brain from your head does not bring a person or any other material thing into being, and putting it into a new head does not destroy anyone or anything. Plausible thought this may be, it raises an obvious pointed question. We are supposing that the remnant person can think and be conscious while he is detached from the rest of you. If the remnant person is your brain, this means that your brain could think and be conscious while detached. The pointed question is this: does your brain think now, in its normal surroundings? Is it now conscious? I don’t mean whether it “thinks” in some attenuated or derivative sense – in the sense of being the organ responsible for your thinking, say. The question is whether it thinks in the strictest and most straightforward sense. There is no very attractive answer to this question. Suppose your brain does now think in the strictest sense. But we are supposing that you are not your brain. You are not, as things are, a three-pound, yellowish-pink organ located entirely within your skull.

32 

Eric T. Olson

Rather, you are the animal your brain is a part of. And you also think in the strictest sense. Surely it couldn’t be the case that you think only in a derivative sense: that you think only insofar as you have a part that thinks strictly speaking. As Chisholm said, if there are now two things thinking your thoughts, one doing it on its own and the other such that its thinking is done for it by something else, you are the one that thinks on its own (1976, p. 104). Or maybe Chisholm was wrong about this, and there is no problem in saying that we think only in the derivative sense of having a part that thinks strictly speaking. In that case, we really can solve the remnant-person problem by saying that the remnant person would be your brain. Your brain thinks now, and is the only real thinker there. If it were removed from your head and kept alive in a vat, it would continue to think – though it would then think only for itself, and not for you. Presumably your brain would count as a person while it was detached, but not now while it remains a part of you, even though there would be no change in its mental capacities. Many philosophers will have no objection to this. Orthodox four-dimensionalism – the view that all persisting things including ourselves are composed of temporal parts – has a similar consequence: it implies that you think now only insofar as a part of you – your current stage – thinks strictly speaking. And although that stage is not in fact a person, it would be were it not surrounded by other stages psychologically continuous with it, even though there would be no difference in its mental capacities. Yet almost no one takes this to be an objection to four-dimensionalism.5 But let us suppose that Chisholm was right: if anything thinks your thoughts in the strictest sense, you do. If your brain now thinks, then so do you. It would follow that every normal human person is accompanied by another being psychologically indistinguishable from her. This article would have two authors, I and my brain, and there would be at least two conscious beings now sitting there reading it. On the Lockean assumption that a person is by definition a being with certain mental properties – rationality and self-consciousness, say – your brain would be a person. You would be one of two people now thinking your thoughts. The transplant operation would separate them, turning one 5

Olson (2007, pp. 122–125) is an exception.

Animalism and the Remnant-Person Problem

33

into a remnant person and the other into a brainless vegetable. You ought to wonder which person you are. Any grounds you might have for supposing that you are the animal person rather than the brain person would seem to be grounds for the brain to suppose that it is the animal person rather than the brain person. How could you know that you’re not making this mistake? Likewise, you ought to wonder whether you are the one who would go with your brain if it were transplanted, or the one who would stay behind with an empty head. Well, suppose your brain does not now think in the strictest sense. But it would think if it were detached and suitably cosseted. This seems to imply that your brain is now prevented from thinking by its fleshy surroundings. Normally it is a mere brute organ, no more sentient or intelligent than a kidney; but remove it from its natural habitat in the right way and it will blossom instantly into a mature philosopher. And putting it back where it belongs would deprive it of these new-found intellectual capacities and restore it to its former state of total oblivion (something that is normally a serious crime). So the sustaining tissues surrounding the brain really do “suppress mental life or the capacity for mental life”. They may not suppress mental life altogether: they don’t suppress it in the organism. But they suppress it in the brain, by preventing the brain from having its own mental life. Yet surely, we want to say, you can’t give something the capacity for thought and consciousness merely by cutting away sustaining tissues; nor can you deprive something of that capacity just by surrounding it with such tissues. That looks just as compelling as the original creation and destruction principles. So if your brain thinks now, there are too many thinkers; if it doesn’t, things can gain or lose mental capacities in an utterly baffling way. That’s the trouble that comes of saying that the remnant person would be your brain. We might call this trouble the remnant-brain problem.

6. However grave the remnant-brain problem may be, it is no reason to doubt whether we are animals. It is not a problem for animalism in

34 

Eric T. Olson

particular, but applies equally to almost any view about what we are. I say almost any view because you could avoid the problem by saying that we are brains: that each of us is literally a three-pound lump of tissue housed within the skull. In that case your brain thinks even now, and removing it from your head would do nothing to enhance its mental capacities. The operation would merely change your surroundings. What about the animal – your body? Doesn’t it think, giving us too many thinkers? Well, if you are your brain, that is presumably because your brain is the only thinking being there. Anything bigger than a brain has mental properties only in the derivative sense of having a brain that has them strictly speaking. This would solve the remnant-brain problem. But no philosopher that I know of thinks that we really are brains.6 If you are not your brain, then our pointed question returns: does your brain now think – really think, in the strictest possible sense? If it does, then there are now two beings thinking your thoughts, your and your brain. On the assumption that a person is a being with such mental properties as rationality and self-consciousness, you are one of two people now thinking your thoughts. You ought to wonder which one you are, and how you could ever know. Or maybe your brain doesn’t think. But it would if it were removed from your head and suitably cosseted. This means that your brain is now prevented from thinking by its fleshy surroundings. And putting it back where it belongs would presumably deprive it of its power to think and restore it to its former state of total oblivion. The remnant-brain problem does not arise only if we are animals. It arises if we are anything bigger than brains. Someone might propose that your brain could never think, even when removed from your head. At most it might constitute a thinker: its matter would make up a thinking being other than your brain itself. Your body – an organism – constitutes you now, but if your brain were removed from your head, that organ would then constitute you, or at any rate it would do so for as long as it continued to realize your psychology. (This is Johnston’s view, and presumably that of most other “constitutionalists”.)

6

Though Hudson (2001, p. 143) says that each of us is a temporal part of a brain. I discuss the view that we are brains in Olson (2007, pp. 76–98). I suppose one might also avoid the remnant-brain problem by saying that we are Humean bundles of impressions, or immaterial substances.

Animalism and the Remnant-Person Problem

35

But this only relocates the problem. Constitutionalists do not think that your brain now constitutes you, or indeed any other thinking being. You do not weigh three pounds. You – not merely your body, but you yourself – extend all the way out to your skin. It is your animal body that now constitutes you, not your brain. So although removing your brain from your head would not give it consciousness or the power to think, it would give it the power to constitute a conscious, thinking being. Your brain is now prevented from constituting a thinker by its fleshy surroundings, and putting it back where it belongs after its removal would prevent it from doing so once more. Those sustaining tissues “suppress mental life or the capacity for mental life” insofar as they prevent the brain from constituting a psychological being. And that seems absurd. This is just the remnant-brain problem in a different form. As far as I can see, any explanation of why a brain could constitute a thinking being when isolated but not in its natural habitat would serve equally well as an explanation of why a brain could think when isolated but not when embodied. It may be, for instance, that a brain cannot in its normal surroundings constitute a thinker because it is a then part of a larger being that constitutes a thinker, namely the animal. In that case, animalists can say that, although the brain could think when isolated, it cannot think in its normal surroundings because it is then a part of a larger being that thinks.

7. So the remnant-brain problem is not a worry for animalism in particular, and simply denying that we are animals does nothing to solve it. If constitutionalists could explain why your brain would constitute a thinker when detached from the rest of you but not in its normal surroundings, then animalists could explain in the same way why your brain would think when detached but not otherwise. If animalists cannot explain why removing your brain would enable it to think, then constitutionalists cannot explain why removing your brain would enable it to constitute a person. If the remnant-brain problem is a reason to deny that we are animals, it is equally a reason to deny that we are material things constituted by animals.

36 

Eric T. Olson

For all that, I have done nothing whatever to solve the remnant-brain problem, or the more general remnant-person problem. Can it be solved – other than by saying that we are brains, or some other desperate ploy? I haven’t much hope for a “psychological” solution – that is, an account of why human brains should have the power to think (or to constitute a thinker) when detached but not otherwise. Another possibility is to deny that there are such things as undetached brains. There are of course particles “arranged cerebrally” within your skull; but they don’t compose anything. Nor would they compose anything when outside your head. There are no undetached brains, and no detached ones either. There are no remnant people. There would be no remnant people even if brain transplants really occurred. So there is nothing in the transplant story whose surprising inability to think needs explaining and the question of how the transplant operation could create and then destroy a person does not arise. We might call this brain eliminativism. If you think it sounds crazy, I don’t blame you. But it may be no worse than the alternatives, including rejecting animalism. In any case, I will devote the rest of this paper to exploring it. The obvious question it raises is why particles arranged cerebrally never compose anything. Why should there be human animals but no human brains? The only way to answer this question is to work out when any particles compose something. How, in general, do smaller things have to be arranged and situated for them to compose or add up to something bigger? There are two “extreme” answers to this question. One is compositional universalism: any things, no matter what their nature or arrangement, compose something. The other is compositional nihilism: no things ever compose anything. There are no composite objects, but only mereological simples. Nihilism is obviously incompatible with animalism, since no organism is a simple. And for reasons I have given elsewhere (2007, pp. 229–232), animalism is not easily combined with universalism. Animalists need to say that some things compose something and others don’t – which is what most of us probably thought anyway. But which ones, and why? Very few answers to this question have been proposed. The best answer I know of is van Inwagen’s (1990, pp. 81–97): those things compose something if and only if their activities constitute a biological life. This implies that the only composite objects are living organisms. I cannot defend this view

Animalism and the Remnant-Person Problem

37

here. But it would explain why there are no remnant people: because a remnant person is by definition not an organism.

8. Whatever its merits, brain eliminativism raises deep issues. Even if the particles arranged cerebrally in the vat would not compose anything, they might still manage to produce thought. At any rate, what went on inside the remnant person (to speak loosely as if there were such a thing) would be indistinguishable from what goes on within a normal human brain. So I have been assuming, anyway. Since you are a philosopher, the remnant person resulting from removing your brain would also be a philosopher. Or rather, those particles arranged cerebrally would collectively produce philosophical thought. This would be thought without a thinker. That possibility might be epistemically troubling. Suppose the particles in the vat were collectively to produce the following line of argument: I could be wrong in thinking that I have hands, or that there are other people, or that anything existed five minutes ago. But even so, I can be sure that I am now thinking. Nothing could mistakenly think that it was thinking. And if I am thinking, then surely I must exist. I could never be mistaken in thinking that I exist. So my own existence, at least, is certain.

According to brain eliminativism, the conclusion of this reasoning would be false. Even if it is logically valid, and anyone who thinks that she exists (or who thinks anything at all) really does exist then, it would not be true in this case that anyone was thinking. The word ‘I’ would not refer to anything. Or at least it would not refer to anything that was thinking. At most it might refer in the plural to the particles jointly producing the thought. But no particle thinks. Of course, in this case no one would be mistaken in giving the argument, since no one would be giving it. But the argument would be unsound, because the premise ‘I am thinking’ would be false.

38 

Eric T. Olson

If this is possible, then I ought to wonder whether I exist. I feel pretty certain that I do. Why? Well, if I didn’t exist, how could I be sitting here writing this? How could I even raise the question of whether I exist? But the particles in the vat could produce the very same reasoning, yet the conclusion would be false. What reason have I got, then, to suppose that my reasoning is not unsound for the same reason – because my thoughts are produced by particles that don’t compose anything? However obvious it may appear to me that I am now thinking, particles arranged cerebrally in a vat could (collectively) find the thought ‘I am thinking’ equally obvious, though it would be false. So for all I know it’s false in my case too. The worry does not require us to take seriously the possibility that I might be a remnant person – that evil surgeons might have snatched Olson from his bed last night and put his particles arranged cerebrally in a vat, and that those particles are the ones producing these thoughts. That scenario is farfetched, to say the least. Maybe we can legitimately ignore it. Even so, brain eliminativism seems to imply that it is possible, in the brain-transplant case, for particles collectively to produce thought just like mine without thereby composing any thinking being. That raises the worry that it might be possible in other cases too, even in ordinary ones. It deprives brain eliminativists of what seems the best argument for supposing that there are any thinking beings, namely that there is thought and that thought requires a thinker. Presumably it would undermine any other argument for the existence of thinking beings in the same way. If this is the way to solve the remnant-person problem, it comes at the cost of undermining our belief in our own existence: hardly a welcome result.

9. Or maybe there could not be thought without a thinker, and particles arranged cerebrally in a vat would be unable to produce thought. (This is van Inwagen’s view [1990, p. 118f.].) Even if they could produce something intrinsinsically indistinguishable from thought, it might not be genuine mental activity: no beliefs, desires, reasoning, or conscious states. Maybe there is nothing that it would be like to be a remnant person, and nothing

Animalism and the Remnant-Person Problem

39

it would be like to be in the states that particles arranged cerebrally in a vat could produce. That might enable me to know that I exist, and that my thoughts are not produced collectively by particles that compose nothing. For I can know that thinking is going on. There seems to be thinking going on, and this seeming is itself a sort of thought. If my particles did not compose anything, as the particles arranged cerebrally in the vat would not, then they could not collectively produce any seeming, and so it would not seem to me that I was thinking. That it seems to me that I am thinking (or that there seems to be thinking going on) would entail that it seems that way to something, and hence that something thinks: me. That might enable me to know that I exist. This would mean that remnant people, or more precisely particles arranged cerebrally that are not parts of an organism, would be able to produce only pseudo-thought and pseudo-consciousness, though these states might be intrinsically identical to real thought and consciousness, and would have similar causes and effects to those of real thought and consciousness. Remnant people would be able to have conversations and write philosophical essays – or at least they could interact in ways outwardly indistinguishable from conversations, and collectively produce philosophical essays. There could be many remnant people hooked up by wireless links to robotic “bodies” whose movements were indistinguishable from those of human beings. There could be a vast society (as it were) of such remotely controlled robots, behaving and interacting in a way indistinguishable from real people, or at least as much like it as their inorganic anatomy allowed: working, chatting, quarrelling, publishing scholarly books, devising arguments for their own existence… It would look for all the world like a community of thinking, conscious beings, yet there would be no thought or consciousness at all. What enables me to know that I exist may make it hard to know whether anyone else does. The possibility that unthinking beings might produce every outward appearance of thinking is nothing new. It seems also to follow from the claim that artificial intelligence is impossible – that no inorganic digital computer, no matter how powerful or cleverly programmed and no matter how it interacted with its environment, could have real consciousness or intentionality. For all anyone knows, it is physically possible to build self-contained inorganic robots – not controlled by remnant people in vats – whose outward behavior and interactions

40 

Eric T. Olson

would be indistinguishable from those of human beings. There could be a vast society of such beings. If they could not really be intelligent, then here too no amount of behavioural evidence – and no amount of “intelligent” behaviour or dispositions so to behave – would guarantee the existence of thought or consciousness. We might even doubt whether there is any importance difference between real thought and consciousness, which may require a subject, and pseudo-thought and pseudo-consciousness, which don’t. Who cares whether the particles arranged cerebrally in the vat, or the robots that behave just like ourselves, could think or only pseudo-think? For that matter, how could I ever know whether I am now thinking or only pseudo-thinking? I proposed that if it seems to me that I am thinking (or that thinking is going on here), then I really am thinking, because seeming is itself a kind of thinking. But can I distinguish pseudo-seeming from genuine seeming? That would again cast doubt on my knowledge of my own existence. Perhaps a good question to ask is whether pseudo-thought and pseudo-consciousness would have any normative status. I have a reason to promote the satisfaction of my desires and to avoid pain, and I have a moral duty to do the same for others. Have I got a reason or a duty to promote the satisfaction of pseudo-desires and to minimize pseudo-pain? I wish I knew the answers to these hard questions, but I don’t. Brain eliminativism may well raise more problems than it would solve.

References Chisholm, R. (1976). Person and Object. La Salle, IL: Open Court. Hudson, H. (2001). A Materialist Metaphysics of the Human Person. Ithaca: Cornell University Press. Inwagen, P. van (1990). Material Beings. Ithaca: Cornell University Press. Johnston, M. (2007). “Human beings” revisited: My body is not an animal. In D. Zimmerman (ed.), Oxford Studies in Metaphysics 3. Oxford: Oxford University Press. Olson, E. (2007). What Are We? New York: Oxford University Press.

Will I ever be a Cyborg?1 Rui Vieira da Cunha*

1. Introduction Eric Olson’s animalist view relies on the premise that person is not a fit candidate to be a substance concept, in Wiggins’s terminology. Instead, he claims, animal is what best serves as the answer to what we most fundamentally are and what determines our persistence conditions. Proposing a thought experiment concerning inorganic replacement, I aim to show that Olson’s animalist view cannot accommodate our very strong intuitions about such cases. My claim is then that animalism either fails on its own grounds or requires some tuning regarding what exactly an organism is and its persistence conditions. I will examine Matthew Liao’s attempt to accommodate such intuitions within an animalist view. I will also describe the basics about Olson’s animalist view but let us first look at the rough outline of my thought experiment. The essentials are as follows, the rest I will provide later, as we go along.

1

*

For comments on an earlier draft of this paper, I am grateful to Arto Laitinen, Sofia Miguens, Mikko Yrjönsuuri, and, most of all, Eric Olson, whose patience and detailed comments were more than I could hope for. Researcher in the Mind Language and Action Group – MLAG – of the Institute of Philosophy, University of Porto – FCT grant number FCT – SFRH/BD/45701/ 2008. Jyväskylä, November 2009.

42 

Rui Vieira da Cunha

2.  A Cyborg2 Thought Experiment Suppose the year is 2020. Tom has just been born. He is a human animal (or just plainly an animal or a member of the species Homo sapiens or an organism). Now suppose, for argument’s sake, that during his career Tom will undergo transformations of the kind that transhumanists envisage as possible (and even likely and desirable) in the future. And let us leap further into the future to describe such transformations. In 2060, Tom has become an enhanced human. He has nanobots running through his bloodstream that fight infections and aging processes and monitor his life functions. Most of the matter that was ever part of his body has been replaced by inorganic parts, far more efficient and enduring. In short, he has technology of the sort some futurists claim to be possible in the very near future and more sceptical scientists assume possible in a more distant future (Bostrom, 2003). We can say that Tom has almost become a full cyborg. Almost every part of his body has been replaced by artificial parts. But not all the parts: Tom’s brain (including his brainstem) has not been replaced. In 2090, Tom is about to undergo another transformation. There are now inorganic brains available and these come with inorganic brainstems, that function as control and coordination centres much in the same way organic brainstems have always functioned for human animals: they regulate the body’s metabolism, the capacity to breathe and circulate blood, etc. The procedure is by then common to everyone: Tom’s brain and brainstem will be gradually replaced, bit by bit and not all at once. At no point in the procedure will Tom’s vital functions ever be interrupted. Whatever memories or psychological features Tom may have, they will be reproduced exactly in the inorganic brain. Nevertheless, Tom’s friends are worried. They wonder if he will survive in the process. 2

I would like to thank Mimosa Pursiainen for pointing out that I am not using the word cyborg in a technical sense. If we take a cyborg to be something partly biological and partly inorganic, then we can only consider Tom a cyborg in the first stage of our experiment, that is, in 2060. By 2090, after the last surgery, Tom will be fully inorganic and he won’t be a cyborg. Still, for practical and stylistic reasons, I chose to go with the word cyborg when I refer to Tom, both in 2060 as in 2090.

Will I ever be a Cyborg?

43

Tom’s friends are worried because they believe Olson’s animalist view to be true. They believe he has persisted through such a long time as the same animal because his life functions have been assured by that brainstem. And they now fear that undergoing the procedure may cause him to cease to exist. In order to fully grasp everything I am pursuing with this thought experiment, let us stop for a while and recall Olson’s animalist view and consider how this thought experiment and the resulting intuitions would be explained on that view.

3.  Olson’s Animalist View Eric Olson’s defence of the animalist view on matters of personal identity rests on the claim that animal is a more adequate concept than person to be a Wigginsian substance concept. Based on his reading of Wiggins (1980, p. 15), Olson assumes that every particular object falls under some kind or concept that tells us, in a special sense, what the object is, and not merely what it does or where it is located or some other accidental feature of it. And that concept determines persistence conditions that necessarily apply to all (and perhaps only) things of that kind. Concepts of this sort are substance concepts. (1997, p. 28)

Thus, a substance concept 1) tells us what the object is, as opposed to merely telling us what it does, and 2) determines persistence conditions that necessarily apply to all things of that kind. Moreover, we can infer that two objects falling under the same substance concept will share the same persistence conditions and that any object that falls under a substance concept will always have those persistence conditions, i.e., “a thing cannot change its criterion of identity partway through its career” (Olson, 1997, p. 29). Substance concepts, according to Olson, can be distinguished from what Wiggins calls phase sortals, or sorts, such as child or athlete or philosopher. These concepts do not tell us what the object most fundamentally is but rather what it is at some point of its existence.

44 

Rui Vieira da Cunha

Furthermore, an object can undergo a change during its existence such that it continues to exist but no longer falls under that phase sortal. To take the example provided, something that is an athlete comes into existence before he or she is an athlete and can cease being an athlete without ceasing to exist. As Olson puts it, “there are such things as former athletes and potential athletes. Athletes don’t have the criterion of identity that they have by virtue of being athletes.” (1997, p. 30). According to the Biological Approach supported by Olson, our substance concept, the concept that best answers the question about what we most fundamentally are, is not person but rather Homo sapiens or animal or living organism. In his own words: “Animal (or organism or human animal) is a paradigm case of a substance concept, and so is an ideal candidate for determining a thing’s persistence conditions.” (1997, p. 36). Olson’s rejection of person as a substance concept is based both on the fact that personhood “is merely a capacity or ability of a thing” (1997, p. 35) and not something that is closely connected with that thing’s “internal, structural, or intrinsic features” (1997, p. 34)3 and on the fact that biological entities like us can persist under the same conditions as many non-people, such as human embryos and human vegetables. A human embryo or a human foetus or a human vegetable cannot be counted as persons but they are still human animals and so, Olson argues, that is what we most fundamentally are: animals. Saying that we are animals is an excellent answer to the question what something is. Person, on the other hand, is but a phase sortal, in the sense that to say that something is a person doesn.t tell us what something is but rather what it does and also in the sense that something (in our case, a human

3

One may of course wonder why a thing’s capacity or ability isn’t closely connected to its intrinsic features. Moreover, one can doubt, along with Nichols (2010), the grounds for the distinction between a thing’s structure and its capacities. In fact, Nichols (2010) has objected to Olson’s line of reasoning on the grounds that (1) human animal, animal, and organism are all functional concepts, and (2) the distinction between what something is and what something does is illegitimate in the reading that Olson’s argument needs. Since my point here is to take the animalist view at face value and argue that it faces a problem concerning inorganic replacement, I will not argue against such assumptions here.

Will I ever be a Cyborg?

45

animal) can exist before coming to be a person and can continue to exist (as a human animal) even after ceasing to be a person4. One aspect of Olson’s account that should be evident by now is his interchangeable use of the terms animal, human animal, and organism. One could also include living organism in this list but it is less often used and, considering that Olson deems life to be a fundamental characteristic of organisms5, it would be redundant to do so, since there could be no non-living organisms. Or so he claims. For the time being, I shall also use those terms interchangeably, although later on some specification will need be made. When attempting to accurately describe the Biological Approach, Olson is clear on a number of issues that concern us. First, it does not exclude other kind of persons – wholly immaterial, like Gods or angels or Cartesian Egos, or even material persons, made “out of nuts and bolts, or wires and diodes” or of other biological species (1997, p. 124). And, of course, it does not exclude that these different kinds of persons will have different persistence conditions (1997, p. 27, p. 124f.); in fact, it implies so. It does exclude, however, that an organism could ever come to be a non-organism or a non-biological organism (1997, p. 125) – and that is something to be explored when we return to my thought experiment. Another important reminder when it comes to Olson’s animalist view is that, as he warns us, it must not be confused with the Bodily View that states that “we are identical with our bodies, or that we persist just in case our bodies continue to exist” (1997, p. 19) or with any modified version of it that would rely on some sort of physical criterion of personal identity, that would somehow focus on the brain and answer any question just by looking where the brain is6. 4 5

6

Olson further illustrates the inadequacy of person as substance concept by means of an analogy with a “Locomotive Criterion” of identity (1997, p. 31). Olson (1997, p. 136): “This proposal entails that an animal necessarily ceases to exist when it dies. In that case there is no such thing as a dead animal, strictly so called. We may call something lying by the side of the road a dead animal, but strictly speaking what is lying there are only the lifeless remains of an animal that no longer exists”. Olson (1997, p. 144) gives a number of reasons against the idea that body could be a substance concept – the fact that an object could be someone’s body for a while and then someone else’s body and continue to exist, for instance. Or the fact that an object can cease to be a human body and continue to exist. The main

46 

Rui Vieira da Cunha

Olson’s Biological Approach makes two main claims: first, that we are animals, members of the species Homo sapiens. Second, that psychological continuity is neither necessary nor sufficient for a human animal to persist through time. Now, the first claim does not seem something that anyone would want to deny. At least, in a weak sense, everybody agrees that we are animals or that we are constituted by an animal or that we stand in some special relation to an animal. What some may be tempted to deny is the stronger sense of that claim, which in fact is exactly the one Olson endorses. In that stronger sense, when we say that you are an animal, we are not just saying that you are constituted by an animal or that you have the body of an animal but that you are numerically identical with an animal, that you are essentially an animal. To be accurate, this should actually count as a further claim, or as a special qualification of the claim that we are animals, since this is in fact the claim that we are essentially an animal. The is in “Tom is an animal” is not the is of constitution but an is of identity. And if that is correct, then we have our persistence conditions by virtue of being animals, members of the species Homo sapiens. Once the first claim is correctly understood in its stronger sense, in this qualified form, the second claim – that psychological continuity is neither necessary nor sufficient for a human animal to persist through time – seems more easily acceptable. Now, for Olson to enunciate those persistence conditions, he needs to provide a more detailed account of what an animal is. This is the point where it seems to me that Olson’s interchangeable use of the terms animal, human animal, and organism requires some specification. In the only two direct statements of identity conditions Olson provides us with, one refers to animal7 and the other to organism8. However, it seems clear to me that he is considering the same question in both of them and although he constantly shifts between these terms, he seems

7

8

issue, however, is put more clearly by Olson (2007, pp. 25–26): “I have never seen a good account of what makes something someone’s body […]. I am unable to complete the formula ‘necessarily, x is y’s body if and only if…’”. “If x is an animal at t and y exists at t*, x=y if and only if the vital functions that y has at t* are causally continuous in the appropriate way with those that x has at t.” (1997, p. 135). “For any organism x and any y, x=y if and only if x’s life is y’s life” (1997, p. 138).

Will I ever be a Cyborg?

47

to show some preference for organism. I will then assume that, even though Olson uses either of those terms interchangeably, it is organism that somehow best expresses what he has in mind9 and it’s an organism’s persistence conditions he is looking for. In doing that, ultimately he relies on biology and so his account is scientifically informed. Metabolism, teleology, and organized complexity are the relevant features of a living organism. What about the identity condition of these organisms, then? In the process of explaining what an organism life consists in, Olson constantly refers to the brainstem, “the organ that is chiefly responsible for directing your life-sustaining functions” (1997, p. 140). That is the reason I mentioned it when I presented my thought experiment and since it is now time to get back to that, I will leave the considerations on the brainstem’s importance for the next section of this paper. Given this description of Olson’s animalist view, I think it can now be asked what it is about my thought experiment that may counter this view. You might be asking yourself what is the problem of a human animal turning into a cyborg. Since the animalist only cares if there is still a human animal living, even if the cyborg were to retain Tom’s psychological characteristics (his memories, habits, and so on), the only question would be whether he is still a living animal or not. Perhaps the animalist can simply reply that the enhancements Tom underwent in 2060 have simply changed some of his features but his life functions have persisted and so has the organ that directs them. And if Tom’s brain and brainstem were to be replaced by an inorganic device in 2090, then the human animal known as Tom would cease to exist. The fact that the cyborg resulting from the procedure could have the same psychological properties Tom had is not a problem for the animalist view, you could 9

I find some further comfort for this view in Olson’s later statement that “[…] animals, including human animals, have more or less the same metaphysical nature as other biological organisms. This is not to deny that some animals may have properties of considerable metaphysical interest – rationality and consciousness, for instance – that no plant or fungus could ever have. But if we ask what organisms are made of, what parts they have, whether they are concrete or abstract, whether and under what conditions they persist through time, and the like, I believe that the answer will be more or less the same for human organisms as it is for plants and fungi. So we need an account of the metaphysical nature of organisms generally.” (Olson, 2007, p. 27).

48 

Rui Vieira da Cunha

be tempted to reply. The resulting being would simply be something like Tom’s copy. It would not have his biological life. Perhaps now is a good time to disclose one essential fact of my thought experiment. If you read it again, you will notice I never mentioned Tom’s choices or thoughts. In fact, nothing even remotely close to Tom’s actual mental life. That is because in 2059 Tom fell into a persistent vegetative state. Does that make a difference?

4.  Back to the Cyborg I think it is clear by now why I decided Tom should have such an unfortunate event in his future. Had I presented the thought experiment in the usual manner, granting that Tom would still have higher mental functions such as memory and reasoning even after becoming a cyborg, the focus would be on the necessity of psychological continuity or connectedness in the matter of personal identity. However, I have stated that my aim is to show that the animalist view either fails on its own grounds or requires some tuning regarding what exactly an organism is and its persistence conditions. Unfortunately for Tom, this purpose is best served if he is in a persistent vegetative state, maintaining all of his vital functions throughout our experiment. So, what can we say about the events in Tom’s imagined future and the way they could – if indeed they could – affect our judgement concerning his numerical identity? What plausible intuitions, if any, arise from those events? And, more importantly, can the animalist view accommodate such intuitions? Consider Tom in 2059, when he lapses into a persistent vegetative state. Let’s assume it is an irreversible one: Tom’s cortex is damaged beyond repair, even if his brainstem is untouched. After the enhancement surgery, we could practically call him a full cyborg, even if his brain (still damaged) and his brainstem (untouched) are not replaced. The most plausible intuition would be to say that Tom survived such a process. That is, of course, if we grant that Tom still persisted after lapsing into a vegetative state. Even if some of us would say that he did not, I believe they would probably say this because of the lack of

Will I ever be a Cyborg?

49

psychological features. But in that case it would be the audience who would reject Olson’s view to begin with. Olson’s approach grants absolutely no importance to those features. Its main characteristic is that it puts biology in the place of psychology (Olson, 1997, p. 16). That is not the audience I am seeking. I want to show that we can be persuaded by the animalist view and still have some very strong and plausible intuitions about this thought experiment that that view cannot explain. And I also believe one of the strongest points for the animalist view is that it grants that in a case such as Tom’s in 2059 (Olson’s Human Vegetable Case) you would persist. The human animal that you are would not cease to exist just because its higher mental functions are gone. I think that is a very strong intuition most of us will share with the animalist view. Let us now turn to Tom’s state in 2090. Could the severely mentally debilitated Tom survive the last surgery? Would it still be Tom after that procedure? I believe we can say so. You may object that such a process of inorganic replacement is neither possible nor conceivable. These procedures do not strike me as impossible or inconceivable. To put it as Parfit does (1984, p. 219)10, I would deem them technologically impossible but not deeply impossible. Regardless of what our position on thought experiments and their validity in drawing conclusions in philosophy, it seems this one is not more far-fetched than any other commonly described in addressing personal identity: Olson himself considers the possibility of inorganic replacement – and even in a situation where consciousness and thought is maintained during the process, as we shall see. If we can consider that, my thought experiment must be a lot easier 10 “It may be impossible for some of these cases to occur, whatever progress may be made in science and technology. I distinguish two kinds of case. Some cases contravene the laws of nature. I call these deeply impossible. Other cases are merely technically impossible. Does it matter if some imagined case would never be possible? This depends entirely on our question, on what we are trying to show. Even in science it can be worth considering deeply impossible cases.” I cannot be so bold as to claim that inorganic replacements of the kind envisaged here do not contravene any law of nature. However, even if they do, there is still the possibility (also sustained by Parfit 1984, p. 219) that we can derive some intuitions from them, as long as we pose the right questions. And I am fairly confident that the scenario imagined here includes and specifies all relevant conditions as Wilkes (1988, p. 9) would demand.

50 

Rui Vieira da Cunha

to consider. And I argue that the intuition it elicits is that Tom or you and I or any other human animal could survive such a process. Remember the story once more: Tom had already lapsed into a persistent vegetative state when his transformations began. If not having any of the other parts removed made any difference, why would the brain and brainstem be more important? His mental life has been gone since the beginning of the experiment, so that is not interfering. If that did not keep us from saying he ceased to exist, why would it deter us now? But perhaps it is not the psychological aspects. Perhaps by now you have become so persuaded by the animalist view that the idea of a brainstem replacement is holding you back. The animalist view is clearly attractive in that it can account not only for our persistence conditions but for those of other animals as well. And it not only coincides with our intuitions about numerical identity in most of our ordinary life but it also seems scientifically informed, as far as possible. So, since you have figured out the importance Olson’s animalist view attributes to the brainstem, you might feel tempted to say Tom would die if he had his brainstem replaced with an inorganic one. But why should the brainstem matter all that much? I believe the explanation of the brainstem’s importance in the animalist view is best given if I present what I take to be two possible readings of Olson’s account of an organism’s persistence conditions. These are not absolutely incompatible readings, I think, but they do contain some discrepancies and might render different results when addressing Tom’s situation. To be thorough, I will begin with another reading, one I wouldn’t even consider a possible reading at all but which, even if I consider it incorrect, I must mention, since there seems to be some room to interpret Olson in that sense. In his explanation of what is an organism’s life, Olson’s writing is increasingly analogical and metaphorical. He compares life to thunderstorms, for instance, to account for the fact that a life permanently integrates new matter into the organism which life it is (“A life is a sort of storm of particles in constant motion” 1997, p. 136). He states at a certain point that “Every organism has a life, and it is hard to see how there could be a life without there being an organism whose life it was. And an organism cannot be animated by two lives, at least not at once.” (1997, p. 137). Some readers might take this as Olson engaging in some sort of vitalist path, some sort of inhabitancy

Will I ever be a Cyborg?

51

of the body by a life, somewhat like a breath of life or something almost as mystical. That, I think, couldn’t be further away from Olson’s intentions and, in fact, from his achievement and I hope it is clear why I dismiss this. Olson explicitly mentions biology and biological issues so many times that it would simply be unbelievable to assume that, not to say downright inconsistent. If it means anything at all, I would venture to say it can be no more than Olson’s – and any philosopher of biology or even any biologist, for that matter – inability to scientifically convey what life is in a biological sense. So then, what are the true possible readings of Olson’s account of an organism’s persistence conditions? The first is the reading that focuses on the brainstem’s importance directly or just by itself. Olson seems to endorse this reading several times: when describing the Human Vegetable Case, for instance, he states that “Neither is the animal “brain-dead”, for those parts of its brain that direct its vegetative functions remain fully intact.” (1997, p. 8). Later on, when considering a case where the brainstem is missing, he states: But a detached cerebrum is not an animal […] because its parts do not coordinate their activities in the way that the parts of an organism coordinate theirs. […] The reason is not just that many life-sustaining organs –heart, lungs, digestive tract, and just about everything else – have been removed from the cerebrum, but also that those organs that once coordinated the life-sustaining functions that went on in the arm or cerebrum have been cut away. (1997, p. 115)

And, more emphatically, he also states “I have suggested that your brainstem, as the organ that is chiefly responsible for directing your life-sustaining functions, is essential to you, for without it there is no Lockean life and no living human organism at all.” (1997, p. 140). There are many other passages of this sort11, which may lead one to think the brainstem’s importance is so that one cannot live without it: in fact, we have just seen Olson stating this. And one may be led to assume that it is absolutely impossible for anyone to have their brainstem replaced and, 11 See, for instance, (1997, p. 132): “That is because the organs that once directed those activities – the pons, medulla oblongata, and hypothalamus, among others – are missing.”); or 1997, p. 10: “Imagine, then, that our surgeon leaves the rest of you intact when she removes your cerebrum, so that your brainstem continues to do its job of directing your heartbeat, circulation, breathing, and digestion […]”.

52 

Rui Vieira da Cunha

therefore, that Tom is no exception. On this reading, then, Olson’s view would accommodate our intuitions about Tom in 2060. At that time Tom is still a human animal. He just happens to have had most of his parts replaced by inorganic ones. He is reduced to his brain and brainstem, that much is obvious. But in Olson’s view that would probably be only an extreme case of amputation, of an animal being pared down to the bare minimum that ensures his survival12. His brainstem, the organ responsible for his life function, has suffered neither any sort of injury nor intervention whatsoever. It still plays the same role as in any other human who has not been enhanced the way Tom was. True that most of Tom’s body is made up of inorganic attachments which are not part of it. But that is not problematic for the animalist view: attaching a prosthesis to Tom cannot bring about his death. And it is definitely not a matter of how many prostheses or how many organs and parts are being replaced but of which organs and parts. Now, Tom in 2090 would be a completely different matter for the animalist view. Remember Olson stated that there would be no living organism without its brainstem. If this reading of Olson’s view were right, then we would have to say that Tom dies in 2090, when he has his brainstem replaced. However, that is not what I would consider Olson’s truest reading, and even those evidences of it are, it seems to me, pretty clear concerning what Olson actually means when he refers to the brainstem. It is not that the brainstem matters per se but that it matters in as much as it controls the vital functions of the organism. In all of the above statements, every reference of the brainstem includes a reference to its role, its job, its activity, within the organism: it coordinates the vital functions, it directs those life-sustaining functions, it is chiefly responsible for those functions. So, the important thing is not the brainstem but the brainstem’s job, the functions it coordinates. In case the quotations above are not convincing enough, we may consider these: Your life-sustaining functions are not disrupted when you lapse into a persistent vegetative state (1997, p. 11);

12 It would be pretty much like the case of a detached head connected to a life-support system Olson considers (1997, p. 133).

Will I ever be a Cyborg?

53

[…] the human vegetable in the story is biologically continuous with you – that your life-sustaining functions continue on in that animal […] (1997, p. 12); What it takes for us to persist through time is what I have called biological continuity: one survives just in case one’s purely animal functions – metabolism, the capacity to breathe and circulate one’s blood, and the like – continue. (1997, p. 16).13

So, if we suspected the first reading to be not quite wrong but not quite accurate either, we can now look at this second reading as the more rigorous one, the one Olson himself would certainly endorse. It is not the case that the brainstem matters by itself but that it matters in a derivative sense, because of its functions14. So, is Olson’s view on this second and more accurate reading now able to accommodate our intuitions about Tom in 2090? It would seem that there would be nothing to object, from Olson’s view, to a replacement of the original brainstem by an inorganic brainstem. If the 13 Other examples could be the following: “[…] you are an animal, and an animal ceases to exist when it dies – when its vital functions cease and its tissues decay beyond the point where they can be reanimated. […] you could be immortal only if it is possible for the life-sustaining functions of a biological organism to continue forever.” (1997, p. 71); “On the Biological Approach, what it takes for us to survive remains the same throughout our careers: like other animals, we persist as long as our life-sustaining functions remain intact.” (1997, p. 89); “[…] its organic functions are continuous with yours: your metabolism and other life-sustaining functions have continued on without interruption and are now the life-sustaining functions of the brainless animal.” (1997, p. 116); “Consider the biological concept of death. […] most biologists would agree that it has something to do with the irreversible cessation of those metabolic and other activities that distinguish living organisms from non-living things.” (1997, p. 119). 14 Note that on the first reading the organ is being enthroned on itself, because of being what it is and not of doing what it does. My stress in this last sentence is merely to draw attention to the fact that Olson’s most suitable reading, the one which is closer to his intentions and which has on its side the overwhelming textual evidence, is one in which the basics of his substance concept are defined on the grounds of what something does rather than what something is. And that was exactly Olson’s reason to reject person as a substance concept in the first place. Of course it can be argued that these are different levels of explanation. But it has also been argued – and very convincingly so, I may add – that the distinction between what something is and what something does is illegitimate on the reading Olson needs to prefer animal over person as a substance concept – see Nichols, op. cit..

54 

Rui Vieira da Cunha

brainstem is granted such importance only because of its functions, one might think that it is irrelevant whether it is something organic or inorganic, as long as it does its job, as long as it coordinates the life-sustaining functions. Unfortunately, even though it may seem unproblematic, the idea of an inorganic brainstem is something the animalist view cannot accept. At a certain point, considering the possible existence of other kinds of persons different than human animals, Olson is very clear on stating that no organism can survive a process of inorganic replacement of its brainstem: It may be possible to replace all of your parts, including your brain, gradually and piece by piece, with inorganic prostheses in such a way that your mental capacities were preserved throughout […] The result would be a wholly non-biological person – with rationality, consciousness, free will, the works – who was both psychologically and materially continuous with you. Nevertheless, according to the Biological Approach that being would not be you, for you are a biological organism, and no organism could come to be a non-organism (or so I shall argue). (1997, p. 125)

The argument Olson is referring to in the end of that statement is his account of what an organism is, as I presented it supra, with the relevant features being metabolism, teleology and organized complexity. On Olson’s view, Tom would survive just until 2090, because his brainstem is controlling his life-sustaining functions. The moment it would be replaced, the organism would perish. However, it’s not the case that Tom would perish just because his vital functions were gone or because there was nothing there to direct them. Tom would also cease to exist because he would cease to be an organism and, given Olson’s insistence that organism is a substance concept, Tom could never cease to be an organism without ceasing to exist. Olson is perfectly aware, I think, of the problem that inorganic replacement poses to the animalist view and so, after he has presented his account of what an organism is, he addresses the issue once more: … imagine that your brainstem is replaced by an inorganic substitute gradually, bit by bit, rather than all at once […] there is never a period when your life-sustaining functions are left without an organ to coordinate them, or when your cerebrum is not aroused and activated in the normal way by the brainstem. As a result, there

Will I ever be a Cyborg?

55

need be no interruption in consciousness throughout the operation (suppose the surgeons use only a local anaesthetic). The result would be a rational, conscious being with your mind. Isn’t it obvious that you would be that being? My view, however, entails that you could not survive this … For something with an inorganic brainstem, I argued, could not be an animal at all (1997, pp. 141–142).

In this statement, Olson is focusing on a situation where consciousness is maintained throughout the whole process and so it seems that the only reason why we intuitively say that the being resulting from the operation would be you is due to psychological factors. In Tom’s case, however, there are no such factors to consider and, I argue, the intuition remains: we still think Tom is Tom and has not ceased to exist just because he has that last surgery in 2090. So, given that Tom can become a cyborg and given that he started out as a human organism, there seems to be only two possible answers (apart from denying that Tom has ceased to exist): either to conclude that Tom was not, after all, a human organism, at least not essentially; or to conclude that an organism can survive inorganic replacement15. Either animalism fails on its own grounds, given that this thought experiment was “untainted” with psychological features, or animalism requires some adjustments concerning what exactly an organism is and what its persistence conditions are. The first answer amounts to giving up the animalist view altogether. The second answer would allow us to accommodate our intuition within the animalist view. But how could this be done? One way of going about it is Liao’s way. Liao (2010) argues for the possibility of accommodating a process of inorganic replacement within the animalist view – or the Organism View, as he puts it. Liao considers that a being, X, is essentially an organism, if a) X begins to exist when the capacity to regulate and coordinate its metabolic and other life processes is there; b) X persists as long as there is what may be called “organismic continuity”, which is the continuing ability to regulate and coordinate its metabolic and other life processes; and c) X ceases to exist when the capacity to regulate and coordinate its metabolic and other life processes is permanently gone. (2006, p. 337)

15 Here I am particularly indebted to Eric Olson, who has shown me the need to clarify this point.

56 

Rui Vieira da Cunha

This account of what an organism is, is not all that different from Olson’s. The main difference is that Liao thinks that “there could be non-carbon-based life forms that have non-carbon-based interdependent parts that are used to regulate and coordinate various life processes such as absorption, assimilation, metabolism, and so on, in order to process certain material into fuel so that they would be able to function. If so, it seems that these non-carbon-based life forms would also qualify as organisms.” (2009, p. 17). If this were correct, then inorganic replacement would no longer pose a problem to the animalist (or organism) view and our very strong intuition that Tom would survive the surgery in 2090 may sit easily with that view. Liao (2010, p. 68) even presents a thought experiment similar to Tom’s except that consciousness is maintained throughout the process and the experiment is in reverse: you would begin your existence as a non-carbon-based life form and then have all your parts gradually replaced by carbon-based functional equivalents. I said in the beginning that my thought experiment poses a problem to the animalist view. Given a case like Tom’s, I think an animalist is left with two options: he can bite the bullet and deny Tom survives the 2090 replacement of his brainstem or he can choose Liao’s solution. If he denies that Tom survives after 2090, he cannot accommodate our intuition that Tom does not cease to exist just because a small organ is replaced. In this case, I argued, my thought experiment is more damaging to the animalist view because they cannot dismiss our intuition just by saying that we are biased or influenced by psychological considerations. That option is not available for an animalist in Tom’s case and so he must admit this is a strong blow to the animalist view. If we somehow think Tom can survive a gradual process of inorganic replacement, it seems we can no longer say that Tom falls under the substance concept “organism”. After all, Tom began to exist as an organism but has ceased being an organism without ceasing to exist. So, is being an organism just a phase in Tom’s existence, just like being a child, an athlete, or a philosopher? It certainly seems so. As for Liao’s solution, I believe it comes at a very high cost for the animalist (or organism) view. The cost of redefining what an organism is in such a broad way that it may cease to be a substance concept. Of particular note, for instance, is Liao’s case for the possibility or conceivability of non-carbon-based life forms:

Will I ever be a Cyborg?

57

But while it may be the case that organisms that are most familiar to us are all carbon-based life forms, there is no reason to suppose that all organisms are necessarily carbon-based life forms. Strictly speaking, organisms are just entities that have interdependent parts. (2009, p. 17, my emphasis16)

You can object that there is nothing wrong with Liao’s solution, that it just broadens the concept of organism. In doing so, you would say, Liao is just providing a more detailed account of what an organism is. Isn’t that what I said my thought experiment was meant to show about the animalist view? That in order for it not to fail on its own grounds, it would need some finer tuning concerning what exactly an organism is and its persistence conditions? That is true. I do not think, however, that Liao’s tuning can be made while still maintaining organism as a substance concept17. Remember, a substance concept in Wigginsian terminology is something that tells us what a thing fundamentally is. It tells us that thing’s most essential metaphysical nature. It helps to set that thing apart from other things which do not share its nature. If we take Liao’s solution, even though his explicit definition of what an organism is seems very close to Olson’s, we are broadening the concept to such an extent that I believe too many different things in the world could count as organisms. Olson himself is very aware of all the problems this poses for the animalist view, I think. Right after addressing the topic of inorganic replacements, in the above excerpt, he states that there is little that he can say to reply to our intuition that we could survive such a process. 16

My emphasis here should by no means be understood as saying that this is all there is to Liao’s definition of organism – which I have already presented earlier. What I want to point out is that given that definition, and the fact that an organism could be non-carbon-based in his account, that aspect – being an entity that has interdependent parts – is one of the main features of an organism. And that is something perhaps too many things in the world possess. 17 I thank Arto Laitinen for suggesting that perhaps Liao’s solution would require us to think of organism as the substance concept and of carbon-based or noncarbon-based as a phase sortal. Though I am tempted to say that would be an accurate rendition of Liao’s proposal, I am not sure if it is. And I am also not sure if that explains away the fact that we do tend to think of organisms and its being carbon-based as one essential feature of it, as something an organism could not cease to be without ceasing to exist. Talk of a non-carbon based organism may be acceptable but doesn’t that force us to count many different things as organisms?

58 

Rui Vieira da Cunha

And he is clear regarding the options that are left: “If you want to insist that you could survive such a thing, you must either deny that you are a living organism, or come up with an alternative account of what it takes for an animal to persist through time” (1997, p. 142). Liao’s solution is probably an attempt at such an alternative account. However, the way I understand it, it can also be seen as falling terribly close to being an outright denial that we are essentially living organisms, that is, that organism is the substance concept under which we fall. I have tried to show how Olson’s animalist view fails to explain our strong intuition about Tom’s survival in my thought experiment. Even after trying two possible readings of Olson’s account of an organism’s persistence conditions, there is no plausible answer the animalist view can offer us. Since I resorted to no psychological or mental continuity or connectedness of any sort to bring about such intuitions, there is no way we can say that we were influenced by a psychological approach nor, I dare say, by practical attitudes about “what matters in identity” (Olson, 2007, p. 43). My claim that animalism fails on its own grounds thus means that it cannot offer plausible answers even to an audience willing to consider only biological factors when accounting for our persistence conditions. I see no other alternative than to conclude that animalism requires some adjustments regarding what exactly an organism is and what its persistence conditions are. Whether these adjustments can be satisfied with something like Liao’s solution, I am not quite sure, considering the possible implied consequences of such a solution on the central premise of animalism: that organism is a substance concept.

5. Conclusion The main premise of Eric Olson’s view on personal identity issues is that animal or human animal or organism is what best serves as the answer to what we most fundamentally are and what determines our persistence conditions. That that is, in short, our substance concept. Resorting to a thought experiment concerning inorganic replacement, I have tried to show that things like you and I can become totally

Will I ever be a Cyborg?

59

inorganic. If organisms cannot become such inorganic things, then you and I are not (essentially) organisms. If organisms can become inorganic things, then we need a better definition of organism. As I claimed at the beginning of this paper, Olson’s animalist view cannot accommodate our very strong intuitions about such cases and the only options left for an animalist, apart from trying to deny that premise – that Tom and you and I can become totally inorganic – are either to give up animalism altogether or come up with a better account of what exactly an organism is and what its persistence conditions are. Liao’s solution, even though cleverly argued, is one that in attempting to achieve the latter, does so in such a broad way that it comes too close to achieving the former.

References Bostrom, N. (2003). Human Genetic Enhancements: A Transhumanist Perspective. The Journal of Value Inquiry, 37, 493–506. Liao, M. (2006). The Organism View Defended. The Monist, 89, no. 3, 334–350. Liao, M. (2010). Twinning, Inorganic Replacement and the Organism View. Ratio, 23, 59–72. Nichols, P. (2010). Substance Concepts and Personal Identity. Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, 150, no. 2, 255–270. Olson, E. (1997). The Human Animal: Personal Identity Without Psychology. New York: Oxford University Press. Olson, E. (2007). What are We? A Study in Personal Ontology. New York: Oxford University Press. Parfit, D. (1984). Reasons and persons. Oxford: Clarendon Press. Wiggins, D. (1980). Sameness and Substance. Cambridge, MA: Harvard University Press. Wilkes, K. (1988). Real People: Personal identity without thought experiments. Oxford: Clarendon Press.

Part II Epistemology and Phenomenology

How Consciousness explains the Self Klaus Gärtner

What is the ‘Self’? This question has a long tradition in Philosophy of Mind. For some Philosophers, like Descartes, it is the entity which holds our consciousness, while others, like Hume, deny its existence. Whatever it is – whether it exists or not – there seems to be no doubt about how we can access something that we are inclined to call a ‘Self’. Since Descartes, this access has been our Consciousness. While trying – but still failing – to obtain an explanatory concept of Consciousness we are still in the state of determining whether the ‘Self’ really exists, and if so in what form. I therefore propose the following hypothesis: A concept of the ‘Self’ depends necessarily (but not sufficiently) on a concept of Consciousness. What will be the aim? Well, to establish a conceptual relation between Consciousness and the ‘Self’ and if possible to obtain a tool able to handle a concept of the ‘Self’. What we can gain from this approach is that our concept of the ‘Self’ will depend on the choices, experiments, ideas and conceptual background that also account for the concept of Consciousness. I will propose the following methodology: we have to introduce an epistemological relation, where Consciousness can be seen as the access to whatever it is we call a ‘Self’. This epistemological approach will depend in part on phenomenological reasoning since this kind of reasoning is able to show limits and problems regarding certain concepts. Furthermore, this approach seems to be integrative and therefore connects a variety of related aspects in Philosophy of Mind. The key to establishing such a relation will be Self-Consciousness.

64 

Klaus Gärtner

Justifying the Connection What is Consciousness? I do not want to discuss all the different theories of Consciousness. Rather, I want to show which is the relevant one to support my argument. Basically there are three general categories of Consciousness: Creature Consciousness, State Consciousness and Consciousness as an entity.1 I will focus on Creature Consciousness only. Creature Consciousness is concerned with how an “[…] animal, person or other cognitive system may be regarded as conscious in a number of different ways.”2 I decided to focus on Creature Consciousness because this is the specific set of characterizations used where we actually describe in what way someone or something is conscious, that is to say what attributes someone has to have to be considered a conscious creature. To explore the ‘Self ’ three criteria seem to be interesting: the Wakefulness –, the What it is like – and of course the Self-Consciousness criterion. There are three more criteria mentioned within the context of Creature Consciousness which are either criticisable, because they can be considered an understatement (Sentience), depend on conscious states (Subject of conscious states) – which were already excluded –, or which involve how consciousness is directed at an object (Transitive Consciousness), but this does not tell us anything about accessing the ‘Self ’. The Wakefulness criterion tells us about the capacity of an organism to respond to its surrounding environment or better one can only be conscious when one is awake and alert. What this means is something or someone cannot be conscious when asleep or in a coma. The What it is like characterization gives us a more subjective notion. When we are conscious it seems to be the case that there is a certain ‘something what is like to be that organism’. This is famously described in Thomas Nagel’s article ‘What is it like to be a bat?’3. The third and most important tool to access our ‘Self’ is the Self-Consciousness criterion. 1 2 3

See Van Gulick (2009). Van Gulick (2009). See Nagel (1974).

How Consciousness explains the Self

65

This is the awareness that one is aware. Those are the three criteria to be used describing access to the ‘Self’. Again, we have to keep in mind that this is not everything there is to say about Consciousness. These ideas seem to be the critical ones when we are talking about accessing the ‘Self’ and when we are trying to show how a concept of the ‘Self ’ depends on a concept of Consciousness. What is the ‘Self’? This question cannot be answered here. It is important, however, to show a relation to Consciousness so that perhaps in the end there might be a solution to this problem. Nevertheless we have to at least characterize the ‘Self’ to see what we are dealing with. The problem of the ‘Self ’ is characterized by the subject of mental phenomena. It is the entity which bears sensations, perceptions, thoughts, desires and actions.4 Usually it is referred to as a first-person perspective. This is not to say that the ‘Self ’ can be called an object, it does not seem to be a thing such as something we refer to when we say ‘it is sunny’. However short and problematic this characterization may be, it seems to be more interesting how one can think about the ‘Self ’. According to Colin McGinn three mature theories of the ‘Self ’ can be identified: theories which identify the self with the body; theories which explain the self in terms of various mental relations; and theories which take the concept of the self to be primitive and not to be explained in terms of anything else.5

I think we have to add a fourth theory: denying the ‘Self’ completely6. The first theory explains that our self-reference refers to the body, an object with mental attributes. The second theory states that our self-reference refers to complex mental states which belong to the ‘Self ’. The third theory claims that the ‘Self’ is a substance different from the 4 5 6

See McGinn (1998). McGinn (1998, p. 144). As an example see Metzinger (2003).

66 

Klaus Gärtner

body – non-reducible to mental states – which is simple and basic. The fourth theory denies the ‘Self’ in all of its forms, which means there are multiple reasons to reject a theory of the ‘Self’ (for example, identity disorders). Those approaches to the ‘Self’ are very general. It is not my intention to go into further detail. This small introduction has only been given to keep in mind that there are different approaches. Now we can get to the real problem: why is a concept of Consciousness necessary (but not sufficient) for any concept of the ‘Self’? The relation between a concept of Consciousness and a concept of the ‘Self’ The questions are now the following: How can we connect a concept of the ‘Self’ to a concept of Consciousness? Furthermore, we also have to ask ourselves: Why does a concept of the ‘Self’ depend on a concept of Consciousness? Well, the answers depend on the three criteria of Consciousness which have something to do with our access to the ‘Self’. So, how can we connect the above criteria – Wakefulness, What is it like and Self-Consciousness – to the ‘Self’? Intuitively, we would say that, well, our Consciousness has to have a certain Subject, Consciousness does not seem to exist for itself and classically that structure has been the ‘Self’. That, however, might not be true anymore, but for a certain form of Introspection or Phenomenology it seems to be necessary. Since at least in these terms all three criteria need a relation to the ‘Self’. Let us start here. It seems to be obvious that the relation between the ‘Self ’ and Self-Consciousness is the strongest. What we have to do are two things: first, we have to find the relation between Self-Consciousness and the ‘Self’ and second, explain how the other two characteristics can be included in that relation. The first part does not seem too complicated. In most of the literature that introduces the ‘Self’, the ‘Self’ is somehow connected to Self-Consciousness (to talk about her ‘Self’ she needs a certain self-awareness). This awareness can be explained by “[…] differences between self-awareness and the awareness of the objects of

How Consciousness explains the Self

67

experience that enter into the content of experience […].”7 Ascribing this capacity to someone seems to fulfil our notion of the ‘Self ’. A subject, for example an animal, that has a certain living body but lacks the idea of a psychological subject appears to be different in nature from this idea – at least that is what we think. That means an animal is a subject that has mental states but only in a living body without access via Self-Consciousness. This Self-Consciousness gives us a special kind of unity which is a unity from the inside. So we can say the following: What this position implies is that the kind of unity a self exemplifies is bound up with self-consciousness; without self-consciousness the mind of a creature has no more unity than that conferred body – from the inside it is just a collection or succession of mental states.8

As we can see, the relation between the ‘Self ’ and Self-Consciousness is established. It is now possible to talk about the second problem. How can we include the Wakefulness criterion and the What is it like criterion into this relation? We can connect both criteria to Self-Consciousness. When we talk about phenomenological approaches to Self-Consciousness we can claim that two kinds of Self-Consciousness exist. It is important that there is a pre-reflective – apart from the reflective – Self-Consciousness which means that “[…] (1) it is not an explicit or thematic form of self-consciousness, and (2) reflective self-consciousness is possible only because there is a pre-reflective self-awareness that is an on-going and more primary self-consciousness.”9 This notion incorporates the idea of What it is like because it is a subjective feeling of self-awareness and therefore implies certain qualities. Thus, this is the notion defended by Thomas Nagel10. So the problem of including the What it is like criterion is basically to give this criterion a Self-Consciousness base. What is suggested is “[…] that first-person experience presents me with an immediate and non-observational access to myself, and that

7 8 9 10

Morrison (1996, p. 550). McGinn (1998, p. 143). Gallagher and Zahavi (2009). See Nagel (1974).

68 

Klaus Gärtner

(phenomenal) consciousness consequently entails a (minimal) form of self-consciousness.”11 This solves our problem of how to entail the What it is like criterion, but still does not give us a solution for the Wakefulness criterion. That is a bit trickier, but can also be solved by the phenomenological notion. What we have to claim here is that the pre-reflective self-awareness has to be distinguished from an explicit self-awareness. This seems to be the case, since we are not talking about an awareness that is observational; rather, it is a non-observational self-acquaintance. We can put it this way: To have a self-experience does not entail the apprehension of a special self-object; it does not entail the existence of a special experience of a self alongside other experiences but different from them. To be aware of oneself is not to capture a pure self that exists separately from the stream of experience; rather it is to be conscious of one’s experience in its implicit first-person mode of givenness.12

This contradicts the idea of some Philosophers who claim that children do not have a Self-Consciousness, while in psychology, the evidences seem to point in a different direction13. This is the basic idea of minimal Self-Consciousness, which is also defended by some contemporary analytical Philosophers14. Minimal Self-Consciousness is primitive and already present from birth. It is prior to the learning of language and the ability to conceptually form judgements and seems to be the basis of a more advanced form of Self-Consciousness. With this minimal notion of Self-Consciousness it is now possible to include Wakefulness in our relation. We just have to claim that every form of consciousness entails this primary pre-reflective Self-Consciousness. Object experiences do not enter our mind in a purely conscious form; there is always some sort of ownership of an experienced object. Excluding this ownership seems to be implausible. Of course phenomenological reasoning cannot give the complete answer. It seems merely impossible to justify our constitution as humans – especially when we talk about the mind – only through certain 11 12 13 14

Gallagher and Zahavi (2009). Gallagher and Zahavi (2009). See Gallagher and Zahavi (2009). See for example Goldman (1976).

How Consciousness explains the Self

69

problems and limits we have but, at the same time, to establish a relation of certain problems and concepts seems to be necessary. We can now try to explain why a concept of the ‘Self’ depends on a concept of Consciousness. The idea is simple and has already been mentioned above. We cannot know anything about the ‘Self’ without accessing it and the way we access our ‘Self’ is via Consciousness. It is therefore necessary to have a concept of Consciousness first. Depending on that concept we can analytically access the ‘Self’. At this point we normally say that is not what is happening: Whatever Consciousness is and how it is grounded, we usually think of it as just a tool for introspection. That might not be entirely right. I argued above that a certain pre-reflective Self-Consciousness forms part of all sorts of Consciousness; there is, at least on a pre-reflective level, a direct phenomenological link to our ‘Self’. Since that is the case and there are possibilities to analyse Consciousness – but the ‘Self’ only via Consciousness – we need a concept of Consciousness first. That turns a concept of Consciousness into a necessary condition for a concept of the ‘Self’. Is a concept of Consciousness also sufficient for a concept of the ‘Self’? Here the answer has to be no. It seems to be an overstatement to claim: accessing the ‘Self’ via Consciousness may constitute the ‘Self ’. The ‘Self’ is only the content or character of our Consciousness15 and we can therefore be wrong. Even if we try to argue that there is a direct link of acquaintance at the pre-reflective level, that is, a relation of sensing a certain ownership within the What it is like, in a way these sorts of sensations may exist, but this does not mean that their content has to be true. It might be that in the end there is nothing more than our Self-Consciousness, which in its pre-reflective form, has a certain What it is like character and that this character is sensed with an ownership, but this still leaves the possibility open that a ‘Self ’ might be an illusion. It is true that it has been argued that this very basic relation forms part of being a certain creature, like a bat16, but even though we need this sensation and this sensation might be different in every life form and subject, it does not follow that the world is the way we sense it. This would leave us with no more than Self-Consciousness and a certain 15 This depends on what theory of Consciousness one defends. 16 See Nagel (1974).

70 

Klaus Gärtner

sensation, but this does not directly imply a ‘Self ’. Therefore, it seems clear that the sufficiency condition cannot be fulfilled.

Conclusion This basically leads to the idea of the hypothesis: a concept of Consciousness is necessary (but not sufficient) for a concept of the ‘Self ’. This implies certain things: first, there is no claim about ontological existence. We cannot say what is first: Consciousness or the ‘Self ’. Second, we cannot decide whether the ‘Self’ does exist or not. What we can argue though is: if we suppose that the ‘Self ’ exists (some people seem to think that it has to) its conceptual form depends on the choices, experiments, ideas and conceptual background that we make for the concept of Consciousness.

References Block, N. (1996). Consciousness. In Guttenplan, S. (ed.), A Companion to the Philosophy of Mind. Oxford and Malden: Blackwell Publisher, 210–219. Brook, A. & Raymont, P. (2009).The Unity of Consciousness. In Edward N. Zalta  (ed.), The Stanford Encyclopedia of Philosophy (Winter 2009 Edition), . Gallagher, S. & Zahavi, D., (2009). Phenomenological Approaches to Self-Consciousness. In Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Spring 2009 Edition), . Goldman, A. (1976). A Theory of Human Action, Princeton: Princeton University Press.

How Consciousness explains the Self

71

McGinn, C. (1996). The Character of Mind. An Introduction to the Philosophy of Mind, New York: Oxford University Press. Metzinger, T. (2003). Being No One. The Self-Model Theory of Subjectivity, Cambridge, MA: MIT Press. Morrison, K. (1996). The Self. In Guttenplan, S. (ed.), A Companion to the Philosophy of Mind., Oxford and Malden: Blackwell Publishers, 550–558. Nagel, T. (1974). What is it like to be a bat? Philosophical Review, 83, 435–450. Van Gulick, R., (2009). Consciousness. In Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Winter 2009 Edition), .

Self-Knowledge, Introspection and Memory1 António Marques

Self-knowledge (S-K) has a variety of dimensions and philosophers have proposed, and continue to propose, a host of different methods to explore or to establish the principles of a very particular form of knowledge such as this. What is quite impressive in the philosophical literature (by which I mean not only that of analytical inspiration but also that of the so-called continental philosophy) is the fact that the usual relationship between S-K, introspection and reflection leaves no space for the essential role of memory. Hence, it is memory’s role in S-K that I wish to look at here. Traditionally, empiricist and rationalist approaches have dominated the epistemic views of S-K. The latter starts with Descartes, or even earlier, and continues until Kant; the former includes Locke and Hume2. 1

2

A first draft of this paper was presented at the seminar about “Knowledge and Ability” coordinated by Franck Lihoreau. I’m grateful for his kind invitation and for the stimulating discussion with the researchers who participated in that session. This is a too simple if not erroneous scheme taking into account the multiple and often contradictory positions that exist even inside the same field. It is enough to stress that Descartes (or what can be designated as Cartesianism), who is pointed out as the classical paradigm of self-knowledge by introspection (S-KI) and the concept of reflection that operates in it, is submitted to a devastating critique in Kant’s first Critique (notably in the “Paralogisms of Pure Reason” of the first Critique). I do not think there is enough evidence in Descartes’ main works to support the picture referred to above, but I shall not follow up this discussion. The fact is that it is in Locke that a clear concept of reflection as an essential piece of introspection and self-knowledge is to be found. The following quotation from Locke’s Essay Concerning Human Understanding illustrates a relatively common concept of reflection which prevails in the literature on these themes: “The mind receiving the ideas mentioned in the foregoing chapters from without, when it turns its view inward upon itself, and observes its own actions about those ideas it has, takes from thence other ideas, which are as capable to be the objects of its contemplation as any of those it received from foreign things” (Essay …, II,

74 

António Marques

In fact, if one thinks of Hume and Kant as the most relevant modern representatives and the culmination points of both approaches to S-K, it would seem perfectly acceptable to suggest that introspection plays a determinant methodological role. In the first case, let us remember what Hume famously said regarding his own mind and what he finds there: “For my part, when I enter most intimately into what I call myself, I always stumble on some particular perception or other, of hot or cold, light or shade, love or hatred, pain or pleasure. I never can catch myself, at any time without a perception, and never can observe any thing but the perception” (D. Hume, 1984, 300). In the case of Kant, in the section “The paralogisms of pure reason”, because he is occupied with the deconstruction of all dogmatic metaphysics about the ‘I’, the introspective tool is also used: “Now in inner intuition there is nothing permanent, for the ‘I’ is merely the consciousness of my thought. So long, therefore, as we do not go beyond mere thinking, we are without the necessary condition for applying the concept of substance, that is, of a self-subsistent subject, to the self as a thinking being. And with the objective reality of the concept of substance, the allied concept of simplicity likewise vanishes…” (Kant, Critique of Pure Reason, B 413, transl., N. Kemp Smith). With these quotations from these two classical authors I only wish to show that knowledge of the self, which should be acquired by introspection, is relatively poor: a particular kind of perception in the case of Hume, and the mere consciousness of the ‘I’ in the case of Kant. I make this remark in order to show that skepticism about the possibility of genuine knowledge already has, to my mind, its roots in classical philosophy, be it empiricist or rationalist. When I refer here to skepticism in relation to the knowledge of one’s own mind (or self), I mean our capacity to know our mind by setting up a dual structure, that is, a mind that looks at itself, a mind that sets itself up as a target. Surprisingly, if one takes the examples of both Hume and Kant (for different reasons),

chapter VI). Objects “as capable to be the objects of its contemplation as any of those it received from foreign things” are precisely what seem to be at the core of any introspective move, but this is precisely what makes it so problematic.

Self-Knowledge, Introspection and Memory

75

this target seems to offer us little: a perception of feelings or awareness of an ‘I’3. Nonetheless, in order to approach the notion of self-knowledge, one needs to understand as clearly as possible the link between reflection and memory, and my argument below is based on the view that the dual structure of reflection has to be revised by the introduction of the role of memory. Consequently, what I suggest is that clarification of what can be considered self-knowledge and introspection requires a consideration of the role that is played by memory and the evaluation of the epistemic nature of retrodictions (that is, statements about past events, in this case, statements about past mental events). Then my argument, which for the moment is only designed to be a draft, goes through the following points: –  I: Introspection is an inescapable tool of S-K and is an instance of direct knowledge of my mental or psychological/mental states, then S-K is in most cases S-K by introspection (S-KI). – II: S-KI is memory dependent (there is a structural time factor in all S-KI that is virtually unrecognized by many philosophers). – III: if S-KI is dependent on memory and at the same time is direct (unmediated) knowledge, it cannot be treated as an analogy with displaced perceptual knowledge (displaced knowledge: for example, to know that person A was in this room by seeing his hat). – IV: S-KI is expressed in retrodictions, that is, statements about my own past mental states (not to be confounded with statements about myself in the past). – V: S-K that is not S-KI does not depend on memory. Typically, it corresponds linguistically to expressions or avowals in the present tense which possess a different epistemic nature, namely they 3

In the same vein, Wittgenstein quoting William James makes a nice (and intentionally over-simplified) picture of introspective operations: “Here we have a case of introspection, not unlike that which gave William James the idea that the ‘self’ consisted mainly of ‘peculiar motions in the head and between the head and the throat’. And James’ introspection showed, not the meaning of the Word ‘self’ (so far as it means something like ‘person’. ‘human being’, ‘he himself’, ‘I myself’), or any analysis of such a being, but the state of a philosopher’s attention when he says the Word ‘self ’ to himself and tries to analyse its meaning. (And much could be learned from this.)” (Wittgenstein, PI §413).

76 

António Marques

are not only authoritative in the first person but are also infallible. Typically, expressions like “I’m cold”, “I’m furious”, “I’ve the intention to do this”, etc., which are expressions of mental/psychological states, require only sincerity conditions. Yet I shall suggest that even expressions or avowals are not totally memory independent if one accepts the working memory function introduced by some psycholinguists. But this issue is not to be discussed here. – VI: S-KI does not have a reflexive structure, in the sense that it is not a second-order thought targeting a first-order one. –  VII: S-KI has first-person authority with respect to the contents of one’s own mind although it is fallible precisely because the retrodictions are linguistically memory dependent. – VIII: Statements like “I was sad in Paris last week”, which possess first-person authority, are fallible and are pieces of S-KI unlike the correspondent statement “I was in Paris last week”, which does not have any authority of that kind and is consequently not S-KI. Nevertheless, let us mention that both statements are fallible. Below I will not deal with each of the above points separately; I only wish to identify some transversal guidelines against the background of the views of some of main authors. Let us begin with Fred Dretske, whose book Naturalizing the Mind (1995) has an entire chapter entitled “Introspection”. An intuitive definition of it would be, for example, “the mind’s direct knowledge of itself ” (Dretske, 1997, p. 39) although he adds that introspection is an “instance of displaced knowledge”. Introspective knowledge is knowledge of the mind – i.e., mental facts. Since mental facts (according to the thesis I am promoting) are representational facts, introspective knowledge is a (conceptual) representation of a representation – of the fact that something (else) is a representation or has a certain representational content. It is, in this sense, meta-representational. Meta-representations are not merely representations of representations. They are representations of them as representations (1997, p. 43). Since mental facts (according to the thesis I am promoting) are representational facts, introspective knowledge is a (conceptual) representation of a representation – of the fact that something (else) is a representation or has a certain representational content (ibid.).

Self-Knowledge, Introspection and Memory

77

Then the examples Dretske gives are analogies with representations of external objects: I can see a photo as a piece of paper, as an object with 2 grams, as a photo of Clyde, and so on. I suggest that this feature of introspection as representation of one’s own mental states as something else seems quite hard to accept. Normally I do not represent me as sad, but I am or I was or I will be simply sad. Cases in which I represent me as sad are typically situations where I say that I represent my face as sad (either in a picture or in a photo or in my imagination). But these situations are not pieces of S-K. Of course a little more introspective attention on my recalling can raise doubts in me about my sadness and then I will ask myself whether what I felt was a genuine feeling of sadness, about the reason I was sad, and so on. Anyway in such cases it would be better to speak about continuous and associated memory operations. However, these cases do not have the double form of a first-order representation, which is set as a target, and another second-order one, that is, a meta-representation of that target. Even a retrodiction of some complexity such as “I was sad in Paris because of this and that, and so on” does not have a dual structure like a meta-representation of first-level representations. It is much more analogous to my direct representation of myself in a mirror. The situation that just simulates the meta-representation would be: I am seeing myself in a second mirror, which mirrors me in a first mirror. But this does not adequately simulate S-KI: remembering my past mental state, when I was sad, does not have any kind of intermediate representation; it is direct knowledge (occurred at a time 2) that essentially consists of the recalling of a past event (occurred at time 1). Recognizing this difficulty Dretske claims that: It may seem as though this account of introspective knowledge – as a species of displaced perception – makes it into a form of inferential, and thus indirect knowledge. If introspective knowledge of oneself – that one represents the world thus and so – has the same structure as knowing that the postman has arrived by hearing the dog bark, then there is an intermediate ‘step’ in the reasoning that makes knowledge of the target indirect. There are two ways of knowing that the postman has arrived: by seeing or hearing him arrive and by seeing or hearing something else (the dog) that ‘tells’ you he has arrived. If, on a representational theory, introspective knowledge is more like the latter than the former, then the representational account fails to give self-knowledge the immediacy that we know it to have (Dretske, 1997, p. 60).

78 

António Marques

Then Dretske adds that this is a relevant objection to the analogy that he claims between introspection and displaced perceptual knowledge, but he insists that in an introspective statement like “I represent k as blue” the introspective quality comes (and this is the fundamental difference between displaced introspective knowledge and displaced perceptual knowledge) from the fact that the truth of the information of my representation of blue is not relevant, if what I see/represent is or is not blue; the relevant fact in that introspective statement is about I am representing k (ibidem, p. 61). To my mind Dretske wishes to preserve the traditional dualistic structure of introspection and self-knowledge as reflection. This leads us to another point about the reflexive nature of introspection and S-K. Like Dretske, T. Burge, or more recently P. Jacob4, accepts this reflexive structure as part of S-K. I have already commented on Dretske’s view of displaced knowledge, on his wish to preserve the reflexive feature of introspection and the implicit difficulties he recognized. In the case of Burge, it is possible to find difficulties regarding S-K as a dual reflexive structure although of another kind to those that are to be found in Dretske. The case of Tyler Burge is quite specific in the sense that he considers that, on the one hand, at least some S-K statements seem not to require a dual structure since the targeted thought coincides in time and content with the subject (second-order) thought, but on the other hand S-K involves necessarily the difference between meta- and first-level thoughts on which is based all reflection. Let us observe in some detail Burge’s approach to this topic. Burge sees the differences between S-K and perceptual knowledge in the personal status of S-K. Namely in S-K acts, one is facing statements that are made simultaneously from and about one first-person point of view. He identifies this feature in the following terms: “the point of view and time of the judgment must be the same as that of the thought being judged to occur” (Burge, 1988, p. 651). And he adds “When I judge: I am thinking that writing requires concentration, the time of the judgment and that of the thought being judged are the same” 4

For a detailed discussion of Dretske’s model of displaced knowledge applied to S-K, see P. Jacob, 2010. The theoretical framework of the discussion is Jacob’s wish to show that “our picture of introspective self-knowledge derives from two traditional sources: rationalist epistemology and empiricist epistemology” (ibid., p. 3).

Self-Knowledge, Introspection and Memory

79

(ibid.). This also means, as Burge makes clear in another passage of the same paper, that “The reflexive judgment simply inherits the content of the first-order thought” (ibid., p. 650). The essential difference in relation to perceptual knowledge consists of the personal status of S-K, or in other words the use of first-person pronouns, and as he notes: “In all cases of authoritative self-knowledge, even in those cases which are not ‘basic’ in our sense, it is clear that their first-person character is fundamental to their epistemic status” (ibid. p. 651). Thus it is first-person authority and not so much its reflexive structure that confers a special epistemic status on S-K if compared with perception. Or perhaps in other words, the correct interpretation of Burge’s view is that it is the reflexive structure of S-K plus the first-person authority that gives it its peculiar epistemic nature. Yet if one accepts Burge’s description of S-K, when one compares the epistemic nature of S-K and that of perceptual knowledge, one feature of the former (the thought I have has the same content and coincides in time with what is thought about) does not place both knowledge types far from each other. After all, in reflexive knowledge (as all S-KI should be) more than one content is not to be found. This means that the first-level (targeted) thought does not have a content on its own5 5. What makes the difference is the place of first-person authority in S-K. Yet does this difference produce drastic epistemic consequences? In Burge’s view it does, but I think it is a mistake: first-person authority does not imply infallibility in the process of introspection. Let us make the nature of this process clearer. 5

This is a remarkable result in order to re-design the traditional concept of reflection and introspection. It is interesting to consider another feature of this redesign since in S-K each targeted event can be seen as a “recognizing” of what happens or of what is happening in me. Wittgenstein’s approach to the nature of recognition is striking inasmuch as it describes the misunderstandings surrounding recognition when memory is taken into account. The result is also “fusion” into one content, as is the case in Burge’s approach. “It is easy to misconceive what is called “recognizing”; as if recognizing always consisted in comparing two impressions with one another. It is as if I carried a picture of an object with me and used it to identify an object as the one represented by the picture. Our memory seems to us to be the agent of such a comparison, by preserving a picture of what has been seen before, or by allowing us to look into the past (as if down a spyglass)” (PI §604). Yet this is a mistaken picture, “Indeed, it is not so much as if I were comparing the object with a picture set beside it, but as if the object coincided with the picture. So I see only one thing, not two” (PI §605).

80 

António Marques

Using such criteria, it seems hard to maintain that S-K is clearly distinct from the usual perceptual knowledge since, after all, in both there is one content and there is no time gap that I can be aware of. My concern lies not so much in the claim for the existence of a unique content in the case of reflexive knowledge (“reflexive judgment simply inherits the content of the first-order thought”), with which I agree, but much more with the claim that in S-K there is a typical reflexive structure with a dual composition based on a targeted object by a second-level thought/representation. As I have tried to show, the reflexive operation does exist in all S-K yet as an introspective move which is memory dependent6 S-K is S-KI with a dual time structure where memory is necessarily involved. By dual time structure I mean the awareness that a past t1 event is targeted by an actual thought/ representation typically expressed in retrodictions at t2. This is exactly why Burge does not see in basic S-K a necessary time gap, which requires the work of memory. As I said above, he sees S-K as possessing the classical dual structure plus first-person authority and it is this characteristic alone that gives S-K a special epistemic nature. I would say that he means something near to infallibility. The following lines illustrate, in my opinion, this conviction: “The source of our strong epistemic right, our justification, in our basic self-knowledge is not that we know a lot about each thought we know we have. It is that we can explicate its nature and its enabling conditions. It is that we are in the position of thinking these thoughts in the second-order, self-verifying way” (Burge, 1988, p. 654). But this is already problematic for Davidson to whom “of course people have beliefs, wishes, doubts, and so forth; but to allow this is not to suggest that beliefs, wishes, and doubts are entities in or before the mind, or that being in such states requires there to be corresponding mental objects” (D. Davidson, 2001, pp. 35–36). So if there are no mental states as objects in the mind that one could grasp in any sense, there is no such a thing as introspection in the usual sense for Davidson 6

The fact that in S-K there is one content (first-order thoughts do not have any content on their own) does not imply that the targeted content that occurred in the past coincides in time with the actual statement/thought. On the contrary, it is essential to S-K that the awareness of a time gap (even the narrowest) is preserved. Yet if I am not wrong, Burge’s position erases any time gap whatsoever.

Self-Knowledge, Introspection and Memory

81

I guess. But Davidson’s point is that such a thing is not needed in order to have an attitude. In Davidson’s words, “having an attitude is not having an entity before the mind; for compelling psychological and epistemological reasons we should deny that there are objects of the mind” (ibidem, p. 36). I believe Davidson has a fundamental point here, but I think he could go deeper in the consideration of these “compelling psychological and epistemological reasons”. It is at this moment that the problem we face can be formulated in the following terms: it is counterintuitive to simply deny the existence of introspection and the correspondent S-K, but one should not count on dual reflexive structures, on target objects in the mind and similar sorts of objects to identify introspective knowledge. It is at this point that it seems to me necessary to introduce the time factor, or in other words memory. S-KI needs a time distance, even a minimal one: the statement about my own mental/ psychological state made at t2 always refers to a t1 event. There is, I agree, a dual structure, but it is a present/past dual one not a reflexive one. Even I “reflect” at t2, for example, as a result of mental states that I felt at t1. There is always a dual time structure that still requires at least two things that are usually required for S-KI, which are 1) to be direct, that is, not mediated like any displaced knowledge, and 2) to have first-person authority. Recent studies on memory and especially short-term memory elucidate us not only about the complex neuro-chemical processes involved and what distinguishes this kind of memory from long-term memory. They also allow us to understand the role that short-term memory plays as it works when representations/thoughts coincide (in time) with their correspondent content ones or when we report something to other people or to ourselves. Psychologists call this very short-term memory “working memory” and it can be defined in the following terms: “Working memory can be thought of as a low capacity information reservoir that is always full, sensations flowing into it continuously at about the same rate that they are forgotten […]. Working memory is an indispensable form of transient memory; it is a moving window of comprehension that allows us to understand the present in terms of the very recent past” (O’Shea, M., 2005, p. 85). This is the case whenever the grammatical form is equivalent to present tense statements. It is enough to think about statements like “I’m understanding this text”, “I’m telling you

82 

António Marques

how much I am sad”, “I’m seeing this sunset as a Rothko picture”, etc. These are statements that do not seem to have the same structure as simple avowals or expressions. Precisely they involve a working memory that eanables cognitive performances and are present in all language learning7. This way of seeing S-KI leads us to the question of the status of the content of retrodictions, which are memory dependent statements and are different from retrodictions that are not memory dependent, or rather that have a truth-value that does not depend on my memory: the statement that I was in Paris last week is a retrodiction of this type. It is not a piece of introspection, and even if it possesses the dual time structure, it lacks the other essential component, that is, the authority of first-person8. My past sadness is not a target of a meta-representation but simply the content of a thought/ representation that in this case is a recall of a past state of mind, experience or whatever. What happens is this: the so-called first level content, i.e. the event that at time1 I was sad, is

7

8

For many psycholinguists working memory is implicated in syntactic comprehension or in keeping track of syntactic dependences. Furthermore, the working memory is what enables cognitive performances such as attention being focused on targets at a particular time. For example, the judgment, “Now I’m understanding this argument” involves working memory, and even implicit elements learned in the past that enable comprehension are, so to say, linguistically hidden. If we consider that judgment is a piece of S -K, then the targeted element is memory dependent and the statement itself possesses a dual time structure. See C. Philip Beaman, “Working Memory: Beyond Language and Symbolism”, Current Anthropology, Vol. 51, June 2010, pp. S27–S38, and Conway, A. R. A., M. Kane, M. Bunting, Z. Hambrick, O. Wilhelm, and R. W. Engle, 2005. “Working memory span tasks: a methodological review and user’s guide”, Psychonomic Bulletin and Review 12, pp. 769–786. First-person authority must fulfil the following condition: a total asymmetry in relation to the truth/satisfaction/conditions of the correspondent third-person statement’s content. Then “I (A.M.) was in Paris last week” has the same truth/ satisfaction condition of “He (A.M.) was in Paris last week”. They are not asymmetric. But “I (A.M.) was sad last week in Paris” has not the same truth/satisfaction condition of “He (A.M.) was sad last week in Paris”. So “I was sad last week in Paris” has a first-person authority which is lacking in the first statement “I was in Paris last week”.

Self-Knowledge, Introspection and Memory

83

recalled at time2. Is the t2 representation a meta-representation of the t1 representation? I would say that it is not and I shall illustrate this claim with a remark of Wittgenstein’s from the Philosophical Investigations. “Does it make sense to ask ‘How do you know that you believe that?’ – and is the answer: ‘I find it out by introspection’? In some cases it will be possible to say some such thing, in most not. It makes sense to ask, ‘Do you really love her, or am I only fooling myself ?’, and the process of introspection is the calling up of memories, of imagined possible situations, and of the feelings that one would have if…” (PI §587). The remarkable thing in this quotation is that for Wittgenstein introspection, so to speak, achieves its target in most cases by an operation of recalling, of memory. In other words, in most cases the structure of introspection requires a time factor, and without it there is no target whatsoever. (In this case, the target element is the quality of a feeling.) On the other hand, when memory does not play a role, it seems that Wittgenstein would reject introspection. Without memory, no target, no introspection. Possibly the cases where Wittgenstein would say that it does not make sense to speak of introspection are such as: believing p and observing my belief seems to be absurd. Let us note that even some statements like “Now I’m believing p” implicate that I perhaps did not believe p before. But this contrast only needs the time factor. In fact, what is more plausible according to the quotation above is that Wittgenstein considered that in some cases where time did not play a role there was no sense in the question. Is memory dependent inquiry truly self-knowledge and are other forms, which exclude memory, pseudo self-knowledge? Wittgenstein does not make this explicit, but one can suspect that it is what he means. Thus, as a non-definitive conclusion, S-KI has at least 3 features: a dual time structure based on memory, it must be direct and it must have first-person authority.

84 

António Marques

References Burge, T. (1988). Individualism and Self-Knowledge. Journal of Philosophy, 85, 649–663. Burge, T. (2007). Foundations of the Mind, Oxford: Clarendon Press. Davidson, D. (2001). Subjective, Intersubjective, Objective (“Knowing One’s Own Mind”), Oxford: Clarendon Press. Dretske, F. (1995). Naturalizing the Mind, Cambridge, Mass.: MIT Press. Hume, D. (1984). A Treatise of Human Nature, London: Penguin. Jacob, P. (2010). Do we know how we know our own minds yet? . Kant, I. (1992). Critique of Pure Reason, trans. N. Kemp Smith, London: Macmillan. Locke, J. (1690). Essay Concerning Human Understanding. O’Shea, M. (2005). The Brain, Oxford: Oxford University Press. Wittgenstein, L. (2009). Philosophical Investigations, (revised 4th edition), Oxford: Wiley Blackwell (quotations from PI are made in the usual form referring to the number of the section in the case of the first part of the work).

Imagination as a Bodily Pattern: thinking about Sartrean’s account of Consciousness Clara Morando

Sartre’s study called L’Imaginaire. Psychologie Phénoménologique de L’Imagination (1940) is a follow-up to another book named L’Imagination (1936), where we can find a) a clarifying list of the imaginative consciousness’s distinctive traits, b) an extensive characterization of various kinds of images (eidetic, hypnagogic), c) long-detailed descriptions about certain psychopathologies, and finally, d) a perspicuous categorization of the remaining consciousness modes as perceiving and conceptualising. Concerning what draws the specificity of imagination, this book adds to a historical view a phenomenological deepening into a peculiar type of intentionality, ascertained by an eidetic treatment. The first part of the book, called “Le Certain”, deals with the image’s intentional structure where Sartre proposes a functional metapsychology claiming that when consciousness works based on an irrealizing scheme connected to non-existent or absent objects, we are then allowed to call it imagination. This particular way of intending things goes on by building up, step by step, the imaginary consistence whose very function is to establish another strand of knowledge constituted by several noetic acts which have in common the imaging movement as a different type of intentionality beyond perceptive knowledge. In order to make the image’s description as an image possible, it is necessary to embrace a second-order-thought, just because it is imperious to understand how the very same object, the imaged object, occurs in our consciousness. Although the Cartesian view considers that only by a conscious act (an intentional act) can imagination be cleared, we have to recognise that Sartre gives us more than we can simply find in Cartesianism as a philosophical doctrine, historically rooted on the distinction between mind and body and on the assumption that consciousness is a

86 

Clara Morando

mind’s indispensable attribute. It is true that only by a reflective act is it possible to achieve a fuller comprehension of what imaging means. However, we cannot forget that Sartre does not say that that same act is incorrigible, i.e., that all we get by reflection is absolutely clear and distinct. The French author also never appeals to an obscure anthropological dualism, believing contrarily that the body is a psychic substance, and the psychic is bodily instantiated. What we get here is that the imaging activity can only be a special corporeal operation, inserted in a vaster field called consciousness, although is not just that simple. He tears apart imaginative consciousness from all other forms of intentionality, formulating suggestive hypothesis about its very nature and putting in practice the Husserlian phenomenological method. However, he does not follow it strictly and refuses to suspend the existence of the natural world. He decomposes the images to what matters in terms of their distinctive traits, making a sort of reduction but without the respective phenomenological epoché that he considers to be a miracle we cannot understand very well. Thus, we have now an open gate to assert that every image is a form of consciousness, although this particular consciousness differs from merely perceptive consciousness. We can also say now that consciousness has nothing to do with a closed and sterile immanence, incapable of reaching the world. Gregory McCulloch stresses that “consciousness is nothing else than a relationship”, and this applies also to a special form of consciousness – the imagination. When I have a perception it means that I have a certain (pre-reflexive) consciousness of it, and, for instance, a perceived chair can be viewed as the object of that particular movement of consciousness. What Sartre greatly ensures goes on to say that when we have an image of the very same chair, what specifically happens is that we are allowed to believe in a sort of identity between the chair as the object of perception and the chair as the object of imaging. According to this, consciousness relates itself with the referred chair in two different ways, and in both ways of dealing with the object ‘chair’ we see as constant the directive movement to its corporeity. In the specific case of imagination, Sartre adds that consciousness establishes a kind of synthetically organized intentional movement directed to the existing chair.

Imagination as a Bodily Pattern

87

We can say that the image’s object does not simply correspond to a mere image of it in terms of what we consider a reproducing copy of the real object. In his analysis he vehemently insists that “imaging experiences do not have images as intentional objects”, they have the very objects themselves. So, an image, as Sartre says, is first of all a relation entailed between the conscious subject and its object, which results after conceiving an image through the so-called consciousness’s intentional movement. The image is the very intentional movement. We must have present the fact that perceiving and conceptualising, with imagining, represent three types of complex structures linked to conscious activity. There are, thus, three possible ways of being conscious of the same object. a) In perception, I am really facing the object in its very existence, although I am not capable of perceiving it in its entirety, i.e., in all of its material richness. I am obliged (structurally) to make, through my perceptive consciousness, an always incomplete adding operation related to the available object’s data. This means that there is a kind of perceptive resistance in the objects that opens to the possibility of a successive annihilation onto their existing characteristics, and the only way we have to partially avoid it (the annihilating process) is to successively produce a series of profiles about the same object. We establish, then, a crucial synthesis that surrounds objects and shows most of all (or only some of) their traits or properties. The idea of the perceived object is represented in a whole-thinking of it, and presupposes a sort of concrete knowledge, firstly without words and images and attached corporeally to what is perceived. This happens also in imaging activity, which appears to locate in the very same side of perception, marked in its constitution by a synthesis of multiple appearances. The only but great difference among them relies on the fact that if perception consists of a slow-building knowledge as Sartre points out, imaging is, on the other hand, an immediate form of knowing that. The latter is then directly connected with the object by an intriguing and instantaneous way that stands for us like a sort of enigma. In perception, contrarily, nothing appears without having a complex branch of connections with other objects that gives us the special feeling that when we perceive things we just only get on them half of the picture from them.

88 

Clara Morando

On image we just find an essential poverty in terms of what we call the synthetical object’s richness. That specific way of intending/ understanding an object leads us to a particular kind of isolation from it among all other objects, imaged or not. The different elements that compose an image and the way they connect each other are not sufficient to establish an internal coherence that could rival the one we encounter on perceived objects. As we said, the worldly existing objects and the impossibility of knowing them in their entirety constitute a strong proof of the infinity of the perceptive process. Consciousness is always inferior to the real world’s richness because it is incapable to get the complexity understood. The perception of the real things is really menaced by the incompleteness signal. b) In imaging, on the other hand, we see a strict correspondence between consciousness and the very image produced by its intentional movement. On image we have nothing to learn, because the object’s imaging intuition seems to happen very rapidly, like lightning, and is not also compelled, by its very nature, to respect physical laws (or just some of them). It is reasonable to assert now that only by imaging can we be totally sure of what we are imaging, just because here we do not need to know if there is a proper adequation between existing objects and imaged objects. In Sartre’s philosophy of imagination we are never allowed to say that images represent mere copies of the existing objects. Images are, instead, just an intentional movement that envisages non-existing objects (absent, unreal) showing us, on the other hand, that they have a certain scale of reality, but now another type of reality, with different powers, and not less intense. To clarify this, the French author suggests that all contents pertaining to an image point out, in a delusive way, to the empirical world. Indeed, on that specific pointing out to a presumable material plan is found to be an almost-collapsing “irrational quality”, simply because in its movement towards a non-existing object, imagination does not commit itself to strictly respect as a whole the physical laws engaged in perception. It remains, in all the image’s contents, a sort of opacity in relation to their structural meaning in a latter sense, due not only to the fact that they are not obliged to follow physical laws, but also to the fact that in the very intentional moment directed to a particular non-existing object

Imagination as a Bodily Pattern

89

we are confronted with an emerging power of consciousness entirely unsuspected. We do not find here regulating essences nor generating rules. We simply find a kind of assessment of the image that is greatly linked to the feeling of certainty we encounter in perception. That is why we are authorized to assume that images are objects of quasi-observation, and we know now this is not because they copy real objects, but because they produce the irreflective belief of a peculiar existence, although we rationally maintain their effective non-existence. This powerful belief, incoherent and outstanding in a minimal first stage, is the very core of imagination, equivalent to intuition. That is why it is impossible to see on images a décalage between objects and consciousness, representing only the result of an intense intuition. We can also say that they (images and consciousness) are just the same thing. It follows then that there is another characteristic one can focus on the image’s status, and it consists exactly of that the imaging consciousness posits its objects all-embedded in nothingness. Or, more rigorously, it grasps them in a progressive movement towards nothingness. We have seen that the precise instant in which we almost believe in the image’s empirical existence is a sort of paradoxical step in relation to what it follows. What comes after is then precisely the stage where we assure the image is a simple image. But is also on that very first step (and we cannot forget it) that relies the imagination’s intriguing and powerful trait. We cannot get erroneously drifted by this descriptive and critical set of statements to the very assumption that in the irreflective level of consciousness we are dealing with a certain type of unconscious. Nothing can be more wrong than this. Even in the irreflective level, that Sartre prefers to call pre-reflective, consciousness possesses of itself merely an immanent consciousness or a non-thetically rooted consciousness. This means that consciousness in its lower-level is simply consciousness of ‘the’ thing; for instance, in the case of perception, it is consciousness of the ‘chair’. More rigorously, it means perceptive consciousness of the existing ‘chair’. Just in the upper-level, the second-order-thought-level (the reflexive level), is consciousness able to posit itself as consciousness of an existing ‘chair’. So, consciousness, in its various categories, is omnipresent. What happens then in the specific case of imagination? How is it possible for the imaging consciousness that originally pertains solely to

90 

Clara Morando

the pre-reflective level of consciousness to posit itself as an image? How does that specific type of consciousness appear to itself as an image if being an image would initially presuppose a certain mode of reflexivity? This last question seems only to make sense in terms of what we understand as the traditional image theory. Sartre objects to it by saying that the image is not a sort of miniature in our minds in relation to objects (he calls this view ‘the illusion of immanence’), but is in the very image’s producing act where there occurs a special way of constituting a particular object as an image in the wide world of things. Thus, is precisely the transcendent consciousness connected with a determined thing (or object), turned into an image, that posits the very same object as an image, which essentially seems to be out of our consciousness. Is uniquely the astonishing intuition of transcendence that is conveyed to all imaged objects, belonging now apparently to the world of things, which leads to the compelling impulse of consciousness to hunt them as if they were exterior to their own producing act of consciousness. This explains why the positional act taken by all imaging consciousnesses in relation to a specific object involves necessarily (and specially) not only a) a character of negation by positing the object in the plan of quasi-observation, as we above said, but involves also b) a kind of “positive element” in that very act of positing. This “positive element” is paradoxically ascertained by a strange and intuitive feeling about what we could name the always-running evanescence of the image. We have now clearly understood that images are themselves objects among other objects, including the perceived objects, and that an image can be defined by its constant negation as “nothingness of being”, although it starts through a belief in a sort of intensity of its proper existence. However, this primary belief is not what a superficial critique would call an unconscious belief, functioning merely as an obscure incentive for the rest of the process, but represents abstractly the belief in the image as a pure and simple object and nothing else. When I say that I have Pierre’s image I also want to say that I do not see anything at all. But it does not mean that I am not envisaging Pierre in an intuitive way just because I am not seeing him. Sartre says then that what my actual intention intends to grasp is precisely Pierre in his corporeity, “that Pierre’s image is a way of not touching him, a way that he has of not being”. So: Pierre’s image’s main trait is to

Imagination as a Bodily Pattern

91

be a kind of “intuitive-absent” object of consciousness, more precisely, an intuited given absent to intuition. How can we have an intuition of nothingness? It is precisely this immediate consciousness of a particular type of nothingness which constitutes that specific corporeity I was pointing to before. Another characteristic figured in imaginative consciousness in its peculiar movement towards emptiness and whose proper mode is to be at the same time and in-itself a non-thetically consciousness involving the object is, precisely, spontaneity. This spontaneity or, more properly, this intriguing spontaneous emergence described as a specific kind of intentionality that progressively denies the existence of a certain object, does not have in its core, as a primary goal, the urgency to be reported to a specific object, i.e., to this or that object. What happens is just that intentional movement represents first of all a type of consciousness that is simply transversal. Sartre uses a metaphor saying that what imaginative consciousness specifically does is to arrange a sort of crepuscular light glued to itself in order to spontaneously produce and maintain the object as an object-in-image, as a nothing-object. Consciousness seems then to create several objects-in-image but abjures simultaneously its creative power on them. This tells us that consciousness simply takes off through negation of the positive element that would affirm the modelling gift of existence. It substitutes then this particular gift by another type of positivity, a paradoxical positivity, essentially linked to the act of empirical negation whose intensity feels sometimes to be so great that can even rival that of objects constituted by perceptive consciousness. To a descriptive static of imaginative consciousness there follows a dynamic way of talking about it, just because it is not possible to phenomenologically study consciousness without seeing it as a temporal synthesis organized between other forms of consciousness. Those other forms of consciousness come after or before imaginative consciousness, composing among all of them a continuum that metaphorically evocates a melody. The pattern followed by this particular type of consciousness seems to have in common with perceptive consciousness the fact that also the first one looks at its object, analogically, through perceptive territory, having special care when determining the sensible elements constituting it. McCulloch stresses that Care is in order here, since there are, for

92 

Clara Morando

instance, two ways in which one might image ‘a chair’. The difference here is that whether at perceptive processes we just find a passive acceptance taken by consciousness in relation to sensible objects, simply representing them as they are (although this is always a partial representation), at imaging processes consciousness adopts towards things on their representations a dynamic posture, over-flowed by an unstoppable creative will.

Hypnagogic Images This astonishing creative will can be so dramatic that it produces sometimes a sort of pathological feeling of over-reality in certain images. These images seem nervously to be more real, more concrete and more palpable, than empirical objects. Once they are hypnotising objects they also seem to have great powers on other things, including us, fulfilling a special kind of hierarchical superiority that totally submits their very creator subject. Therefore, these special images constitute a transcendent menace over us. Sartre calls these hypnagogic images menacing visions saying also that our consciousness posits them dangerously among all other objects, images or not, simply because when we perceive things, our consciousness does not primarily ascertain objects into the perception level. We just put the representation as simply being a representation. To perceive a thing is solely to put it in the middle of all other things, not having the concern to know if they are real or not. Although this is done with a specific concern of internal coherence, what we only achieve is a particular coherence that cannot be said to be empirical. So hypnagogic images rely on a positive belief about the likeness of their very structure, engaging themselves also with the quasiobservation phenomena; however they are not obliged subject to the exactly same requirements than in perceptive objects or plausible images (according to the world of perception). It is true that we can talk about an internal coherence on hypnagogic images; however, that does not imply a sense of full respect concerning physical laws. As we said before, for Sartre those images contain

Imagination as a Bodily Pattern

93

a fantastic character, caressing concise frontiers on their forms, ghostly-like. The strict law of individuation is not applicable to them, neither are to other laws of perception like, for instance, the “perspective law”. Sartre says that there is something inexplicable in them, although they reflect only contents of the specific knowledge we possess. What happens is that in hypnagogic images, the subject abstains from making judgements that would rectify, organize and stabilize the image contents. We just do that in perception, and to do that we need more or less some distance in relation to the perceived object. It is true that in plausible images sphere (which are much likened with sensible objects) that specific judging distance is smaller, but that does not mean that it fails to appear. What we see now is that in hypnagogic images the essential décalage is inexistent and consciousness is then able to feel some of the elements that compose its knowledge, multiple-versions-combined, in a way so intense that they seem to constitute sensible evidences. The appearing of the object and the certainty of its existence make just one. And we are not talking about a slow movement of consciousness, as occurs in perception, but about a dazzling and “fatal” appearance resulting from an instantaneous sum of several past consciousnesses that form between them a temporal unity of a longer-persisting type of consciousness – the consciousness of hypnagogic images. What intrigues us here is not the process of summing up through multiple and finished consciousnesses. What perhaps constitutes motive for reasonable pre-occupation is precisely the arbitrary and devastating way in which this particular form of consciousness passes through all over the rest. It’s like a big hole in the middle of perceptive and imaging plausible representations, using strangely their own materials to compose in an unpredictable way new things, completely bizarre or even monstrous. Ontologically speaking, it seems that nothingness opens a breach through our consciousness but disguises itself in a total contradictory appearance – as an all-powerful being, in this case, the hypnagogic image. It is not its content that relates to nothingness, it is precisely the movement of consciousness producing it that relates to nothingness through an interior process of negation, though very intense. Consciousness here is captured by hypnagogic images (it is equivalent to the hypnagogic image), and in this very equivalence we can say that consciousness is captured by itself. Consciousness is then a

94 

Clara Morando

nothingness that produces paradoxically several forms of nothingness. The producer of judgements about things, the I, taken as transcendent to consciousness no longer means a transcendence, an indispensable fiction created by consciousness in order to be reflexive, and is now plunged in the territory of the transcendental consciousness, i.e., the only form of consciousness. Aron Gurwitsch describes it as non-egological or completely anonymous. The I matches then with the pre-reflective level of consciousness, and this consists of a longer-type of consciousness, as we explained above, simply because consciousness is fascinated by the presumable contents of that specific image. It is ultimately fascinated by its own movement as consciousness in a sort of self-suggested paralysis in-itself, not being able to create properly the (transcendent) I as the equivalent to the reflective moment of consciousness. We stay fascinated by the hypnagogic image in a sort of physical and mental trap. Other psychologists have studied these very phenomena and they are explicitly referenced in Sartre’s work. The object has then no longer any type of exteriority in face of consciousness and its presence coincides totally with the surface of transcendental consciousness. Consciousness “superficiality” is due to the fact that it is, as Sartre says in BN, pure wind empty of contents – a mere translucent movement of intentionality towards objects with no inhabitants within. Thus, what happens when we have hypnagogic images is that consciousness, as stated before, does not possess a contemplative power in itself – does not find a way to keep at distance the influence of those very images. Consciousness operates then, “miraculously” or instantaneously, a bizarre synthesis over accumulated materials (memories, feelings and thoughts) that functions similarly to an absurd categorical imperative. All of this is done with a “sense of despair” taken by consciousness about its impossibility to guarantee a decent inferential unity. The only way seen to keep something likened to reality is to confer an inedited meaning to the hypnagogic image in order to make clear that there is, maybe, some reasonability, although those images disrespect several laws of physics. When my body is paralysed and my thoughts are fascinated by the hypnagogic image, then there appears, in Sartre’s words, a new structure of consciousness: the captured consciousness. And this happens often in moments that appear before,

Imagination as a Bodily Pattern

95

for example, sleep (in a light form), or that happen in schizophrenia and other mental disturbances. Human consciousness is left with no alternatives in ontological terms; it no longer has the power of negation to deny what is intended, i.e., the possibility of the very annihilating possible becomes unreachable. The self as a distance born through the creative power of consciousness as a strategy has to avoid being hypnotized by objects (whatever they are: images, concepts, perceptions, memories) ceases completely its efficacy. Moran (Introduction to Phenomenology) accuses Sartre of being all-Cartesian due to his constant refusal about the possibility of the unconscious. Consciousness is always conscious of itself, including in its pre-reflective level, even if it is there only consciousness of its objects. It seems that we can sometimes be, or better yet to say that we are most of the time, caressing beings of reflective awareness and that essential caress must be, in a certain way, or if we prefer, in pathological cases, a recurrent state or a permanent state. Although Sartre’s book departs from Husserl’s theory of imagination, relating the latter with the perception sphere and recognising in it inferiority in face of perceptive processes, the French author rethinks it and agrees to say that imagination is an independent type of consciousness that cannot be reduced to perception. As Dermot Moran stresses, and appealing to Sartre’s words, specific activity of consciousness – the imagination activity – generates sui generis objects. In order to get through this Sartrean statement Moran recommends that we are more attentive about what is meant with the puzzling activity of imagination. He focus his reasoning on what is said about the very act of imagining (which is also imaging). More specifically he states that the act of imagining is a magical act, i.e., is a sort of enchantment intending to produce the object of thinking that each one of us has, and that imagination process must be exactly like this, because we need, when we imagine, to literally possess the imagined object. We are the imagined object. I would add to Moran’s observation to the fact that this magical act depends upon an “immediate awareness of its nothingness”. That is why imagined objects are never totally coincident with perceived objects. That is why we also can talk of a transversal intentionality that goes through images and through all objects of imagination. And that

96 

Clara Morando

is why it is possible for us to be compossible with those very kinds of images – through that nothingness which is paradoxically a positivity as we have seen. In the case of images, what happens is that when our consciousness produces them it does not do so in a way that we could totally believe to be disconnected from a material sense. Moran adds that the particular operation of recalling the material content of an image in our minds involves a certain type of corporeal presentation, although this presentation is not really or actually present. When we imagine we do it in a corporeal way, without any kind of platonic separation between our bodies and our minds, without any artificial distance within our all-beings. Sartre stresses in L’Imaginaire that when I think about Peter I think about him corporally, although I cannot see, touch or hear him. And I also know at the precise moment I recall him that he is not materially present. We are then allowed to believe that the body engraves on imagination its marks. This unavoidable fact is visible, for instance, in the case when I see an arbitrary bunch of lines and I find there a face. This happens often when we look at clouds, as everybody knows. The referred example gives me reasons to think that when we imagine we do it under powerful influence of our bodies, i.e., that we follow strictly on images-building a sort of generality concerned with corporeal movements. In this sense, another example given by Sartre also shows the very ocular movement as the provider of rules to find (to imagine) in chaotic or meaningless drawings any kind of familiar knowledge, whatever it is and whatever it costs. This explains why we have to hunt (firstly) for the object-in-image in the world, as said before, knowing now that that imaginative hunt must obey the requisites of an embodied consciousness. Our judgement about what is seen is determined by the physical apparatus of our eyes, and this is a judgement that is not merely representative but means essentially a corporeal way of sensing objects traditionally referred to the internal arena of our minds. As Merleau-Ponty recognises, in his Phénoménologie de la Perception, Sartre sees that the body’s-spatiality implies that all things are observed in a relational of the build-over human’s physical dispositions, and everything what we deal with is precisely accounted for in the interior of that same net. Sartre calls this the hodological space. In

Imagination as a Bodily Pattern

97

a radical version it also includes the products of imagination, and it is only in this sense that we can say that imagination has something to do with perception. The problem here is that it is not possible to have any point of view about our bodies without making at the same time an infinite regression. Sartre underlines an absolute incapacity of the body to be simultaneously self-known and transcendent, because the very body is always the instrument that cannot be used by another instrument, i.e., the particular point of view about which I cannot have another point of view. So: all of the stated arguments lead to a conclusion establishing the pre-reflective level of consciousness (spontaneous, a first-order-thought awareness) as being no longer the body’s self-consciousness, although the body belongs permanently to the non-thetic consciousness structures board. If we admit imagination as a specific form of consciousness belonging in a primary order to irreflective consciousness, then we have to ask how is imagination not a way of escaping from the world, but an essential way of living more sensible in it. In Being and Nothingness Sartre makes explicit that if is true that our consciousness is not able to recognise at the same time its body’s-being without making of that specific assumption a transcendence, it is no less true that in all irreflective moments of consciousness (which always and necessarily occur when we are in a reflexive state) the usage of the verb to exist should be done in a transitive mode, and that consciousness exists in its body. The type of relation established here is an objective relation between the body as a point of view and things, whether between consciousness and its body we can find differently an existential relation. What we cannot forget at this point is that they are simultaneous (the two types of relation – objective and existential), so, knowing about the Sartrean-Cartesian consciousness Omni-presence commitment, we also conclude for this double-relation in images sphere. The body is a conscious structure of my own consciousness, and body’s-consciousness means, more particularly, a lateral and retrospective way of sensing the world. We face here a strong equivalence between body and consciousness. The mind is its own sensations, but it never gets totally identified with them, simply because in that very movement of self-identification with a corporeal object – the body,

98 

Clara Morando

consciousness needs to open a distance within itself in order to be sophisticatedly aware of it, of what it is. Thus, to have consciousness of our own body we need to face it as if it were another person’s body. We have to reflexively deal with it just inserted in an objective relation which means precisely that we never reach it reflexively through an existential relation. But the latter exists and the only way we have to prove it is to assume indirectly its effects severely induced in corporeality. Kathleen Wider states that “Sartre’s development of this bodily self-presence to oneself as subject as well as object is minimal and inadequate. This is because of his adherence to a division between the lived body and the body as an object in space, in the world, open to the look of another”. As Phyllis Sutton Morris says, we are unable to introspect the subject, objecting strongly to Russell’s claim that experiencing consists of a relation, that we find there an experienced object and who goes through the experience – the subject. The body is only an implicit term in the analysis of our own conscious experience. However, as Sartre claims, the same body is a centre through which objects are intentioned, and in this very claim we cannot put the body as an additional object of experience. The sensation is a hybrid notion between the subjective and the objective, conceived as an exterior object but then applied to the subject. The Being-for-itself has to be all body and all consciousness; it cannot be united to a body. This is very familiar with what we have seen about images. We could add here the example given by Sartre when he talks about schematic drawings. He states that what is interesting about the scheme is the very fact of being an intermediary between the image and the sign. However we can place it on the side of an image. The intention as initially perceptive becomes imaged, but for that it is necessary for the subject to interpret the scheme in a way that allows consciousness to create. On that very creation it is necessary for the body to adopt a certain attitude, i.e., to play a symbolic pantomime in order to animate the bunch of traits constituting the scheme. We can see it, for instance, in the different moments of a bird’s flight or in a ballerina’s movements. Not forgetting also that the object’s relation with the retina is neutral and the only way of giving an answer to the movement of the real subject is to have an existential relation with it. And imagining is another form of doing it.

Imagination as a Bodily Pattern

99

We can now finish these very lines about what in Sartre’s philosophy the activity of imagination means and implies by pointing out several suggestions: If the body is implicitly perceived within its relation with other perceived objects (including “its own body” in just an objective relation), indicating a distance as a necessary condition to be aware of the objects, then what happens with the objects we call images? It seems that we can discern in the very activity of imagination an attempt of consciousness to grasp absent or non-existent objects in an always-corporeal way, because imagination lends life to what is feeble, to what is surrounded by nothingness, conferring to a negative trait (the object’s inexistence or absence) a positive character which is much stronger than its material concreteness. That is why imagination is also much more powerful than conceptualisation; because it is a dynamic form that operates having as reference the possibilities of the existence, of the corporality, although both of them are negated. In hypnagogic images this very process reaches such a higher level that there is no distance between our embodied consciousnesses and images, and we can say here that we are pathologically our images. So, in conclusion: 1) If the body can provide a unified subject in single acts of consciousness, as Morris sees in Sartre’s proposal, because in perception the same body is at the centre of different situations, spatially, we could say the same of imagination. Also in it there is consciousness unity, even in hypnagogic images – that represent, as we said, a desperate attempt to give some (corporeal) reasonability to the appearing objects. 2) Summing up these traits which were pointed out, we just have to say, as Sartre did, that consciousness is not a black box, but a continuous relation that a particular body has with objects – perceived, conceptualised or imagined (the hodological space). In order to be coherent with Sartre’s perspective we also have to assume that in imagination we similarly relations with objects; the only difference is that they are not real. But that does not mean there is an evanescent way in dealing with images. To prove against that prejudice we take the schematic drawings case, for instance, and we are compelled to see that the body-subject inexorably puts its marks in images. 3) So, if consciousness has a spatial origin referring itself permanently to that same origin, an identical situation must happen in a

100 

Clara Morando

special form of consciousness as the imagination, which is a form of consciousness, probably the most important.

References Gurwistch, A. (1964). Field of Consciousness. Duquesne: Duquesne University Press. Merleau-Ponty, M. (1945). Phénoménologie de la perception. Paris: Gallimard. McCulloch, G. (1994). Using Sartre. An Analytical Introduction to Early Sartrean Themes. New York: Routledge. Moran, D. (2005). Husserl. Founder of Phenomenology. Cambridge: Polity Press. Morris, P. S. (1976). Sartre’s Concept of a Person. An Analytic Approach. Massachussetts: University of Massachussetts Press. Sartre, J. P. (2003). La transcendance de l’Ego et autres textes phénoménologiques. Paris: Vrin. Sartre, J. P. (1993). O Ser e o Nada. Ensaio de ontologia fenomenológica. Lisbon: Círculo de Leitores.

Feelings and the Self Dina Mendonça

All philosophers recognize how crucial the notion of self is for emotional processes – that is, to be emotionally aroused is to feel the self implicated in some way. Both self and emotions are at the center of much philosophical debate and there are on-going debates and disagreements and many unclear issues regarding both topics. However, both concepts seem to share a common trait: there is a sense that both emotion and self are such that their design is never fully complete – that is that they are open-ended entities. Even though both the concept of emotions and the concept of self are continuingly being analyzed and discussed in philosophical debates, this open-ended character of both may be one way to better grasp their nature. In this paper I want to look upon their relationship by focusing as much as possible on their open-ended nature. The paper argues for a situated approach to the nature of emotion by showing first how a situated approach can bring to the surface interesting connections between the self and the emotional world, and second how such an approach allows us to understand how emotions design the self. The first part of the paper lays down the situated approach to emotion and the many ways in which the implicated self can be understood. The second part of the paper looks at some emotions (fear, love, pride and jealousy) in order to illustrate and elaborate on the conceptual map constructed in the first part of the paper, and indicate some of the open-ended nature of both concepts.

102 

Dina Mendonça

Part I: Situations and the Self i. Situated approach to emotions One way to grasp the intentionality of emotions and avoid attaching them solely to the sentient subject is to take a situated approach to the nature of emotions. De Sousa takes such approach in his book The Rationality of Emotions (1987) when he introduces the notion of paradigm scenarios. De Sousa writes, My hypothesis is this: We are made familiar with the vocabulary of emotion by association with paradigm scenarios. These are drawn first from our daily life as small children and later reinforced by the stories, art, and culture to which we are exposed. Later still, in literate cultures, they are supplemented and refined by literature. Paradigm scenarios involve two aspects: first, a situation type providing the characteristic objects of the specific emotion-type (where objects can be of the various sorts identified in chapter 5), and second, a set of characteristic or “normal” responses to the situation, where normality is first a biological matter and then very quickly becomes a cultural one (De Sousa, 1987, p. 182).

I think it is accurate to state that we become acquainted with the vocabulary of emotion through paradigm scenarios but I also think that the dynamics of paradigm scenarios is more complex. First, the story of how these paradigms are drawn is more complex than De Sousa describes them. It is not simply that stories reinforce paradigm scenarios, though I’m sure that experience before story telling is crucial for the emotional relevance of stories. However, I think they also can introduce new paradigms that are reinforced by daily life existence (or not), that is, stories point out possibilities of paradigms as well as complexity of paradigms. There is probably a creative process between stories and daily events that form these paradigm scenarios. And only this explains that, as De Sousa writes, “A paradigm can always be challenged in the light of a wider range of considerations than are available when the case is viewed in isolation.” (De Sousa, 1987, p. 187) Second, the assumption of normal responses to a situation requires detailed explanation avoided by De Sousa. How is this normal determined? De Sousa writes that it is first a biological matter and then a cultural one, but the question remains: how is it determined? Biologically

Feelings and the Self

103

men and women procreate; does that mean that it is normal for couples to have children in situations of love and abnormal not to have children? What is the status of the abnormality? No doubt biological and cultural structures make some general boundaries for responses to emotional situations: biologically, one cannot feel like expressing love with kisses if one does not have a mouth to kiss with; and culturally yawning is to be avoided (or disguised) to prevent offending people in certain situations in countries where it is taken as a sign of boredom. Nevertheless, given that there are issues about the normativity of emotional processes, as well as questions regarding the source and function of normality, much more needs to be said about the status of these biological and cultural structures and their role in establishing boundaries and norms for emotional processes. Finally, in De Sousa’s description the person who experiences is placed outside the scope of the situation, and this forces De Sousa to end up dividing the world of paradigm scenarios in two: objects and normal responses. Such division obscures one of the greatest advantages of the situated approach to emotion, namely the movement and a complex dynamics of sequences of events in emotional processes. In order to maintain the richness of De Sousa’s paradigm scenarios without the problems identified above I want to complement it by using John Dewey’s conception of situation (Logic: The Theory of Inquiry, 1938) in order to indicate a different understanding of the connections between self and emotion, and the genealogy of such a connection. The notion of “situation” is crucial for the understanding of Dewey’s philosophical work. For the purpose of this paper, it suffices to point out that Dewey’s redefinition of experience, as primarily an affair of doing/undergoing, shifts the focus of attention from the functioning of our sense organs to the situated, complex structure of our actions as living creatures. That is, we base our analysis of experience on the complex structure of situations, instead of basing it solely upon the functioning of our sense apparatus. Another important consequence follows from this description of experience as an interaction: experience is no longer primarily cognitive. Stating that experience is no longer primarily cognitive means that experience is now primarily an active-passive affair (MW 9, p. 147), and there are more activities than the cognitive ones. Ultimately, this provides a reason to show that when we reflect

104 

Dina Mendonça

upon reality we are not reflecting merely upon an epistemological account of experience, and therefore knowledge is no longer the only ruler available to measure reality. The idea of being situated allows us to clarify that organism and environment are not understood as independent entities. In fact, we can only understand organism as part of its environment, and we can only understand environment as a dynamic process that involves the organism. This means that we cannot fully understand the implicated self of emotional processes without seeing it as partly defined by the emotional processes it undergoes, and similarly the emotional process is defined by the character of the self. Consequently, self and emotional processes are never totally taken to be given and complete once and for all for, as Tallisse writes, in his simple and clear book on Dewey, “the term ‘environment’ does not denote some permanent, independent entity; rather, it is shorthand for an array of interrelated and active forces or factors that collectively constitute the conditions within which we, at some particular time and in some particular place, live.” (Tallisse, 2000, p. 19) Likewise, organism does not stand as something permanent, an independent entity; rather it is shorthand for an array of interrelated active forces or factors that collectively constitute modes or ways in which one lives at some particular time and in some particular place. A common objection that is raised to Dewey’s philosophy and which is of particular interest for the philosophy of emotions is the charge of subjectivism. The claim is usually supported by the fact that Dewey’s description of experience always leaves unanswered the question “who is the subject of the experience?” Dewey’s view of experience in terms of situations does not seem to completely cut across the subject-object distinction, for the subject necessarily characterizes situations. For example, the notion of a problematic situation requires Dewey to accept that a problematic character be identified by a concrete and determinate subject who has the experience (Mackay, 1942, p. 394). The difficulty may be formulated in the following question: “Would the situation still be problematic without a subject to disclose it?” If we answer yes, we are forced to wonder why Dewey chooses such terminology; for the word “problematic” implies someone to find it such. However, if we answer negatively the problematic character of the situation is subjective because the subject of the experience necessarily determines it. In

Feelings and the Self

105

addition, it may be argued that within Dewey’s description the problematic character of situations is dependent on the abilities of the subject to identify problems. According to Dewey, the objection relies on a confusion regarding the importance of the subject in the experience and on our bias towards objectivity. Dewey writes: Experience shows that as a matter of fact objective reference precedes subjective reference. Reference to a subject instead of an object is extrinsic and reflective. It is indeed only another mode of objective reference; that is, some tediousness of the object is accounted for in terms of an unusual state of the object. Otherwise to say “I am bored” and “It is tedious” are merely two phrases to express exactly the same fact. (LW 2, p. 91)

No doubt, with Dewey’s terminology, we can establish a distinction between the sentences “I am bored” and “It is tedious.” In fact, it is precisely because we can translate experience in terms of situations that we can draw such a distinction and imagine, for instance, both a scenario where one is bored without facing a tedious situation, and one where someone is not bored within a tedious situation. Therefore, Dewey’s terminology is not subjectivist but, on the contrary, allows us to escape the type of subjectivism that invalidates any exchange of information and makes criticism impossible. Dewey claims that the objection arises mainly because we take for granted that one gets bored when and only when a situation is tedious. We assume that when the situation is of a certain kind it accordingly affects the subject who experiences: if the situation is tedious the subject feels bored; if the situation is pleasant, the subject is pleased; if the situation is confusing the subject feels puzzled; etc. We may add that it is precisely this that makes us think that something is wrong when, for some reason, the subject does not have a certain disposition in a given situation; for instance, when a scary situation does not frighten the subject. For Dewey, asking the question “Whose experience?” carries another implication, namely that the nature of ownership is such that rational discourse about what is owned is unattainable. As Dewey explains, this implication is absurd for “it would be to infer from the fact that houses are usually owned, are mine and yours and his, that the possessive reference so permeates the properties of being a house that

106 

Dina Mendonça

nothing intelligible can be said about the latter” (LW 1, p. 179). We can talk about experiences and situations just as we can talk about the existence and the properties of a house independently of being owned. The quality of belonging is an additive that is important to note, both concerning a house or a situation, with respect to certain consequences. It can be argued that this additive is not as simple as it sounds. Hocking, who in general accepts Dewey’s description of experience, asserts that the addition of the subject is both a gain and a loss. He argues that, This addition of the psychological “I-think” gives us a more complete view of the whole situation. But – and here is the puzzle – with the addition something seems to have escaped. To assert “John is a rascal” is less complete than to say, “I think John is a rascal,” but it is more forceful: the prefix “I think” may relieve me from an action for libel, because in telling the truth more circumspectly it withdraws the force of assertion from the outer world. (Hocking, 1940, p. 412). Hocking’s objection allows us to state the importance of ownership more clearly. The notion of situation requires a more detailed description than one given by a mere sentence “I think John is a rascal.” In order to fully develop such an example we would have to describe what prompts such a statement: what John did, why he did it, when he did it, etc. The notion of “situation” helps us to see not only that ownership is not a private and exclusive affair, but also, that ownership is capable of denotation when the situation is fully described. For, as Dewey writes, “the quality of belonging to some one is not an all-absorbing maw in which independent properties and relations disappear to be digested into egohood,” (LW 1, p. 179) and “in some specifiable respects and for some specifiable consequences, these selves are capable of objective denotation just as are sticks, stones, and stars.” (LW 1, p. 179) What this discussion reveals is that Dewey’s notion of situations transforms the sense of subjectivity of experience into possible objective reference because the subject can be observed, named and described. When we need to ask “whose experience?” we have simply not described the situation accurately. Accordingly, experience is no longer something private, only accessible to the subject who has the experience. And Dewey’s notion of situation is a rich way to grasp the dynamics of emotional processes for it gives us the connection to the self without limiting such a connection

Feelings and the Self

107

to a private and inaccessible realm. In order to fully profit from the insights of Dewey into experience requires a more complex description of paradigm scenarios. The pattern of emotional activity that I will present in a very brief form aims at providing such theoretical vocabulary such that we may be able to focus our attention on the lively activities of mind instead of its results, and enable us to understand that the clear and compact words we have for some emotions do not exhaust the thought provoking reality of emotional life (Dewey LW 10, p. 49). ii. Self and The Pattern of Emotional Activity Before exploring the way in which we can map out the pattern of emotional activity it will be helpful to consider which way we should understand the notion of self so that we can more clearly understand its role in emotional activity and see how emotional activity contributes to the notion of self. In an overview article on the several notions of self, Gallagher (2000) summarizes the continuous ongoing reflection by philosophers, psychologists and other cognitive science experts, focusing on two important aspects of the self: the minimal self and the narrative self (Gallagher, 2000, p. 14) The Minimal Self is considered phenomenologically, that is, in terms of how one experiences the sense of self and how consciousness of oneself one takes oneself as an immediate subject of experience, unextended in time. As Gallagher writes, “the minimal self almost certainly depends on brain processes and an ecologically embedded body, but does not have to know or be aware of this to have an experience that still counts as a self-experience.” (Gallagher, 2000, p. 15) One of the aspects of the minimal self is that it is the home of the first-person pronoun “I” in its self-reference form such that it does not allow a mistake. That is, it has the feature of immunity to error principle, such that when a person refers to herself she can never make a mistake about this (Gallagher, 2000, p. 15). The other important conception of self that Gallagher identifies, the Narrative Self, is defined as “A more of less coherent self (or self-image) that is constituted with a past and a future in the various stories that we and others tell about ourselves.” (Gallagher, 2000, p. 15) The narrative self grants not only an abstract

108 

Dina Mendonça

entity of narration as an identity extended in time, but also a sense in which this abstract narrative is in a sense open (not only because future events transform past history) but also because, as Gallagher writes, “we cannot prevent ourselves from ‘inventing’ ourselves.” (Gallagher, 2000, p. 19) The design of the pattern of emotional activity, made in analogy to the Deweyan pattern of inquiry, provides a theoretical background to seeing how emotions and self (both minimal and narrative conceptions) interact with one another. After an explanation of the pattern of emotional activity it is possible to illustrate in which ways we can consider the self to be implicated, granted that there is considerable difficulty in clarifying the interrelations between the minimal self and the narrative self as Damasio’s work testifies (Gallagher, 2000, p. 20). I will go through the pattern quickly, pointing out the different moments with an illustration. Pattern of Emotion (Mendonça) 1. Settled adjustment 2. Indeterminate situation; disturbance 3. Identification of a Tension a. Feel b. Create families of emotions c. Compare/Contrast (other situations) d. Construct a Narrative e. Work out possible sequences 4. Assert Judgment (label situation) The first moment of the pattern (settled adjustment) aims to point to how things stand before the situation occurs, but this moment is somewhat false. That is, in order to conceive a situation one needs to be able to separate it and distinguish it from the flow of occurrences and, consequently, one needs to assume some type of settled flow of events where nothing is disturbed. However, as every moment of life is full of uncertainties, tensions etc, such that we can say that there are always several situations involved. Yet, when we think of a situation, for example the death of a close friend, this appears separated from the rest of experience. The moment before receiving the bad news (for example, imagine someone sitting in her living room reading a book) would be

Feelings and the Self

109

taken as the settled adjustment before the situation, and it is in this sense a settled adjustment is conceived. Nevertheless, the separation of something as a situation depends on being of relevance for the subject of the situation, though it does not require that the subject, the self, be conscious of that as a situation. The situation becomes indeterminate when this previous flow of events such as sitting in the living room reading a book is disturbed when the telephone rings and the person is told: “Roger died in a motorbike accident.” From the moment the person hears the words to feeling tension there is a space, that is, the tension is not immediate. So, the situation is indeterminate with respect to its issue (LW 12, p. 110), consequently the situation is not one of doubt (or fear, or sadness) in a subjective sense (LW 12, p. 110): at that point the situation is simply disturbed, confused, obscure (LW 12, p. 109). The pattern of emotion is felt as beginning with the recognition of a tension, that is, the indeterminacy is formatted in terms of a situation of tension. The identification of a tension is not a quick process and as the picture of the pattern illustrates it involves a process with many stages, and the sense of identification may vary in degree and may not require full consciousness. That is, the items “a-e” should not be seen as a rigid sequence or one that must be fulfilled completely in order for a situation to be identified as an emotional process. Instead, the identified items should be taken as a dynamic occurrence where one may return to the first step given (a: feel) after having gone through the process of (i) through (iii). In our example, it may be that from the moment she hears the words to feeling the tension there is a space – that is, the tension is not immediate. But once tension settles, she feels sad, wrapped up in the news. She begins to form a family of emotions: sadness, bewilderment, anger, pity, curiosity, etc. She compares and contrasts the news of his death with other situations of loss, and with memories of him (situations of no loss). She works out possible consequences of tension (future family events, future possible loses, what she will do next time she sees her aunt and uncle, etc.). By doing this she constructs a narrative; his death slowly becomes a story that points to future action: calling her aunt and uncle, remembering him more than she has done over the past year, etc. In Dewey’s pattern of inquiry the end is reached when the indeterminate situation becomes settled, but the pattern of emotion ends when

110 

Dina Mendonça

the initial tension is integrated. This requires some explanation. Integration is the assimilation of the emotional situation as an experience (to use Dewey’s terminology). Using my previous illustration, we can understand the notion of integration by seeing that though her cousin died a few days ago it may take her months, years to feel that grief is integrated. One acknowledges this sense of integration of grief by realizing that it takes time for people to be able to talk about someone they have lost. This is not only valid for the case of grief but one can also think of other emotional processes that sometimes take years for the person who lived through them as an experience to be able to talk about them. The pattern of emotion was made with the startling conclusion that within Dewey’s description emotions behave like ideas. But why would thinking through situations be helpful for a theory of emotions? There are many advantages to the philosophy of emotions (Mendonça, 2012) but here I simply want to concentrate on those advantages that allow a richer description between the self and the emotional process felt. Given the description of the pattern of emotion we can see that first, the pattern allows for the emotional character of a situation to change with reflection on it. This is because the construction of a narrative may transform the initial family of emotions of a certain situation such that a situation that is initially labeled as anger taken from the perspective of the minimal self, can be later labeled as sad taken from the narrative self perspective. For example, I may feel initially angry about a harsh reply from a friend but label the situation as sad by attaching it to a lack of communication between us months after the event occurred. This means that the person who forms part of the situational whole is also a potential source of change of the emotional situation. Consequently, we can consider personhood in its complexity allowing for continuity and simultaneously accepting the possibility of change. Second, the ability to identify change in situational processes in the pattern of emotion opens the way for an explanation of emotional maturity, since experiencing sad situations will hopefully change the impact of sadness when sadness is part of the family of an emotional situation of grief. That is, the narrative self-perspective will have an impact on the minimal self. Third, it provides a more complex way to explain how emotions resonate. Feeling like crying because I see someone else crying varies

Feelings and the Self

111

in intensity to the extent that the situation resonates with me. And the situation resonates with me depending on the person I am (have I felt similar things, does my “self ” incorporate this other person) and the person I see in an emotional situation. Given the interconnectedness that the pattern of emotion allows us to establish within a situational whole it may be interesting to consider that the way we conceive personhood, among other things, will be more accurate when thought of as the sorts of situations that the person has undergone and what type of expectations one has of future experiences. In fact, there is a part of our common sense description of other people that captures the connection made clear by the pattern of emotion: we refer to people saying, “Joanna is a solid person, someone you can trust in a situation involving danger,” or “John is at loss in situations involving change,” etc. What I hope the previous words have offered is that there is a dual open-ended connection between the self and person and the emotions. On the one hand, the emotional situations a person undergoes modify, change, fortify the person contributing to the ongoing creation of who we are both at the minimal and narrative level of the self; on the other hand, while someone undergoes an emotional situation they contribute towards the modification and strengthening of the paradigm scenario at hand. That is, while people are building and creating selfhood they are simultaneously constructing and perfecting paradigm scenarios.

Part II: Some Emotions and the Implicated Self In the second part of the paper I want to consider a few emotions and point to how their open-character is connected to the open character of the narrative self. This reflection does not aim to give a complete analysis of each of these emotions but to point out how emotional complexity implies certain types of openness. Ultimately this means that there are ways in which emotional processes are open to modification, and that these transformations are incorporated into the narrative self in such a way as to contribute to the ongoing invention of the self. Underlying

112 

Dina Mendonça

all the descriptions given in this second part there is the attempt to establish the open-ended character indicated above. Firstly, a short look at fear will show that even the emotions considered as simple may carry complexity and when we acknowledge them we realize that their open-endedness is only visible while we consider the narrative self and it is the self that is implicated. Secondly, when we look upon love we will recognize that part of this open-endedness is linked to contrary elements within the same emotional process. Thirdly, the look at pride and jealousy aims to point out that what we take as part of our self or not changes the emotional processes available to us, and that taking something as part of ourselves can result in different formats. i. Fear Now you may ask how it is that people are constructing these paradigm scenarios for fear is fear no matter what people undergo. Even if one accepts that certain emotions are so complex that they are constantly under change one would have to admit that the basic emotions are not changeable like that. Typically, the emotion of fear is portrayed as a quick response to a perceived threat: we see a snake and we flee, or shake, or scream. It is considered a basic survival mechanism that identifies dangerous situations. Yet we are scared of things like dogs, fire, snakes but also of bureaucracy, speaking in public and engaging in relationships, and though there is clearly something in common among these examples they also raise the insight that there is an element of complexity that is not grasped by the description of fear as a quick response to danger. I want to take the complexity of fear a step further by using a book published by the Portuguese philosopher José Gil entitled Portugal, Hoje. O Medo de Existir (Portugal, Today. The Fear of Existing) where, among other things, the writer considers the inheritance of fear such that this emotional state is no longer something felt in situations of danger but as part of the character of a person (in his book Gil refers to the character of the Portuguese people). This heritage of fear passes from parents to parents, generation to generation, similar to the way we have learn to react fearfully to sight of a snake. Gil claims that, “today, thirty

Feelings and the Self

113

years after the regime of fear, we continue to live with it. Portuguese society and the Portuguese have not lost the fear, even though (or maybe because of that) the new generations know little about the past of the Salazar period”1 (Gil, 2004, p. 78). The descriptions of fear that Gil provides currently existing in the Portuguese are not the things that typically come to mind in the philosophy of emotions, but they testify to how even in a basic emotion like fear, the way we learn is crucial for future experience of emotional situations, and how complex an apparent simple emotion can be. However, I do not want to deny the difficulty in changing and modifying the experience of fear, but to point out that fear may be a much more complex emotion if considered within the scope of the narrative self as José Gil’s book does. This complexity points to an open-ended aspect of emotional processes, namely that the inheritance of emotional processes may be such that emotions become part of an identity (self or nation) as a background emotional mood, but this is only visible when we consider the self in its narrative format. ii. Love It is arguable whether love can qualify as an emotion but it clearly is an emotional process that is felt by the minimal self, such as when a mother looks at her children, and felt by the narrative self as when the mother looks at how she has loved her grown up children. The mysterious nature of love is a witness to the complexity that surrounds the emotional process of love. Maternal love is unquestionable, yet I once heard a son tell his mother about the issue of hitting his children: “It’s okay to discipline them, as long as they know you love them.” In fact some people actually believe that it is not possible to discipline without doing this. Fortunately, this method has slowly become unacceptable. But what I want to consider here is that the love of a mother, such an unquestionable type of love, has been created as an image associated 1

My translation of the original quotation: “Hoje, trinta anos depois do fim do regime do medo, convivemos ainda com ele. A sociedade portuguesa, os portugueses não perderam o medo, ainda que (ou talvez por isso) as novas gerações pouco saibam do passado salazarista.”

114 

Dina Mendonça

with a type of behavior that one would classify as opposed to loving. That is, when someone thinks of the love of parents, in the abstract or concretely, it is seldom the case that one thinks of times or moments of hitting, or speaking violently, etc. Yet, the relationships with parents are the unavoidable proof that this sentiment of so-called love holds much more other feelings than loving ones. There is no doubt that the way we have first learned to experience such complex emotional situations such as living with the ones we love and share routines with, combine wishes and desires, and negotiate tensions and problems, must determine the way we, as adults, form and continue relationships with our children. The importance of identifying this inescapable characteristic of the emotional process of love is that it makes us realize that while we are not free to change the past, the future lies open-ended and its open-endedness is related to how we view what has happened to us. That is, the character of the emotional process changes when the person of the situation changes her perspective – that is, the one that is able to identify the hidden contradictory feelings within an emotional process may be capable of renewing their emotional processes (for instance, avoiding hitting as a means of protection in love and finding other ways to deal with difficulties). It is not a problem in itself that situations of love include elements of power struggles, fights, irritations, etc. for it is not necessarily problematic that emotions frequently incorporate contradictory elements even when we are not aware of this. In sum, the contradictory feelings that may lurk within emotional processes grant us ways of modify the feeling of those same processes. iii. Pride Pride requires a lot of cognitive processing related to the self and overall results from a positive self-evaluation (Lewis, M., Takai-Kawakami, K., Kawakami, K., & Sullivan, M. W., 2010). It is considered one of the self-conscious emotions and is associated with behavioral cues that have innate elements that are automatically perceived by others (Shariff AF, Tracy JL., 2009). The positive self-evaluation of pride

Feelings and the Self

115

results in seeing the self and the object as separate. As Michael Lewis writes, The emotion I have labeled “pride” is the consequence of a successful evaluation of a specific action. The Phenomenological experience is joy over an action, thought, or feeling well done. Here the focus of pleasure is specific and related to a particular behavior. In pride the self and the object are separated, as in guilt. Unlike shame and hubris, where subject and object are fused, pride focuses the organism on its action; the organism is engrossed in the specific action that gives it pride. (Lewis, 2004, p. 630)

However, the separation between subject and object is connected to the self in a relevant way such that what is considered the object, standing in relation to the subject belongs to it in a narrative sense. A person who is proud of her country’s achievements reveals a sense of self that requires a connection to her country, as opposed to a person who does not feel part of her country. In this sense the variety of things that can be the object of pride (oneself, an action, a member of the person’s family, a person’s children, an action by a person’s children, other people, a person’s possessions such as a house, history, etc., a person’s connections such as her country, her friends, etc.) vary to the extent that one takes something as part of one’s sense of self, and reveals the open character of emotional processes through the change of valence it holds. While it is mostly considered something to feel and there is also a tradition of seeing it as a virtue, it stands for a sin and negative emotion in many religious traditions. The point of this last remark is not to make a judgment on how to consider pride but to point out that the valence given to it clearly depends on the type of narrative self at hand. iv. Jealousy Finally, I want to take a very short peek at jealousy because in a way it stands at the opposite side of pride concerning the relation of the self with something other than the self. While with pride subject and object stand apart, in jealousy subject and object are united. In Jealousy what stands in relation to the self is taken as interior and belonging to the self, not through a relevant connection but it is taken to belong to the self,

116 

Dina Mendonça

such that a threat to this belonging anticipates a sense of loss of something the person values as part of herself, such as friendship, or love, or a working relation. Similar to the case of pride, the way the narrative self is understood and the way the emotional situation is understood will change the existence of jealousy, its impact and its design.

Conclusions What the paper here presents is a suggestion that the notion of an implicated self should be examined in more detailed so as to find some of the ways people free themselves from certain implications or how they create new ways of being implicated. Though it may be clear that the boundaries between exterior and interior to the self are not fixed, the very existence of emotional processes seems to indicate that patterns of establishing borders exist and that these may be transformable. In addition, if the first part of the paper holds true then the connection of being a self and learning to speak about emotional processes may hold more interesting insights than the ones that have so far been offered concerning the connection between language and emotional processes. Finally, hopefully the paper has made clear that it would be most suggestive to produce a map of different conceptions of the self with the connections to their different theoretical consequences regarding the experience of emotional processes (for example, the Stoic perspective of self admits the possibility of not feeling grief, yet for Rousseau that would possibly be the mark of a cruel self). It may be the case that our philosophical heritage may provide some keys for the connection between the contemporary Minimal self vs. the Narrative Self, for somewhere in our past reflection may lie the answers for our present models.

Feelings and the Self

117

References De Sousa, R. (1987). The Rationality of Emotion. Cambridge, MA: MIT Press. Dewey, J. The Early Works, (1882–1898). Edited by Jo Ann Boydston. 5 vols. Carbondale and Edwardsville: Southern Illinois University Press, 1969–1972. Dewey, J. The Middle Works, (1899–1924). Edited by Jo Ann Boydston. 15 vols. Carbondale and Edwardsville: Southern Illinois University Press, 1976–1988. Dewey, J. The Later Works, (1925–1953). Edited by Jo Ann Boydston. 17 vols. Carbondale and Edwardsville: Southern Illinois University Press, 1988–1991. Faber, A. & E. Mazlish, (1982), How To Talk So Kids Will Listen & Listen So Kids Will Talk, New York: Avon Books. Gallagher, S. (2000). Philosophical conceptions of the self: implications for cognitive science. Trends in Cognitive Science, 4(1), 14–21. Gil, J. (2004). Portugal, Hoje. O Medo de Existir. Lisboa: Relógio d’Água. Hocking, W. E. (1991). Dewey’s Concepts of Experience and Nature. In John Dewey The Later Works, 1925–1953. Edited by Jo Ann Boydston, 14, Carbondale and Edwardsville: Southern Illinois University Press, 411–426. Lewis, M. Takai-Kawakami, K., Kawakami, K., & Sullivan, M. W. (2010). Cultural differences in emotional responses to success and failure. International Journal of Behavioral Development, 34(1), 53–61. Lewis, M. (2004). Self-Conscious Emotions: Embarrassment, Pride, Shame and Guilt. In The Handbook of Emotion, Third Edition, Michael Lewis, Jeannette M. Haviland-Jones and Lisa Feldman-Barrett (eds.), New York, NY: Guilford, 623–636. Mackay, D. S. (1991). What Does Mr. Dewey Mean by an ‘Indeterminate Situation’? In John Dewey: The Later Works, 1925–1953. Edited by Jo Ann Boydston, 15, Carbondale and Edwardsville: Southern Illinois University Press, 393–401.

118 

Dina Mendonça

Mendonça, D. (2012). Pattern of Sentiment-following a Deweyan suggestion. In Transactions of Charles Peirce Society, Vol. 48, No. 2, Spring 2012, 209–227. Mendonça, D. (2003). “When Heads Happen – Exploring the Deweyan Conception of Mind”, presented at SIFA October. Mendonça, D. (2003). Human Nature in the Making. Cadernos de Filosofia, IFL, ed. Colibri, nº 13, 29–41. Shariff A. F., Tracy J. L. (2009). Knowing who’s boss: implicit perceptions of status from the nonverbal expression of pride. Emotion. 9(5), 631–9. Talisse, R. B. (2000). On Dewey: The Reconstruction of Philosophy. Belmont, USA: Wadsworth/Thomson learning.

Part III Cognition, Psychology, Neuroscience

De Se Attitudes and Semiotic Aspects of Cognition Erich Rast1*

Overview In this article, I will re-examine some of the classical puzzles for de se attitudes that have been laid out by Hector-Neri Castañeda, David Lewis and John Perry in various articles and compare them with Jackson’s Knowledge Argument. The origins of these puzzles go further back to work by Russell on egocentric particulars, by Frege in ‘Der Gedanke’, Wittgenstein’s considerations on subject-uses of I in the Blue Book, and work by Roderick Chisholm. Nevertheless, it is fair to say that de se puzzles gained widespread popularity only later due to publications by Castañeda (1967), Perry (1977, 1979), and Lewis (1979). The article starts with a survey of well-known de se puzzles: Perry’s supermarket example, his Rudolf Lingens example, and David Lewis’s Two Gods thought experiment. I will then discuss Jackson’s Mary example, which bears a striking similarity to de se puzzles. After this primarily exegetical part, I will address the question regarding what these puzzles have been taken to show and what they really show. My central thesis is that typical de se puzzles reveal an important and epistemically irreducible aspect of thinking, but do not allow for any conclusions regarding physicalism and the Mind-Body problem. As I will argue, there is a special kind of introspective knowledge, the existence of which is fully compatible with physicalism and this special kind of *

An earlier version of this paper was presented at the conference ‘Metaphysics of the Self ’ at the Institute for the Philosophy of Language (IFL), Lisbon, in December 2009. I would like to thank Klaus Gärtner, Jorge Gonçalves, Franck Lihoreau, Daniel Ramalho, António Marques and Dina Mendonça for fruitful suggestions and commentaries. This research was conducted under a postdoctoral fellowship from the Portuguese Fundação para a Ciência e Tecnologia.

122 

Erich Rast

knowledge results from the fact that all sorts of episodal thoughts plays a particular role in thinking and cannot be replaced by other kinds of thoughts. On the basis of this insight, I will suggest a trivializing interpretation of de se puzzles: A thought of a certain type, say α, is a necessary condition for the occurrence of a corresponding α-behaviour or action, simply because episodal thought tokens are divided into distinct classes according to the role they play in cognition. Correspondingly, it is highly unlikely that thoughts of type α, which present subjective experiences in cognition to the one currently thinking, could play the same role in that person’s thinking as thoughts of another type, say β, which present physical knowledge to the one currently thinking, and vice versa. This does not mean, however, that physical knowledge cannot explain thoughts of type α or that instances of α and β belong to different ontological categories. Before going on, some terminology must be clarified. Talk about thinking is often ambiguous between thinking in an episodal sense and dispositional thinking in the sense of having the ability to entertain certain thoughts or having a dispositional belief that p for some embedded proposition p. In what follows, I have the former in mind when talking about thinking here; this kind of thinking might also be called cognition in order to set it apart from the dispositional reading. Cognition is timebound and actual. In contrast to this, in what follows belief and other propositional attitudes should be understood in the common dispositional sense. When an agent thinks (viz. cognates) that p this means that he currently entertains a p-thought or is having a p-thought. In contrast to this, when someone believes that p he has a disposition to act as if p were the case. Without further qualification the terms episodal thinking and cognition leave the question open whether the respective agent having the thought endorses the embedded proposition or not, i.e. whether she considers the embedded proposition true or not. I shall understand these terms in their non-philosophical sense in what follows, according to which the agent indeed takes the embedded proposition to be true. Understood in this sense, when someone thinks that John is 32 years old he also dispositionally believes that John is 32 years old and does not just ‘think the thought’ without being committed to its truth. The converse does not hold. From the fact that someone dispositionally believes that p it does not follow that he thinks that p at a certain time. I

De Se Attitudes and Semiotic Aspects of Cognition

123

will often talk about dispositional knowledge instead of belief, since the kind of beliefs about one’s own thoughts I will discuss result in some form of introspective knowledge as long as the person in question does not suffer from serious mental illness. My main points could be made entirely in terms of belief instead of knowledge, though, and so not too much weight should be given to this terminology.

Puzzles of De Se Attitudes The supermarket example laid out by Perry (1979) is one of the clearest cases for the de se attitudes: I once followed a trail of sugar on a supermarket floor, pushing my cart down the aisle on one side of a tall counter and back the aisle on the other, seeking the shopper with the torn sack to tell him he was making a mess. With each trip around the counter, the trail became thicker. But I seemed unable to catch up. Finally it dawned on me. I was the shopper I was trying to catch. (Perry, 1979, p. 3)

This scenario has two different, though related aspects. On one hand, Perry might cognate in various ways about himself without realizing that his episodal thoughts are about himself. For example, he might think that someone/the only bearded philosopher in a Safeway store west of Mississippi/John Perry/this man (+pointing gesture) in the mirror is making a mess without realizing that it is he himself whose sack of sugar is torn. As long as John Perry is sufficiently amnesiac, doesn’t remember his name or that he hasn’t shaved himself in the morning, and doesn’t recognize himself in a mirror, none of these ways of thinking about himself seems to explain his behaviour, until he starts to think: It is me, who is making a mess, I am producing a trail of sugar! On the other hand the example also has a linguistic aspect. The different ways in which John Perry might linguistically realize his episodal thoughts don’t seem to have the same explanatory power for his action than Perry’s utterance of I am making a mess. For example, in order to express the same thought using his proper name, John Perry would have to additionally believe that he is called John Perry, which he doesn’t believe

124 

Erich Rast

in the given scenario. This has led Perry and others to the conclusion that there is an essential reading of the first-person indexical that is irreducible with respect to its power for explaining certain changes in behaviour and actions. Another well-known example can be found in Perry (1977): An amnesiac, Rudolf Lingens, is lost in the Stanford library. He reads a number of things in the library, including a biography of himself, and a detailed account of the library in which he is lost. He believes any Fregean thought you think might help him. He still won’t know who he is, and where he is, no matter how much knowledge he piles up, until that moment when he is ready to say, ‘This place is aisle five, floor six, of Main Library, Stanford. I am Rudolf Lingens.’ (Perry, 1977, p. 492).

Again, the problem in this example is that it always seems to be conceivable that the epistemic agent in question gathers all kind of knowledge about himself from external evidence, but only when he realizes that this information is about himself will he be able to act accordingly. So it appears as if, from an epistemic point of view, the agent learns something when he realizes that he is John Perry or Rudolf Lingens respectively, he himself is making a mess on the floor or he himself is located at aisle five, floor six, of Main Library, Stanford, and so on. Lewis (1979) has provided a famous variant of these examples that, albeit being highly contrived, is rather instructive from a logical point of view, because it is spelled out in terms of possible worlds semantics for propositional attitudes like dispositional belief: Consider the case of two gods. They inhabit a certain possible world, and they know exactly which world it is. Therefore they know every proposition that is true at their world. Insofar as knowledge is a propositional attitude, they are omniscient. Still I can imagine them to suffer ignorance: Neither one knows which of the two he is. They are not exactly alike. One lives on the top of the tallest mountain and throws down manna; the other lives on top of the coldest mountain and throws down thunderbolts. (Lewis, 1979, pp. 520–521)

The idea behind this thought experiment is as follows. In epistemic logic based on normal modal logic an agent’s epistemic state is represented by the set of possible worlds, i.e. maximally truth-making doxastic alternatives, that are accessible from the actual world by an accessibility relation for that agent and the respective kind of propositional attitude.

De Se Attitudes and Semiotic Aspects of Cognition

125

Suppose the belief set Γ is the set of possible worlds accessible by an agent’s current belief relation from the actual world. Consider now a case when the agent is in doubt whether p is the case or not. In a possible worlds setting, this boils down to saying that there are some w1, w2 in Γ such that p holds at w1 and ¬p at w2.When the agent learns that p is the case, the revised accessibility relation will yield a set of possible worlds from which all ¬p worlds have been removed. The more an agent learns about the universe, the smaller becomes his belief set. If only one world is left and as in Lewis’s scenario that world is the actual world, then the agent is omniscient (see Figure 1 for illustration). Whatever he can learn about the state of nature he has already learned. Nevertheless, Lewis claims that for example Zeus in this scenario still doesn’t know that he himself is Zeus, that he himself is sitting on the tallest mountain and is throwing down manna, and so on. Lewis has devised this scenario to illustrate a limitation of the usual truth-conditional semantics for dispositional attitudes based on modal logics with Kripke or Scott-Montague semantics. He concluded from the thought experiment that possible worlds alone do not suffice to represent what an agent might learn, hence do not suffice to represent epistemic states and corresponding dispositional attitudes in general, and for this reason he proposed a more fine-grained model in which having a relational attitude is modelled as the self-ascription of a property by the respective agent. Since properties in Lewis’s view are more fine-grained than sets of possible worlds, more attitudes can be distinguished by the propertyascription view and so his approach can adequately represent the assumption that the two gods in the scenario might have different epistemic attitudes despite being omniscient about all the ‘external’ facts. Lewis’s solution was not very appealing to many philosophers, and rightly so, because he did not present enough details about the properties he had in mind. Lacking a detailed and positive metaphysical account of properties his suggestion remains unsatisfactory. Fortunately, many other solutions to the Two Gods puzzle have been proposed during the past few decades, and, technically speaking, any way to make dispositional belief more fine-grained will do as long as one merely strives for descriptive adequacy with respect to de se puzzles. One of the simplest solutions, which was also discussed by Lewis (1979), is using centred possible worlds. Instead of bare possible worlds ordered pairs

126 

Erich Rast

consisting of an agent and a possible world are taken as semantic base entities. With this small change the lack of omniscience of the two gods can be easily represented. For example, Zeus’s belief set {‹w, Zeus›, ‹w, Jahwe›} would contain two tuples, one containing himself and the other one containing Jahwe as long as he hasn’t realized that he himself is Zeus. When he learns that he himself is Zeus he removes the pair containing Jahwe from his belief set (see Figure 2). There are many other approaches to hyperfine-grained attitudes that may be used for dealing semantically with Lewis’s scenario: structured propositions from von Stechow (1982) and Cresswell (1985), approaches based on substituting reified propositions for truth-values as in Thomason (1980), using impossible worlds as in Hintikka (1975), and property theory of Bealer & Mönnich (1989). To make a long story short, it is fair to say that de se puzzles nowadays no longer pose any particular technical challenge to a logician who is interested in a descriptively adequate truth-conditional semantics for belief and similar attitudes. However, none of these approaches give a satisfactory answer to the question of what exactly the agent learns when he realizes that he himself is called by a certain name or satisfies a certain property. The power of Lewis’s example can be illustrated further by thinking of possible worlds for a moment in an old-fashioned way as possible constellations of matter and the forces that hold it together. By assumption, in the example both gods know everything there is to say about the constellation of matter in the universe and the forces that hold between matter. Still, so it is claimed, they can learn something. Consequently, whatever there is left for them to learn cannot be an aspect of matter and the forces between that matter. As Stalnaker (2004) remarks, there is a striking similarity between de se puzzles of this kind and the Knowledge Argument by Jackson (1982, 1986), which was originally intended to show that qualia are not physical and therefore physicalism is false. Jackson (1986) has laid out this example in the following, much cited passage: Mary is confined to a black-and-white room, is educated through black-and-white books and through lectures relayed on black-and-white television. In this way she learns everything there is to know about the physical nature of the world. She knows all the physical facts about us and our environment, in a wide sense of ‘physical’ which includes everything in completed physics, chemistry, and

De Se Attitudes and Semiotic Aspects of Cognition

127

neurophysiology, and all there is to know about the causal and relational facts consequent upon this, including of course functional roles. If physicalism is true, she knows all there is to know. […] It seems, however, that Mary does not know all there is to know. For when she is let out of the black-and-white room or given a color television, she will learn what it is like to see something red, say. (Jackson, 1986, p. 291)

Just like the gods in Lewis’s scenario Mary is omniscient about the external world, yet she doesn’t know what it feels like to experience red, before she has been exposed to a red object. In de se puzzles the agent still needs to learn that it is he himself/she herself who has a certain property or is called by a certain name. According to the Knowledge Argument, the agent is also supposed to learn something in addition to her physical knowledge about the world, namely the way in which a certain experience feels to her. So do these thought experiments indicate that there is something nonphysical like an irreducible I-thought or a particular red quale? As I will argue in the following section, the answer to both questions is No. Neither de se puzzles nor corresponding arguments for qualia show that the physical world is not closed or that certain phenomena in episodal thinking are not reducible to physical states. Both kinds of puzzles show, however, that there is a special sort of introspective knowledge which is fully compatible with physicalism and based on remembering past experiences or thoughts. In order to have this kind of knowledge, an agent must store an experience or thought in a way such that the retrieval of this memory fulfils a certain role in his thinking; when something else is stored instead, it might not fulfil the same role when it is retrieved. Investigating this kind of knowledge in combination with some general semiotic considerations will pave the way for a trivializing interpretation of de se puzzles.

Assessment of the Thought Experiments It is worth noting that the Two Gods thought experiment does not strictly speaking present a positive argument for the thesis that there are de se attitudes. Instead of acknowledging that each of the gods lacks some

128 

Erich Rast

knowledge one could also take the example as illustrating that the gods do not lack any knowledge, since there are no epistemic alternatives left for them to consider. By assumption Zeus knows everything there is to know about the physical world, hence one might argue that he also knows who he is, what his name is and which properties he has, that it is himself who is currently thinking, whether he is throwing down manna or thunderbolts, and so on. If the mental supervenes on the physical, the defender of physicalism might argue, then Zeus also knows everything there is to know about his mental life and consequently does not lack any knowledge about his own episodal thoughts either. From this point of view Lewis’s premise that the two gods lack knowledge could be regarded as false. Let us call this the denial reply. It does not seem to be very convincing. On the one hand, the thought experiments have an intuitive power that lures us into believing that the agent can indeed learn more. Using thought experiments in this way has rightly been criticized by Dennett (1991) as intuition pumping in an area where our intuitions may be utterly misleading. On the other hand, the denial reply neglects an important distinction between the static theoretical knowledge an agent has at a certain time and the knowledge an agent might gain when he or she has an insight about him/herself. Nothing in the description of the scenario warrants that one of the gods actually has the insight that it is he himself/she herself who has such-and-such properties. Jackson’s Mary example has been criticized by Dennett (1991) in a similar fashion as in the denial reply to the Two Gods example. According to Dennett, Mary does not learn anything new when she is exposed to a red object for the first time and we only find this claim counter-intuitive because we cannot adequately imagine what it means to have exhaustive knowledge about the physical world. Many other replies have been given to the original argument, ranging from a critique of ‘intuitions’ to the denial of one or more of its premises, and Jackson himself has by now converted to physicalism. For example, according to the well-known ability hypothesis Mary does not gain new knowledge but an ability – see Lewis (1983, 1988/1990), Nemirow (1980, 1990), and Churchland (1985, 1989), cf. Coleman (2009) for a comprehensive critique. Others such as Conee (1994) and Tye (2009) have defended the so-called acquaintance hypothesis according to which acquaintance with certain objects or experiences cannot be explained in terms

De Se Attitudes and Semiotic Aspects of Cognition

129

of knowing that, or abilities. I will in the following paragraphs lay out a position according to which (a) in the classical de se puzzles and in Jackson’s Mary example the agent in question learns something new in the sense of gaining new knowledge, but (b) neither de se scenarios nor Jackson’s Mary example speak against physicalism, (c) the examples are described incorrectly, since the world changes in each of them, and (d) episodal thoughts of a certain kind can trivially not be substituted by episodal thoughts of another kind and dispositional knowledge cannot replace episodal thinking. In a nutshell, what I claim is that in order to know the episodal thought that you’ve been thinking (or are just thinking), you need to think it first, and thinking a certain episodal thought is trivially different from an explanation of that thought or an episodal thought that explains that thought. To back up these claims let me start with an analogy. Imagine a little robot toy that contains sensors that react to light. A sensor function s: T→{l, r, f} from time to three values works as follows: When there is more light to the left than to the right, the sensors return l; otherwise if there is more light to the right than to the left they return r; otherwise they return f. The robot automatically stores each sensor input in its memory and can also perform three actions: move forward F, move forward left L, or move forward right R at a time t. The device is programmed as follows: (1) Store s(t) in memory. (2) If s(t)=r then R, else if s(t)=l then L; otherwise F at discrete times t measured by an internal clock. If designed correctly, this little device will follow the strongest light source until it bumps against something, malfunctions, or its battery runs out. In its memory a certain sequence is stored, and which one depends on its environment. If, for example, the robot is never exposed to light from its left, it may go in circles to the right and the symbol l will never be stored in its memory. Now suppose your colleague from the local A.I. department wanders by and laments that this device is much too simple. He upgrades the robot’s software and hardware to allow for running complex programs on it, including conditional actions based on the contents of the memory and devising simple plans such as ‘If the last sequence of actions was llrll, then R.’ Despite this gain in abilities and ‘cognitive powers’, as long as the robot is never exposed to light from the left none of its routines that introspect l states and act conditionally upon them will do anything. Suppose further that

130 

Erich Rast

your colleague from the A.I. department becomes famous by inventing a rather complicated program that, when hooked up with sensors and running on suitable hardware, becomes conscious and self-aware. (As a philosopher, you remain rather sceptical about this conjecture but the dispute ends in a stalemate: neither can you falsify his claim that the running program is conscious nor can your colleague falsify your claim that it is not conscious.) Out of generosity your colleague upgrades your robot toy, uploads the program, and immediately once the program is started the robot begins to explore the world. Being equipped with sophisticated A.I., this robot bears some resemblance to RoboMary described by Dennett (2005, pp. 122–129), but it illustrates a different point here. The point is this: Even though our robot is now conscious, self-aware, and capable of sophisticated planning and other higher cognitive abilities, what it can experience, learn, and do is restricted by its environment. If you mount a strong light on its right side, the robot will never store the symbol l. Since the robot is conscious, this means that it will never have an l experience, or, to put it in other words, the robot will have no possibility of knowing how it feels like to have an l experience. This example shows two things. First, it is obvious that the robot does not lack any ability when it is never exposed to light from the left. It has the same abilities when it is never exposed to light from the left as it has when it is exposed to light from both sides and does not lack any know-how in either case. So the so-called ability reply to the Knowledge Argument does not seem to be convincing if the above example illustrates basically the same problem as Jackson’s Mary example. Secondly, there does seem to be a sense in which the robot learns something new when it is exposed to light from the left for the first time, even though its blueprint and the software running on it doesn’t change. It has a certain experience that it has never had before and can now store the fact that it has had this experience in its memory. It acquires memory of a past experience, which is, by definition, a learning process. Keeping the robot in mind, let us return to the Mary example. Does Mary learn something new when she is exposed to a red object? The answer is clearly Yes. She can now remember an experience that she didn’t have before. Does that mean that qualia are not physical or that there is some irreducible phenomenal knowledge? The answers which one gives

De Se Attitudes and Semiotic Aspects of Cognition

131

to these questions depends on the stance one takes about red experiences viz. l-experiences. Someone who thinks that the robot can only have an l-experience if it is conscious, i.e. consciousness is a necessary condition of having an l-experience, would without doubt be inclined to say the same about Mary. In this view, which has been criticized by Dennett and others for its reliance on unstable intuitions, Mary learns something new when she is exposed to red, but only if she is conscious. In this point of view, machines cannot be conscious as they merely follow a fixed set of instructions. So when our robot acquires knowledge and eventually becomes just as brilliant a scientist as Mary, it will still not have an i-experience even if it is exposed to light from the left. Notice, however, that neither of the thought experiments supports this view or its opposite. Certain types of computational structures, when running on some hardware, or certain types of physical structures could be conscious without anyone ever knowing for sure that they are. Suppose this were the case. Then there would still be a difference between Mary’s epistemic state before she has been exposed to red and afterwards and the robot’s epistemic state before it has been exposed to light from the left and afterwards. Moreover, suppose the robot was omniscient about the physical world and not conscious. Still, the robot’s epistemic state would be different before and after it has been exposed to light to the left, because it can store l-input from the sensors, which is a learning process, and being able to remember a past input is a kind of knowledge. But in order to be able to remember a past input or experience the respective agent needs to have had it. So there is indeed a special kind of knowledge at play that one might label as ‘phenomenal’, but the conditions for acquiring this knowledge are entirely independent from the question whether qualia are physical or not and from the question whether an agent is conscious or not. If qualia are nonphysical and exist, then the occurrence of a nonphysical phenomenon is a necessary condition for having knowledge of qualia experiences, but the Knowledge Argument does not show this. The Knowledge Argument merely shows that certain knowledge can only be acquired by having certain experiences, be these reducible to physical structures or not. A very similar point can be made about the Two Gods example. Since each of the gods is omniscient about the physical universe, they do know who they are and are able to identify themselves. However, there is

132 

Erich Rast

still a sense in which each of them lacks knowledge, as long as they do not make use of their ability. For instance, unless Zeus thinks the episodal thought I am Zeus he cannot get to know that he has been thinking that thought. Regarding his dispositional belief this means that the occurrence of the episodal thought I am Zeus is a necessary condition for his having the dispositional de se belief that he himself is Zeus. The thought has to occur in actual cognition first, and only after it has occurred can the agent have the corresponding dispositional de se attitude. If this is so, then there are two ways to interpret the puzzle. First, one may claim that as a consequence of having complete theoretical knowledge about the universe an agent will invariably come to episodically think corresponding thoughts of the form I am F. Since we are talking about arbitrary properties F that the agent possesses, this view does not seem to be very compelling. If at all, it only makes sense under very strong rationality assumptions and when the agents in question are highly idealized like the two gods. Since no actual, resource-bound agent can have infinitely many episodal thoughts, and thus every actual agent lacks knowledge about herself, another response seems to be more appropriate. According to this view, the complete theoretical knowledge of the agent in question does in itself not warrant that she has corresponding thoughts of the form I am F. If the agent had such a thought, then she would be able to recognize that she has had it even from a purely 3rd-person perspective, provided that episodal thoughts are entirely reducible to certain kinds of physical structures, but since having the dispositional knowledge does not cause or otherwise warrant that the agent actually has an episodal thought of the kind needed for having a corresponding de se belief, she might not recognize that she herself is F. If, on the other hand, episodal thoughts are not entirely reducible to certain kinds of physical structures, then an agent that is omniscient about the physical world might not recognize from a 3rd-person perspective that she has had that thought. In both cases, however, the agent can only recognize that she has had such a thought after she has actually had the thought. Again, the puzzle does not decide anything about the ontological status of the mental. All it says is that you need to actually think certain thoughts in order to be able to retrieve from memory that you have thought them, and, as in case of the robot, the ability to retrieve memories of past mental events, be they ultimately physical

De Se Attitudes and Semiotic Aspects of Cognition

133

or not, can be considered a form of knowledge. This sort of knowledge may be called introspective knowledge.

Semiotic Aspects of Cognition From what has been said so far, a number of conclusions can be drawn. Firstly, the Two Gods and Jackson’s Mary example are fully compatible with physicalism and neutral about the question whether physicalism holds or not. Secondly, the descriptions of the examples are incomplete. Suppose physicalism is true and Mary is experiencing something red. Then her brain state changes, too, and consequently her previous knowledge about the physical world is outdated. Likewise, if physicalism is true and an agent has the episodal thought I am F, then having this episodal thought corresponds to a change in the agent’s brain state, rendering his previous knowledge incomplete. A static picture of the universe, as is presumed by using some simple, tenseless possible worlds semantics for propositional attitudes, is not entirely adequate for describing these scenarios. When for example Mary’s knowledge is updated according to the changes in the world after she has been exposed to a red object for the first time, she would be able to deduce from her physical knowledge alone that she has had a red experience. Nevertheless, as simple as this may sound, the fact that she has had a red experience is a necessary condition for her to realize, on the basis of physical knowledge about the world only, that she has had a red experience. As I will lay out in more detail below, even under the premise that physicalism is true, recognizing the red experience by means of measurement and physical knowledge and subsequently forming knowledge about its recognition by remembering it is not the same as remembering the red experience itself. This position is similar to what has been proposed by defenders of what Nida-Rümelin (2009) calls the new knowledge/old fact view but with one rather crucial difference: in the present view, the fact is new, although it will be possible to fully explain it in physical terms if physicalism is true. If on the other hand physicalism happens to be false,

134 

Erich Rast

then having a red experience trivially amounts to a new phenomenal fact, namely the fact that the person in question had this particular red experience. In both cases, the world changes and so the knowledge of the omniscient agent in question needs to be updated. What about the updated knowledge then? Does the updated dispositional knowledge suffice to explain a particular red experience or an agent’s insight that he himself or she herself has a certain property? The correct answer is: Why not? Why should the updated knowledge not suffice to explain the respective experiences provided that physicalism is true? Explaining in this sense means nothing more than knowing the conditions under which the agent in question has the respective experiences and therefore is able to form introspective knowledge about them. However, there is no reason whatsoever to believe that dispositional knowledge of an agent or someone else about an agent’s experiences or thoughts can substitute or replace in any way the agent’s having certain episodal thoughts that present experiences, insights or memories thereof to him/ herself in his/her cognition. Even under the assumption that certain episodal thoughts directly correspond to dispositional, physical knowledge about a particular red experience this red, i.e. when we would confine further considerations to whatever corresponds to dispositional knowledge about this red in actual cognition – like the episodal thought that the episodal thought this red is such-and-such –, there is no reason whatsoever to believe that these kinds of episodal thoughts could or should play the same role in cognition as the thought this red that an agent has when he is confronted with a red object or the thoughts he might have when he imagines or remembers a red object. Likewise, various ways of identifying oneself in actual cognition in 3rd-person ways might not play the same role in cognition as the thought circumscribed as being of the form I am F that occurs in cognition when an agent realizes that he has the property F. The resulting perspective on de se attitudes is trivializing. Certain episodal thoughts are necessary conditions for a certain kind of behaviour, whereas other sorts of episodal thoughts are not – no matter what we know about episodal thinking and how it should be described. An explanation of our robot’s inner workings does not substitute the machine’s actual having and processing certain sensor inputs or fetching data from its memory even if the robot itself processes this explanation

De Se Attitudes and Semiotic Aspects of Cognition

135

and accurately measures its own state. Correspondingly, Mary’s knowledge about the physical world cannot substitute her having a red experience. Generally speaking, episodal thoughts of a certain sort play a certain role in thinking that thoughts of another sort might not. This thesis should be relatively uncontroversial, yet it has been overlooked consistently in the assessments of respective thought experiments. Episodal thinking has another important property that is relevant for the assessment of the examples: it does not work like a language. Too see why, consider the robot example again. In a well-functioning robot an l symbol occurs when the left sensor receives more input than the right sensor, and correspondingly for the processing of r and f symbols. The symbols are stored in a memory module of a given capacity. Suppose now that the robot’s program says ‘When the memory contains the sequence lll, remove it from memory and execute R for the next two time units.’ Another part of the program could possibly use another symbol, say a, to refer to a sequence lll. However, in order for the symbol a to have the same effect as in the above programming instructions it has to be assigned in an appropriate way to sequences of lll’s stored in memory; no matter how this connection looks like in detail the robot must remove three l’s from memory and give the R signal for the motors to go right for the next two time units in order to run the same program on the basis of the alternative representation a. In this respect, although the choice of the symbol l is completely arbitrary, a sequence of three l’s in memory is the most basic representation of three inputs from the left sensor on this machine with respect to the particular program running on it. While it would be possible in this example to replace three l symbols with a single one such as a, which is stored whenever the left sensor has received three subsequent inputs in a row, this strategy only works because the program only acts upon three received l symbols. It does not work in general. Suppose the robot gets the following, more complicated instructions: ‘Whenever there is a sequence of n symbols l followed by a sequence of n/2 symbols r, remove the whole sequence and drive right for n time units.’ This rule involves the counting of sensor inputs, and for this purpose at some level of description the robot must store the number of subsequent l symbols in memory. To store the number of subsequent l inputs, one may for example store n symbols l for n inputs from the left sensor. Alternatively, binary states could be

136 

Erich Rast

used. The number 3 is then represented by setting two states to on (see Figure 3). No matter which representation for n inputs of a kind is chosen, though, in order to run the program successfully it must be able to produce motor output for n time units. Taking into account the complete device, including the software running on it, we may thus say that the desired representation of n sensor inputs encodes the number n. Most symbols do not encode what they stand for. Consider for example the Arabic digit 3. In order to be able to represent three things this symbol must be mapped to three signals, objects, or other entities of a suitable kind. The symbol 3 does not itself exemplify what it stands for and it may thus be called a purely representational symbol, or, in the terminology of Langer (1951), a discursive symbol. From these symbols Langer distinguishes presentational symbols such as the three dots in Figure 3. Presentational symbols exemplify what they stand for in addition to representing it. In other words, they encode directly at least certain aspects of what they represent. Notice that the two binary gates are a tricky case. They seem to be more presentational than the digit 3, yet without a rule that maps the on-state of one of them to two objects or signals and the on-state of the other one to one signal or object, the gates cannot for themselves present three objects directly. If the mapping mechanism is included in the description of the symbol, then it may be considered presentational but otherwise it is purely representational just like the symbol 3. I do not claim that this distinction can be generalized easily to cases that are more complicated than the encoding of natural numbers. A simple isomorphism will most likely not suffice as an explanation of more complex presentational symbols and iconicism is one of the most vexing problems of semiotics. Instead of grounding the desired distinction on the notion of an isomorphism or making similar attempts to give a precise account that would require much more support, I will only assume some less stringent and more tentative definitions in what follows. Let us speak of a purely representational sign if the connection between sign and the signified is conventional and (in this sense) arbitrary. Let us, following H.N. Castañeda to some extent, assume in contrast to this that a presentation encodes finitely many aspects of what it represents in a way that makes the connection between the sign and the signified not just conventional and arbitrary. (The reason for allowing such a vague

De Se Attitudes and Semiotic Aspects of Cognition

137

definition here is that an explanation of the exact meaning of the phrase ‘not just conventional and arbitrary’ should be left to neurophysiology if physicalism is true and psychology if physicalism is false. That at least some such presentations exist has been illustrated by the example of finite presentations of natural numbers.) Given this distinction, it is clear that languages are purely representational, for it is well-known that even allegedly iconic expressions like onomatopoeia in natural languages are conventionalized. Cognition, on the other hand, cannot be purely representational. If cognition worked like a language and were also purely representational, then each thought token would stand for something else. As I briefly laid out in Rast (2007, pp. 270–275), this view would lead to an infinite regress, since an agent would then have to check to which entity a particular thought token refers in order to ‘understand’ the token, and so on for the symbols the token refers to, for the symbols the referent of the token refers to, etc., until a presentational symbol is encountered. This modern Homunculus problem shows that episodal thinking cannot in general be only representational; it must at some point involve presentations of objects, experiences, and so on. Moreover, as I’ve mentioned before and as H.N. Castañeda has emphasized throughout his work (see for example Castañeda, 1989, 1990), humans are finite and therefore presentations of objects or experiences in episodal thinking must be finite, too. Consequently, when an agent thinks a thought that may be circumscribed in public language as I am F for some property F, the I-presentation of herself in her cognition encodes finitely many aspects of herself and she attributes the property F to this presentation. Likewise, a presentation of a particular red experience in cognition presents finitely many aspects of a particular red object or a region of an agent’s visual field in her cognition. If physicalism is true, these presentations are encoded and processed directly by an agent’s brain – an open, analogue, massively parallel, and presumably non-deterministic computational system – and the result of such a computation is itself finite. But even if physicalism is not true, there does not seem to be any alternative to the popular view that episodal thinking is finite. Although not the central tenet of this article and somehow independent of the previous considerations, I would briefly like to discuss the popular yet sometimes misunderstood thesis that episodal thinking is also computational in nature, which further closes the gap between

138 

Erich Rast

the robot and the Mary example. By ‘computational’ I mean that the mind or brain can within the limits of its finite resources compute only functions that are representable by terms of the λ- or π-calculus and their reduction to canonical form, or another suitable sequential or parallel Turing complete formalism. While there are alternatives to this view, they are hard to reconcile with what we know about the physical world. First of all it must be noticed that the brain or mind can compute certain computable functions within its physical limits. It is therefore not a good idea to claim that the brain or mind is not at all a computational device. So to reject the computational model one has to claim that the mind/brain is higher than a Turing machine in the hierarchy of computational systems. Hypothetical devices that can solve problems that a Turing machine cannot solve are called hypercomputers, and there is a whole hierarchy of them. A relatively weak hypercomputer would for example be the ‘accelerated Turing machine’, which is based on ideas already considered by Russell. An accelerated Turing machine completes its first step during computation (such as moving the head on the tape, printing a symbol to tape, etc.) at time 1, the second step at 1+1/2 time units, the third one at time 1+1/2+1/4, and so on. When it reaches the limit at time 2 it might have solved a task that provably no ordinary Turing machine can solve. There are more credible descriptions of hypercomputers, but none of them seems to be fully compatible with current physical knowledge (see Lokhorst, 2000). Other alternatives such as Penrose (1989, 1997) are more elaborate, but don’t offer any convincing explanation of how the mind might work. Yet other alternatives such as claiming that the mind is neither computational nor hypercomputational nor based on quantum-physical phenomena amount to sheer mysticism. Now the fact that the computational model seems to be the best explanation of how episodal thinking works does not mean that it is the right explanation. The issue is clearly undecided. Still, the best explanation is the best explanation unless someone comes with a better one, and we should prefer the computational model unless someone comes up with direct counter-evidence to it. Notice that even if physicalism were false a computational model of the mind would still be more attractive than any of the alternatives mentioned; in that case, however, some mysticism could not be avoided due to the problem of explaining mind-matter interaction.

De Se Attitudes and Semiotic Aspects of Cognition

139

If the computational model is basically the right picture of how episodal thinking works, then invariably certain episodal thoughts fulfil a different role than others within the computational system as a whole. Although some episodal thoughts may in principle fulfil the same role as others if they are ultimately linked to the same sensor inputs and motor outputs, as the a versus lll example has illustrated, it is not very likely that a particular this red presentation or a particular I am F thought in human cognition could be replaced by some radically different thoughts, such as those that represent some realization of dispositional knowledge in cognition, and still fulfil the same role within the given computational system. What conclusions can be drawn from these general considerations concerning de se puzzles? Let us return to Perry’s supermarket example mentioned at the beginning of this article. In the thought experiment certain kinds of thoughts are connected to a certain kind of behaviour. For example, John Perry looks at himself in the mirror without recognizing himself, thinks what may be circumscribed in public language as This guy is making a mess and attempts to follow the person in order to tell him that his sugar package is damaged. Suddenly he thinks I am making a mess, which must here be understood as a rough and deficient natural language circumscription of what actually goes on in his brain (or mind, if physicalism is false) when he has the insight and correspondingly starts to clean up the sugar. According to the view I have suggested, this is so because a particular finite presentation in John Perry’s cognition lead to a particular behaviour, whereas another presentation leads to another kind of behaviour. The connection of this phenomenon to natural language is loose, though. Take any double-vision puzzle like Quine’s Ortcutt example in Quine (1956) Perry’s ship example in Perry (1979, p. 483), or Richard’s phone booth example in Richard (1983). Being forced to base their judgments on finite presentations of objects, people use certain expressions for some presentations and others for other presentations without realizing that these presentations present the same object in different ways to them. Hyperfine-grained belief can be used to model such cases, but outside of attitude ascriptions the respective natural language expressions retain their public language reference and refer to objects with infinitely many properties. As countless discussions about propositional attitudes have shown, even when the

140 

Erich Rast

respective natural language expressions occur inside attitude ascriptions it is controversial whether hyperfine-grained attitudes should be used to truth-conditionally encode an agent’s tendency to use certain expressions as opposed to others for certain presentations in cognition. Language is purely representational and public, whereas cognition is predominantly presentational, private, and its intricacies are mostly unknown at the time of this writing. For that reason, mappings between these symbolic systems remain inaccurate and leave room for variations due to different theoretical goals and modelling purposes. Things look different, however, when one is interested in logical representations of epistemic states of thinking or computing agents independently of natural language semantics. Certain thought tokens form equivalence classes on the basis of the role they play within the computational system as a whole, including the program currently running on it, and tokens from one equivalence class can by definition not be substituted by tokens from another one. Thus, when instances of certain classes of tokens are stored for later retrieval they play an essential role in the constitution of introspective knowledge of a certain sort that can trivially not be replaced by knowledge based on instances of another class of thought tokens within the computational system. De se attitudes and the special status of subjective experiences within an agent’s thinking as opposed to 3rd-person knowledge about them are symptoms of this general property of computational systems. So from an epistemic point of view hyperfine-grained attitudes seem to be unavoidable. However, being closely tied to purely representational formal languages, sometimes even to their syntax, existing accounts of hyperfine-grained attitudes lack a certain explanatory adequacy concerning these phenomena even when they are descriptively adequate. An explanatory satisfying account of hyperfine-grained attitudes for humans would have to be based on a comprehensive theory of presentational, episodal thinking in general, and no such theory is available at the time of the writing of this paper. If what has been said above is correct, according to such a theory episodal thinking is radically different from what we are used to calling a language. Finally, some things have to be said about the talk about ‘necessary conditions’ in the previous sections. I have not laid out any principal reasons why thought tokens that represent physical knowledge about

De Se Attitudes and Semiotic Aspects of Cognition

141

the world could not substitute thought tokens for subjective experiences or tokens that present thoughts like I am F for a property F in the sense that they might, under certain circumstances, fulfil the same role in episodal thinking as the thoughts they are considered to replace. There is no principal reason why within a computational system one sort of signals could not play radically different roles dependent on the overall state of the system, and I have only claimed that assuming such a dual role for certain thoughts in human thinking is an implausible attempt to explain away introspective knowledge. My reply to Dennett and the denial reply to the Two Gods puzzle is not that it is infeasible that certain theoretical knowledge could necessarily not play the desired role in thinking; my reply is rather that it actually doesn’t. How then, as one might ask, does this support the much stronger thesis that certain thoughts are a necessary condition for certain actions? The answer is that this depends on the way thoughts are categorized. The way typical de se puzzles are set up no thought other than I am F plays the same role for subsequent I-behaviour, even though another thought such as John Perry is F could in principle result in the same behaviour. I-thoughts lead to I-behaviour by definition, and, as one might continue, John Perry-thoughts lead to John Perry behaviour, this guy-thoughts lead to this guy-behaviour, and so on. If someone would claim that in the supermarket example a John Perry-thought might lead to the I-behaviour, a defendant of de se attitudes could reply that this is not possible, because thoughts that result in the I-behaviour, i.e. in the agent having the insight about themselves respectively described by the scenarios, are I-thoughts. They fulfil a role in episodal thinking and memory thereof which manifests itself as the I-behaviour. Understood in this way, having an I-thought is a necessary condition for the occurrence of an I-behaviour, because thoughts have been classified in this way. This is the trivial aspect of the suggested interpretation of de se puzzles, but as I have shown above they also exemplify the non-trivial formation of introspective knowledge. Using the same trivial ‘cognitive a priori’ definition of thoughts in the Mary example would render it rather incomprehensible, because in this way of talking we would have to say that if Dennett was right then thoughts that represent an omniscient scientist’s knowledge about colour vision would just be thoughts that represent a certain colour experience. Not even Dennett talks this way. We usually talk in another way about

142 

Erich Rast

cognition, namely in the way that thoughts are not individuated by their role; so in this way of talking two different thoughts might (at least in theory) play the same role, yet remain different from each other.

Summary and Conclusions Regarding the Mary example, I have outlined an approach that Dennett (2005) calls ‘thick materialism’. Mary learns something new and this new, introspective knowledge is compatible with physicalism. The particular insight of an agent about himself that is described in typical de se examples and an agent’s memory of this insight likewise result in introspective knowledge. In both cases the agent stores a certain thought token in memory and by having had the particular thought and being able to remember having had it in some way the agent acquires introspective knowledge. Within a computational system certain signals or thought tokens play a certain role in the system as a whole that other thoughts might not play. If t is the symbolic presentation of a certain sensory input within a given computational system, then a presentation t’ of knowledge about t can only play the same role within the computational system as t if it is suitably linked to other parts of the system in the same way as t. For example, when t consists of three symbols or signals lll that represent three inputs from a certain sensor and within the system at that time there are also some actual presentations t’ of the system’s dispositional knowledge about lll symbols, their storing and retrieval, and the various roles they can play in relevant subroutines of the computational system then under usual circumstances t’ cannot substitute t in the sense that t’ cannot play the same role as t within the computational system as a whole for the roles that are fine-grained enough to be of interest for a role-based explanation of the system. Concerning human thinking it seems even less likely that thoughts presenting sensory experiences could be replaced by thoughts that encode dispositional knowledge while retaining the same role within the system as a whole. If phenomenal concepts are taken as a condition for the ability of an agent to form introspective knowledge of the kind laid out in this

De Se Attitudes and Semiotic Aspects of Cognition

143

article, then it follows from what has been said above that these can be understood in purely physicalist terms without resorting to ontologically or metaphysically irreducible phenomenal properties. However, there does not seem to be any need for any special conditions of this sort, since the fact that humans have thoughts of different kinds in combination with the fact that they have the ability to remember them (to some extent) suffices as a general explanation as to why humans have the ability to form introspective knowledge.

Figures

Figure 1: An agent’s belief set is shrinking when he accepts new propositions.

Figure 2: Using centered propositions for representing de se belief.

144 

Erich Rast

Figure 3: Different ways of representing the number 3 and their respective decoding: (a) Arabic digit, (b) three dots, (c) two active binary states.

References Bealer, G. & Mönnich, U. (1989). Property Theory. In Dov Gabbay (ed.), Handbook of Philosophical Logic. Dordrecht: Kluwer, 133–251. Castañeda, H.-N. (1967). Indicators and Quasi-Indicators. American Philosophical Quarterly, 4, 85–100. Castañeda, H.-N. (1989). Thinking, Language, Experience. Minneapolis, Minn.: University of Minnesota Press. Castañeda, H.-N. (1990). Indexicality: The Transparent Subjective Mechanism for Encountering A World. Noûs, 24(5), 735–749. Churchland, P. (1985). Reduction, Qualia and the Direct Introspection of Brain States. Journal of Philosophy, 82, 8–28. Churchland, P. (1989). A Neurocomputational Perspective: The Nature of Mind and the Structure of Science. Cambridge, MA: MIT Press. Coleman, S. (2009). Why the Ability Hypothesis is Best Forgotten. Journal of Consciousness Studies, 16(2–3), 74–97. Conee, E. (1994). Phenomenal Knowledge. Australasian Journal of Philosophy, 72, 136–50. Cresswell, M. J. (1985). Structured Meanings. Cambridge, MA: MIT Press.

De Se Attitudes and Semiotic Aspects of Cognition

145

Dennett, D. C. (1991). Epiphenomenal Qualia? In Consciousness Explained. Boston: Little, Brown and Company. Dennett, D. C. (2005). Sweet Dreams. Cambridge, MA: MIT Press. Hintikka, J. (1975). Impossible possible worlds vindicated. Journal of Philosophical Logic, 4(3), 475–484. Jackson, F. (1982). Epiphenomenal Qualia. The Philosophical Quarterly, 32(127), 127–136. Jackson, F. (1986). What Mary Didn’t Know. The Journal of Philosophy, 83(5), 291–295. Langer, S. (1951). Philosophy in a New Key. Cambridge, MA: Harvard University Press. Lewis, D. K. (1979). Attitudes De Dicto and De Se. Philosophical Review, 88(4), 513–543. Lewis, D. K. (1983). Postscript to ‘Mad Pain and Martian Pain. In David Lewis. Philosophical Papers Vol. I. New York: Oxford University Press, 130–32. Lewis, D. K. (1988). What Experience Teache. In Proceedings of the Russellian Society. Sidney: University of Sidney. Lewis, D. K. (1990). What Experience Teaches. In William Lycan (ed.), Mind and Cognition. Oxford: Blackwell, 499–518. (Reprint) Lokhorst, G.-J. C. (2000). ‘Why I am Not a Super Turing Machine’. Manuscript of a talk given at University College London on 24. May 2000. Retrieved in December 2010 from . Nemirow, L. (1980). Review of ‘Mortal Questions’ by Thomas Nagel. Philosophical Review, 89, 473–77. Nemirow, L. (1990). Physicalism and the Cognitive Role of Acquaintance. In William Lycan (ed.). Mind and Cognition. Oxford: Blackwell, 490–99. Nida-Rümelin, Martine (2009). ‘Qualia: The Knowledge Argument’. Published online in Edward Zalta (ed.): Stanford Encyclopedia of Philosophy. Retrieved on 12.2.2010 from . Penrose, R. (1989). The Emperor’s New Mind. New York: Oxford University Press. Penrose, R. (1994). Shadows of the Mind. New York: Oxford University Press.

146 

Erich Rast

Perry, J. (1977). Frege on Demonstratives. Philosophical Review, 86, 474–497. Perry, J. (1979). The Problem of the Essential Indexical. Noûs, 13, 3–21. Quine, W. V. O. (1956). Quantifiers and Propositional Attitudes. The Journal of Philosophy LIII (5: March), 177–187. Rast, E. H. (2007). Reference and Indexicality. Berlin: Logos. Richard, M. (1983). Direct Reference and Ascriptions of Belief. Journal of Philosophical Logic (12), 425–452. Stalnaker, R. (2004). ‘Knowing Where We Are, And What It Is Like’. Manuscript of a talk given at NYU’s La Pietra Conference on Consciousness, Florence 2004, version of 2006 retrieved from . von Stechow, A. (1982). Structured Propositions. Technical report of the SFB 99. Konstanz: Universität Konstanz. Thomason, R. (1980). A Model Theory for Propositional Attitudes. Linguistics and Philosophy 4, 47–70. Tye, M. (2009). Consciousness Revisited. Cambridge, MA: MIT Press.

The Division of the Mind: Paradoxes and Puzzles Vasco Correia

1.  Beyond divisionism The hypothesis that the mind is divided is often put forward as a means to make sense of irrationality, whether in the cognitive sphere (delusional beliefs) or in the practical sphere (weak-willed actions). In essence, the “divisionist” argument states that we cannot understand irrational actions and beliefs without assuming that the mind is composed of different sub-systems. Donald Davidson (1985b, p. 353), one of the most prominent proponents of divisionism, explicitly endorses this methodological assumption: I have urged in several papers that it is only by postulating a kind of compartmentalization of the mind that we can understand, and begin to explain, irrationality.

In a similar vein, Aristotle (2009, p. 20) introduced the distinction between the “rational” and the “irrational” part of the soul to be able to account for the phenomenon of akratic action (or “lack of self-control”), in which the agent deliberately acts against his or her best judgment. And it was also in an attempt to solve the problem of irrationality that Plato argued in The Republic that there are three components of the soul (nous, thumos and epithumia). Irrational actions, he argues, stem from a conflict between those components. Plato (1992, p. 115) evokes the case of Leontios, a man who was unable to resist the temptation of staring at a bunch of dead bodies, despite deeming such voyeurism morally wrong. What seems puzzling about this sort of behaviour is that the agent is aware that it would be in his best interest to perform the course of action A, all things considered, but chooses to do B nonetheless.

148 

Vasco Correia

Philosophers have always questioned the very possibility of such attitudes. Can one see the best option, and approve of it, and nevertheless, in full awareness, choose the worst option? How is that possible? Although a number of philosophers deny the possibility of akrasia thus described, the proponents of the divisionist hypothesis argue that it is indeed possible to act contrary to one’s best judgment, insofar as there are different instances operating within the mind. In the case of Leontios, for example, Plato argues that a part of him wanted to stare at the gruesome scene, while another part of him repudiated such an attitude1. Many authors make a similar claim with regard to specific cases of cognitive irrationality, such as self-deception and wishful thinking, arguing that this type of phenomenon also requires some sort of differentiation in the mind2. In this view, there is a formal analogy between self-deception and akrasia, to the extent that both phenomena seem to imply some degree of inconsistency between the agent’s attitudes: the akratic agent is someone who believes that A is the better option and nonetheless decides to do B; and the self-deceiver, likewise, is someone who believes that p is the most likely hypothesis and nonetheless decides to believe that not-p. After all, if the self-deceiver did not initially hold the unwelcome belief that p, he or she would have no reason to make the effort of embracing the opposite belief. For example, if John did not know “deep inside him” that he is an alcoholic, he wouldn’t go to great pains to persuade himself that he doesn’t have a problem with alcohol. To that extent, as Davidson suggests, it seems reasonable to suggest that the initial belief that p, in concert with the desire that not-p, causally sustains the contradictory belief that not-p: “Self-deception is notoriously troublesome, since in some of its manifestations it seems to require us not only to say that someone believes both a certain proposition and its negation, but also to hold that the one belief sustains

1

2

Hence Plato’s metaphor in the Phaedrus (2003, p. 28) depicting the psyche as a “composite pair of winged horses […] one of them noble and of noble breed and the other ignoble and of ignoble breed”, constantly torn between the impulses of desire and the recommendations of reason. In particular, Donald Davidson, Robert Audi, David Pears, Herbert Fingarette and Sebastian Gardner.

The Division of the Mind: Paradoxes and Puzzles

149

the other”3. According to the so-called “intentionalist” account of self-deception, at any rate, the self-deceiver must hold two contradictory beliefs simultaneously: he or she must belief that p is true, while also believing that p is false. And this is precisely why the divisionist hypothesis is so appealing to many authors, since it solves the paradox of contradictory attitudes coexisting in the same mind. Even at a cognitive level, it doesn’t appear paradoxical to hold two contradictory beliefs simultaneously, so long as the beliefs in question belong to different sub-systems of the mind. In this sense, divisionists seem to avoid what Alfred Mele (1998, p. 38) calls the “doxastic paradox” of self-deception, i.e. the possibility that the same individual believes that p and that not-p simultaneously. According to Robert Audi (1982, p. 137), this is possible because contradictory beliefs coexist at different levels: the subject believes that p at a conscious level, to be sure, but he or she “unconsciously knows that not-p (or has reason to believe, and unconsciously and truly believes, not-p)”. To that extent, it seems possible not only that the subject holds two contradictory beliefs simultaneously, but also that he or she remains unaware of that contradiction (the two beliefs “failing to clash”, so to speak). There are nevertheless two crucial difficulties that undermine the divisionist postulate. The first was pointed out by Sartre in his famous analysis of the mauvaise foi (or self-deception) and concerns primarily Freudian-like accounts of the divided mind. Sartre’s argument is presented as a dilemma, known as the “paradox of repression”, and questions the idea that a part of the mind (supposedly unconscious) could somehow perform the complex task of preventing another part of the mind from contemplating harmful representations. The second difficulty concerns all kinds of divisionism, Freudian-like or not. It is commonly referred to as the “homunculus fallacy” and was initially stressed by Wittgenstein, who argued that the hypothesis that there are relatively autonomous sub-systems coexisting in the mind compromises the very notion of personal identity. 3

Davidson, “Deception and Division”. In Le Pore and McLaughlin (eds.), Actions and Events. Perspectives on the Philosophy of Donald Davidson, Oxford, Basil Blackwell, 1985, p. 138.

150 

Vasco Correia

In this paper I shall examine some of the most influential versions of the divisionist account, namely the models developed by Freud, Fingarette, Pears and Davidson. I argue (1) that each of these accounts leads to a specific paradox or set of paradoxes, and also (2) that the divisionist hypothesis is NOT necessary to explain ordinary cases of irrationality, such as self-deception, denial, rationalization, wishful thinking, and the like. This is not to say that it is impossible for our minds to suffer any sort of division (as indeed happens in pathological cases of mental dissociation), but simply that we do not need to presuppose such divisions to be able to account for ordinary cases of irrationality. My view is that most cases of irrationality, whether practical or cognitive, stem from the influence that desires and other emotions exert upon our cognitive processes, and thereby upon our judgments. If this hypothesis is correct, irrational thought and irrational behaviour are typically caused by a conflict between individual mental states (for example, a conflict between a desire and a belief), rather than from a conflict between differentiated parts of the mind.

2.  The paradoxes of the Freudian account We may begin by examining Freud’s influential model of divisionism, since it admittedly inspired Fingarette’s, Pears’ and Davidson’s more recent versions of this approach. As Freud (1923, p. 19) often stresses, the cornerstone of Psychoanalysis is the claim that a significant number of our representations remain unconscious or repressed, forming what he initially calls the Unconscious (das Unbewusste) and later the Id: “the division of mental life into what is conscious and what is unconscious is the fundamental premise on which psycho-analysis is based”. However, he also specifies that unconscious contents remain excluded from the consciousness, not because our consciousness accidentally neglects to contemplate them, but because they cannot become conscious, insofar as a given strength represses them: “Analysis of these examples of forgetting reveals that the motive of forgetting is always an unwillingness to recall something which may evoke painful feelings” (Freud, 1901,

The Division of the Mind: Paradoxes and Puzzles

151

p. 332). In fact, the process of repression needs to be understood in the light of the fundamental principles that rule the psychic apparatus, and particularly the so-called “pleasure principle”, which implies that the psyche spontaneously seeks pleasure and avoids pain (caeteris paribus). Thus, a representation is repressed either because it is deemed too painful to be made conscious or because it constitutes a threat, directly or indirectly, to the equilibrium of the psychic apparatus as a whole. In L’être et le néant Sartre famously argues that the Freudian account is intrinsically paradoxical. The kernel of Sartre’s argument is that the so-called Unconscious would have to behave as a sort of “second consciousness” to be able to repress harmful thoughts (Sartre, 1943, p. 89). After all, Sartre observes, in order to withhold dangerous information from the consciousness, the Unconscious would have to perform the highly complex task of assessing the potential impact of each given representation in the economy of the psyche. Yet such a task could not be achieved successfully without the unconscious being aware of the representations in question, for otherwise it would lack the means to decide which information is suitable to be made conscious, and which information needs to be repressed. But if that were indeed the case, the unconscious would have to know the negative representations in order not to know them, which seems paradoxical4. The second problem with Freud’s account is that it undermines the concept of personal identity, since it attributes propositional attitudes such as beliefs, desires, representations, memories and emotions to mere parts of the mind, despite the fact that these attitudes only seem appropriate to describe the person as a whole (the in-dividual, strictly speaking). As Wittgenstein (1953, p. 281) points out, “only of a human being and what resembles (behaves like) a living human being can one say: it has sensations; it sees, is blind; hears, is deaf; is conscious or unconscious”. Freud’s description of the mind’s components often falls prey to what Anthony Kenny calls the “homunculus fallacy”, in other words the tendency to describe the alleged parts of the mind as though 4

Simon Boag (2007, pp. 421–447) has argued that the resolution of this paradox hinges upon the recognition that the process of repression inhibits knowledge of knowing that which is repressed. The problem with this solution, however, is that a third degree knowledge would then be required to determine which of the knowledge is to be inhibited, and which is not.

152 

Vasco Correia

they were different persons within the person. In The Ego and the Id, in particular, Freud (1923, p. 146) develops an anthropomorphic model of the psyche which depicts the ego, the super-ego and the id as different “homunculi” coexisting and competing in accordance with their specific set of demands. Regarding the ego, for example, he writes that “we can see [it] as a poor little creature subjected to servitude in three different ways, and threatened in consequence by three different dangers – one posed by the external world, one by the libido of the id, and one by the harshness of the super-ego”. Likewise, in An Outline of Psychoanalysis, Freud (1940, p. 173) goes as far as to suggest that the psychoanalyst ought to make an alliance with one of the mind’s components, namely the ego, in an attempt to protect it from the demands set by its two rivals: “The analytic physician and the patient’s weakened ego, basing themselves on the real external world, have to band themselves together into a party against the enemies, the instinctual demands of the id and the conscientious demands of the super-ego. We form a pact with each other”. But this raises of course the problem of personal identity, given that the components of the psyche do not seem to have much in common. Under such a description, in effect, one may wonder what sort of principle would be able to ensure the overall unity of the several sub-systems. Sartre (1943, p. 89) also stresses this point: “By rejecting the conscious unity of the psyche, Freud cannot but presuppose everywhere a magic unity linking distant phenomena across obstacles”. Irvin Thalberg (1984, pp. 253–254) makes a similar objection by raising the “who’s who?” question regarding the alleged constituents of the psychic apparatus: “Is our ego awake or asleep when we sleep?”; “Which ‘self’ does my ego have the duty of protecting?”; “Whom does my super-ego watch when it engages in ‘self-observation’ – me, my ego, itself?”; “Whose enjoyment do [my instincts] ‘strive’ for?”; and “Whose interests is the ego trying to protect in its repressions?”. Far from solving this serious difficulty, Freud only seems to aggravate it in his later writings, bringing forward the hypothesis of a so-called “splitting of the Ego” (Ichspaltung). In the posthumous text “The splitting of the Ego in the process of defence”, remarkably, he claims that the ego itself is liable to suffer an inner division. When the conflict between the ego’s desires and what reality demands becomes unbearable, Freud (1938, p. 276) writes, “The two contrary reactions to the conflict persist

The Division of the Mind: Paradoxes and Puzzles

153

as the centre-point of a splitting of the ego […] The synthetic function of the ego, though it is of such extraordinary importance, is subject to particular conditions and is liable to a whole number of disturbances”. Although Freud initially confines this hypothesis to the specific cases of fetishism and psychosis, he later extends it to neurosis in general5. But, of course, if the components of the mind may themselves be sub-divided into sub-constituents, the paradoxes of divisionism become all the more difficult to surmount.

3.  Fingarette: “ego” and “counter-ego” In spite of these difficulties, Herbert Fingarette’s analysis takes up Freud’s idea of a splitting of the ego. However, he gives it an ingenious twist by suggesting that the ego’s inner conflicts lead to the formation of what he calls a “counter-ego”. According to Fingarette (1984, p. 224), this instance is formed during the emergence of the person’s self and coexists with the ego in the mind throughout life: “The defensive outcome, then, is to establish what we may call a counter-ego nucleus, this nucleus being the structural aspect of counter-cathexis”. Aware of the dilemmas that Freud’s theory seems to face, the author explains that “such paradoxes arise because both earlier and later versions [of Freud’s theory] are parallel in insisting correctly on the fact that there is a split in the psyche, but fail in defining the nature of the split adequately” (Fingarette, 1984, p. 223). The adequate way to define the nature of the split, in his view, would be to suggest that every ego is essentially divided (and not just the neurotic or the psychotic ego), and also that the upshot of such a division is the formation of two basic sub-systems, the ego and the counter-ego. In addition, Fingarette contends that the origin of such a split is not the process of repression, but a process of “disavowal” through which the counter-ego is split off from the ego. He illustrates this process by 5

Cf. Freud, Outline of Psycho-Analysis, The Standard Edition, Vintage, The Ho­ garth Press, London, vol. XXIII, p. 202.

154 

Vasco Correia

evoking the example of a man who resents his employer’s attitude toward him, but is unable to accept (or avow) his own resentment because he considers it immoral to feel such an anger toward someone else. As a result, he argues, the subject refuses to identify himself with the attitude in question and claims: “It is not ‘I’ who am angry; from henceforth I will dissociate myself from it; it is repugnant to me” (Fingarette, 1984, p. 218). The process of splitting off would begin at an early age, through the constraints of education and social interactions. Negative feelings such as shame and guilt, in particular, would play an important role in causing the person to “disavow” certain emotions and desires, progressively inducing the creation and development of a counter-ego nucleus in the mind. The advantage of Fingarette’s account is that it seems to avoid the paradox of repression, given that the phenomenon of disavowal, unlike Freud’s mechanism of repression, is a conscious and deliberate act performed by the person’s Ego: “the defensive process is a splitting of the ego which is not something that ‘happens’ to the ego but something the ego does, a motivated strategy” (Fingarette, 1984, p. 224). There is however a fundamental dilemma in Fingarette’s account: assuming that the counter-ego is not repressed (nor unconscious) but merely “splitoff” from the ego, it follows that the same consciousness must serve the purposes both of the ego and of the counter-ego. But how could the Ego and the counter-ego remain effectively dissociated if they share the exact same consciousness? Here lies the paradox: on the one hand, the subject’s consciousness is meant to deny the set of beliefs and emotions which seem unacceptable to the ego; but on the other hand, as part of the counter-ego, the consciousness must be aware of such beliefs and emotions. In the previous case, for example, the subject has to be aware of his anger toward his boss, for otherwise he wouldn’t bother to deny it. But if the subject’s ego knows about the unacceptable emotions, how can it not recognize them as its own? And what sense does it make to say that the subject’s consciousness acknowledges certain attitudes as a counter-ego, but not as an ego? An additional problem with Fingarette’s account lies in the assumption that people are capable of disavowing inconvenient realities both consciously and voluntarily. This philosophical hypothesis, known as doxastic voluntarism, is in fact ruled out as a psychological impossibility by most philosophers and psychologists, who argue that no one

The Division of the Mind: Paradoxes and Puzzles

155

seems to be able to decide at will, hic et nunc, what to believe or not6. As Jonathan Bennett (1990, p. 96) points out, although it may not be conceptually impossible to cause oneself to believe something at will, it seems impossible de facto to achieve such a task: There could be simpler, quicker, more reliable means for causing beliefs in people without giving them evidence. I passionately want to spend the evening in a state of confidence that the weather will be fine tomorrow (I have my practical reasons), so I give myself the thought of tomorrow’s weather being fine while snapping my fingers in a certain way, and sure enough I end up convinced that the weather will be fine tomorrow. We have no such fast, reliable techniques for producing belief without evidence, but they are not conceptually ruled out.

Finally, one could also question the very existence of a so-called “counterego”, much like Freud’s opponents have questioned the existence of an unconscious. Even assuming that the process of denial may occur intentionally, it seems doubtful that it should result in the formation of a counter-ego which most of us are unaware of.

4.  Pears: the rival “centres of agency” In a sense, one could describe David Pears’ version of divisionism as a compromise between Freud’s founder model and Fingarette’s sophisticated account. On the one hand, Pears agrees with Fingarette in saying that the divided mind is essentially composed of two sub-systems, instead of the three Freud describes, which he generically refers to as the “main” and the “secondary” sub-system. But on the other hand, Pears (1982, p. 269) adopts the Freudian assumption that one of those sub-systems must remain unconscious: There is then one centre of activity in the subject’s contemporary consciousness and another in the reservoir […] We have to suppose that each of these two centres includes any information needed to give its desire a line of action […] This 6

See for example Pascal Engel, “Volitionism and Voluntarism About Belief ”. In A. Meijers (ed.), Belief, Cognition and the Will, Tilburg University Press, 1999.

156 

Vasco Correia

extremely economic hypothesis is based on Freud’s fundamental concept of a boundary dividing the conscious from the preconscious and the unconscious.

Pears (1982, p. 268) suggests that this is “the most economical form of the hypothesis of the divided mind”, not only because it refrains from engaging in an anthropomorphic description of the mind’s sub-systems, but also because it does not postulate that such a division is an essential feature characterizing each and every mind. According to Pears, we only need to assume that the mind suffers this kind of differentiation in the case of subjects who are prone to some degree of irrationality, whether cognitive or practical. In principle, the cause of the division is the emergence of a strong desire which turns out to be incompatible with the subject’s predominant preferences. For example, the desire to have an extra-marital affair with an attractive person is of course incompatible with the aspiration to remain a faithful partner and a moral person7. If strong enough, the desire in question may eventually form what Pears calls a “rival centre of agency”, which is able to compete with the main sub-system and liable to induce irrational behaviour and irrational thinking. The interest of Pears’ account is that it avoids the homunculus fallacy, given that the secondary sub-system of the mind is described as a purely functional entity. To begin with, it is not an essential feature of the mind, but a characteristic which only applies to irrational minds; secondly, it is not defined idiosyncratically, given that any type of desires could in principle characterize the rebellious sub-system; and finally, for that very reason, it may actually differ from one person to the other (unlike Freudian sub-systems, which are supposedly universal). Nevertheless, Pears’ model seems unable to avoid the paradox of repression. After all, how could the secondary sub-system, which is unconscious, be able to assess the impact of each mental state on the psychic apparatus? As a matter of fact, it seems difficult to conceive that an unconscious instance should possess the awareness and the discernment required to predict the positive or negative impact of every belief, every desire and every representation upon the subject’s mental equilibrium. 7

It is worth noting, however, that the rebellious desire could in principle be of any type: not just a sexual drive, as in many of Freud’s analysis, but also professional ambition, desire for fame and glory, desire to get rich, and so forth, depending on the subject’s specific set of preferences and values.

The Division of the Mind: Paradoxes and Puzzles

157

Aware of this difficulty, Pears (1982, p. 275) argues that the act through which the mind excludes a given content from consciousness is not only unconscious but also “self-reflective”: “The only possible reply is that the suppression of the consciousness that normally accompanies a particular type of mental event may itself be unconscious without having succumbed to a previous act of suppression. To put the point in another way, suppression can be self-reflective”. Ingenious as it may sound, Pears’ solution seems insufficient to neutralize Sartre’s criticism, for even if we accept the idea that there are self-reflective acts of repression, the question of how exactly such mental acts are able to perform the elaborated task of separating the wheat from the chaff, so to speak, sorting out what might be beneficial and what might be harmful for the mind, remains to be answered. As Mark Johnston (1989, p. 82) rightly points out, this would imply that an unconscious sub-system of the mind should be able to manipulate the main system in the best interest of the mind as a whole: The question arises how the protective system could do all this without being conscious of (introspecting) its own operations. After all, it has to compare the outcome it is producing with the outcome it aimed for and act or cease to act accordingly. Any consciousness by the protective system of its own operation is “buried alive”, i.e., is not acceptable to the consciousness of the main system.

An additional problem is that Pears’ definition of the secondary sub-system remains somewhat vague. He writes that, for the mind to be divided, “there must be two conflicting desires which set up rival centres of activity” (Pears, 1982, p. 268), but neglects to specify which other conditions must be combined for the division to occur. In particular, the reader is left wondering whether there are as many secondary sub-systems of the mind as there are strong conflicts between desires. In the book Irrationality Alfred Mele (1987, p. 143) stresses this point with an amusing analogy: “Explaining doxastic irrationality in Pears’ fashion is rather like explaining how a football team held another scoreless by saying that the former strategically rendered all of the latter’s scoring attempts ineffective. The football fan wants much more than this. He wants to know how team A rendered team B’s attempts ineffective”.

158 

Vasco Correia

5. Davidson: the mind’s “compartments” Although Donald Davidson extensively acknowledges the extent to which his account of irrationality is inspired by Freud’s analysis, the divisionist hypothesis he brings forward is in fact weaker than Freud’s, and arguably even more minimal than Pears’. To begin with, Davidson doesn’t assume that there is such a thing as an unconscious part of the mind allegedly inaccessible to the subject’s consciousness. He doesn’t go as far as to deny the existence of an Unconscious, but suggests that we do not need to presuppose the existence of contents inaccessible to the consciousness to be able to make sense of irrationality. All we need to presuppose, he argues, is that there are different “territories” or “compartments” in the mind which the consciousness cannot survey simultaneously. In addition, while Pears and Fingarette suggest that the secondary sub-system emerges in the mind as an organized centre of action, competing with the main system and aiming at distinct goals, Davidson (1982, p. 182) refuses to speculate about that aspect, thereby avoiding the homunculus fallacy. All that is required, Davidson suggests, is that “within each [department of mind] there is a fair degree of consistency, and where one element can operate on another in the modality of non-rational causality”. Thus, if an agent holds contradictory beliefs, as in the case of self-deception, one should assume that there is a subsystem of mental states consistent with the belief that p, and another sub-system of mental states consistent with the belief that not-p. More specifically, Davidson suggests that the partition of the mind occurs whenever a non-rational causality arises, i.e. whenever a mental state A causes a mental state B without there being a (good) reason for B. For example, Jack’s belief that Jane is in love with him is irrational in the case where this is caused by his feelings for Jane (rather than objective information), inasmuch as his desire to be loved by Jane is surely not a good reason to believe that Jane is in love with him. If Jack’s irrational belief stems from a genuine process of self-deception, and not from sheer wishful thinking, this would mean that a compartment of his mind believes that Jane is in love with him, while another compartment of his mind believes the opposite. And in a sense, Davidson explains,

The Division of the Mind: Paradoxes and Puzzles

159

it’s precisely because Jack believes that Jane probably doesn’t love him (and because he cannot bear that reality) that he is motivated to embrace the belief that she does – the former belief causally inducing the later. Despite the advantages of Davidson’s account relative to other divisionist models, it seems to encounter a new set of problems. The first difficulty is that it undermines Davidson’s own holistic theory of mind, which contends that an individual mental state can only be understood in the light of the whole set of attitudes constitutive of the subject’s mind, and that there must be a large degree of consistency between those attitudes. Davidson (1982, p. 184) himself highlights the difficulty: There is no question but that the precept of unavoidable charity in interpretation is opposed to the partitioning of the mind. For the point of partitioning was to allow inconsistent or conflicting beliefs and desires and feelings to exist in the same mind, while the basic methodology of all interpretation tells us that inconsistency breeds unintelligibility.

A second insufficiency lies in the fact that Davidson’s description of the mind’s sub-systems remains too vague. As Pears (1991, p. 395) points out, “[t]he drawing of a fault-line through a point at which internal irrationality occurs is only the beginning of the theory, and it is not enough to identify a sub-system”. As a matter of fact, Davidson states that whenever irrational causality occurs in the mind one must assume that a correlative partition is consumed; but he doesn’t specify how that partition comes about, nor the terms in which the sub-systems co-exist and interact. Besides, as Pascal Engel (1991, p. 16) rightly suggests, the vagueness of Davidson’s analysis extends to the very dynamic underlying the mind’s alleged differentiation: “[Davidson’s model] would need the equivalent of what Freud calls the dynamic aspect of the unconscious to account for the interactions between the sub-systems”. In other words, what Davidson’s model seems to lack is a descriptive account of irrationality, rather than a merely normative one. And, finally, many authors have questioned Davidson’s claim that ordinary cases of irrationality, such as weakness of will and self-deception, necessarily involve some sort of inconsistency between the agent’s mental attitudes (which, in turn, would justify the claim that such cases necessarily entail a partition of the mind). With regard to weakness of will, authors such as George Ainslie and Jon Elster

160 

Vasco Correia

convincingly argue that one needs not to conceive the phenomenon as the result of an action contrary to the agent’s own judgment, as if the action and the intention to act were strictly opposed. More plausibly, what seems to happen when an agent yields to temptation in spite of a previous decision is that he or she revises his or her initial judgment in a sudden and ill-considered fashion under the pressure of an urging desire. Thus, for example, my strong desire to eat a dessert is liable to affect my judgment and lead me insidiously to revise the decision to lose weight. Even if, all things considered, the option of losing weight might be the one which maximizes my well-being, the craving for a more immediate reward induces a biased perception of my preferences and causes me to believe, for a short and decisive instant, that it is preferable to eat the dessert. Ainslie (2001, p. 32) calls this cognitive effect “hyperbolic discounting bias”, which he defines as a sort of myopia regarding preferences over time: caeteris paribus, people tend to overrate the value of immediate rewards and to underrate the value of future rewards. According to this hypothesis, the agent who chooses to go on a diet and eventually fails to act accordingly does not act against his or her own judgment, strictly speaking, for at the time he or she eats the dessert his or her (biased) judgment assesses that option as being preferable. In this sense, there is no inner inconsistency in the agent’s mind (no simultaneously contradictory intentions to act), but simply an evolution in time of the agent’s preferences; in other words, a revised decision, however irrational and temporary it may turn out to be. The upshot of this analysis is that divisionism is not required as a presupposition to understand irrational action, given that a diachronic inconsistency of preferences does not raise the paradox of contradictory attitudes. Likewise, several philosophers and psychologists have argued that self-deception does not imply some sort of inner inconsistency in the subject’s mind either. Alfred Mele and Ariela Lazar, in particular, have argued that self-deception does not entail the subject holding two contradictory beliefs at the same time, but simply a hiatus between what the available evidence clearly suggests, on the one hand, and what the subject is led to believe, on the other hand. In their view, the self-deceived subject does not initially know the true proposition that p and at some point decides to believe the false proposition that not-p. Instead, what happens is that the subject’s judgment is biased by some desire (or other

The Division of the Mind: Paradoxes and Puzzles

161

emotion) which induces a distorted perception of reality. If Jack believes that Jane is in love with him in the teeth of evidence, for example, the problem is not that Jack intentionally decides to believe what seems more convenient, as Davidson suggests, but rather that his feelings for Jane surreptitiously distort his interpretation of her attitudes toward him. A simple sign of friendship may then appear to be an evidence of love. But, again, if there are no contradictory beliefs in the agent’s mind, if the irrational belief simply stems from a biased perception of reality, there is no compelling reason to assume that his or her mind is divided and that one part of the mind is aware of the truth while the other one is being deceived. Furthermore, as Mele (1987, p. 121) observes, it would be paradoxical to suggest that one part of the mind is able to employ a deceptive strategy upon the other, given that a “potential self-deceiver’s knowledge of his intention and strategy would seem typically to render them ineffective”.

6. Conclusion: toward a unitary solution In essence, our critical analysis brings about two significant conclusions. First, that each version of the divisionist account leads to a certain number of paradoxes. This is not to say that the divisionist view is intrinsically paradoxical, but simply that, to my knowledge, no coherent model of divisionism has been developed so far. Second, and perhaps more importantly, it would seem that it is not necessary to postulate the division of the mind to be able to account for ordinary cases of irrationality. As we have seen in the last section of this paper, most cases of cognitive and practical irrationality can be accounted for in unitary terms, without appealing to the divisonist postulate. Crucially, my claim is that both irrational behaviour and irrational thinking can be explained as the result of a conflict between individual mental states, rather than the result of a conflict between differentiated parts of the mind. We have seen this regarding cognitive cases of irrationality, such as self-deception, denial, rationalization, wishful thinking, and the like, which arguably stem from the influence our desires

162 

Vasco Correia

and emotions are liable to exert upon our judgments. And we have seen it regarding practical cases of irrationality, and particularly akrasia (or weakness of will), which can also be understood as the result of a biased judgment, given that the phenomenon of motivated irrationality is also liable to affect the evaluative judgments upon which we base our choices. Here too, it’s simply an individual mental state, generally a desire, which induces an irrational assessment of the feasible options and subsequently an irrational action. Even though the hypothesis of mental partitioning is not required to explain ordinary cases of irrationality, however, it may apply to extreme cases of irrationality, and particularly to pathological cases of “dissociative identity” (DSM IV) or “multiple personality disorder” (ICD 10). To that extent, it would be wrong to rule out the divisionist hypothesis as a paradox per se. Instead, it seems fair to suggest that divisionism remains a puzzle to be solved. It would be interesting, in particular, to examine what exactly makes the difference between “normal” and “pathological” cases of irrationality with regard to identity disorders.

References Ainslie, G. (2001). Break-Down of Will, Cambridge: Cambridge University Press. Aristotle, (2009). The Nicomachean Ethics, transl. D. Ross, New York: Oxford University Press. Audi, R. (1982). Self-Deception, Action and Will. Erkenntnis, 18, 133– 158. Bennet, J. (1990). Why is Believing Involuntary. Analysis, 50, 87–107. Bird, A. (1994). Rationality and the Structure of Self-Deception. European Review of Philosophy, 1, 19–37. Boag, S. (2007). Realism, Self-Deception and the Logical Paradox of Repression. Theory and Psychology, 17(3), 421–447. Bouveresse, J. (1991). Philosophie, mythologie et pseudo-science, Combas : L’éclat.

The Division of the Mind: Paradoxes and Puzzles

163

Davidson, D. (1982). Paradoxes of Irrationality. In R., Wollheim, & J., Hopkins (eds.), Philosophical Essays on Freud, Cambridge: Cambridge University Press. Davidson, D. (1985a). Deception and Division. In Actions and Events, E. Lepore and B. McLaughlin (eds.), New York: Basil Blackwell. Davidson, D. (1985b). Incoherence and Irrationality. Dialectica, vol. 39 (no. 4), 345–353. Davidson, D. (1998). Who is Fooled? In Dupuy (ed.), Self-Deception and Paradoxes of Rationality, Stanford: CSLI Publications, 1–19. Elster, J. (ed.), (1985). The Multiple Self, Cambridge: Cambridge University Press. Engel, P. (1991). Avant propos. In Davidson, Paradoxes de l’irrationalité, Éclat, Combas, 1–20. Engel, P. (1999). Volitionism and Voluntarism About Belief. In A. Meijers (ed.), Belief, Cognition and the Will, Tilburg: Tilburg University Press, 9–25. Fingarette, H. (1984). Self-deception and the splitting of the ego. In R., Wollheim, & J., Hopkins, (eds.), Philosophical Essays on Freud, Cambridge: Cambridge University Press, 212–226. Freud, S. (1920). Beyond the Pleasure Principle, trans. J. Reddick, London: Penguin classics, 2003. Freud, S. (1901). Psychopathology of Everyday Life, trans. A. Brill, T. Fisher, London: Unwin, 1914. Freud, S. (1923). The Ego and the Id, transl. J. Riviere, London, Hogarth Press and Institute of Psycho-analysis, 1927. Freud, S. (1938). The Splitting of the Ego in the Process of Defense, trans. James Strachey, The Standard Edition, vol. XXIII, London: Vintage, 1964, 271–279. Freud, S. (1940). An outline of Psycho-analysis, The Standard Edition, Vintage, London, 2001. Johnston, M., (1989). Self-Deception and the Nature of Mind. In Rorty and McLaughlin (eds.), Perspectives on Self-Deception, Berkeley, L.A., London: University of California Press, 63–91. Kenny, A., (1984). The Homunculus Fallacy. In The Legacy of Wittgenstein, Oxford: Blackwell, 125–136. Mele, A. (1987). Irrationality: An Essay on Akrasia, Self-Deception and Self-Control, N.Y., Oxford: Oxford University Press.

164 

Vasco Correia

Mele, A (1998). Two Paradoxes of Self-Deception. In J.P. Dupuy (ed.), Self-Deception and Paradoxes of Rationality, CSLI Publications, Stanford, California, 37–58. Pears, D. (1982). Motivated irrationality, Freudian theory and cognitive dissonance. In Wollheim R. and Hopkins J. (eds.) Philosophical Essays on Freud, Cambridge: Cambridge University Press, 279–288. Pears, D. (1984). Motivated irrationality, Oxford: Oxford University Press. Plato (2003) Phaedrus, transl. S. Scully, Newburyport: Focus Philosophical Library. Plato (1992). Republic, transl. G.M.A. Grube, Indianapolis : Hackett Publishing Company. Sartre, J.-P. (1943). L’être et le néant, Paris: Gallimard. Talbott, W. (1995). Intentional Self-deception in a Single, Coherent Self. Philosophy and Phenomenological Research, 55, 27–74. Thalberg, I. (1984). Freud’s Anatomies of the Self. In R., Wollheim, & J., Hopkins (eds.), Philosophical Essays on Freud, Cambridge: Cambridge University Press, 241–263. Wittgenstein, L. (1953). Philosophical Investigations, trans. Anscombe, Oxford: Blackwell.

Empirical and conceptual clarifications regarding the notion of ‘Core-Self’ from Gallagher’s and Merker’s Behavioural-Neuroscientific Proposals João Fonseca

I. Introduction: Conceptual Confusions and Methodological Fragmentation The notion of Core-Self’ has been used and formulated by several authors in different contexts and with different meanings, using diverse methodologies and approaches ranging from traditional conceptual analysis, Phenomenology, Cognitive Psychology and Neurosciences (Damasio, 1999; Gallagher, 2000; Northoff & Bermpohl, 2004; Pankseep & Northoff, 2009). This conceptual and methodological fragmentation threatens the very ambitions of the scientific task at stake. As a result, the current scenario concerning the studies on the nature of the (Core) Self presents us with a confusing profusion of concepts that tends to obscure the intellectual endeavour itself. This problem has recently been identified as such by some of the most prominent authors in the field. (Gallagher & Zahavi, 2008; Praetorius, 2009). I share the authors’ worries and also their diagnosis regarding the main causes of the problem: the proliferation of different methodological and disciplinary approaches to the subject. For instance, Gallagher & Zahavi state that: This disparity, which is both problematic and productive, is directly related to the variety of methodological approaches taken within philosophy and related interdisciplinary studies of the self. They include introspection, phenomenological analysis, linguistic analysis, the use of thought experiments, empirical research in cognitive and brain sciences, and studies of exceptional and pathological behaviour. One problem to be posed in this light is whether different characterizations of

166 

João Fonseca

self signify diverse aspects of a unitary concept of selfhood, or whether they pick different and unrelated concepts. (2008, pp. 197–198)

Although originally restricted to the more general notion of Self, this same observation applies neatly to what currently happens concerning the more specific notion of ‘Core-self’. In the recent literature, ‘Core-Self’ is loosely identified with the more basic and primitive form of self-consciousness present in all conscious experiences (such as basic emotions or conscious cognitive functions). But this is where the consensus among authors ends – there is not one consensus even at this basic and loose level of definition. Some authors suggest that the notion of Core-Self is a basic primitive (Zahavi, 2009) where others consider it to be a complex-layered concept (Panksepp & Northoff, 2009). To complicate this issue even more, some authors go as far as to propose the elimination of the very notion of ‘Core-Self’ (Dennett, 1991; Metzinger, 2003, 2009). The fundamental aim and motivation of this article is to bring some conceptual clarification to the notion of ‘Core-Self ’ taking into consideration what is stated in the last paragraph of the quotation above, i.e., “One problem to be posed in this light is whether different characterizations of self signify diverse aspects of a unitary concept of selfhood, or whether they pick different and unrelated concepts.” I intend to develop this line of questioning concerning the notion of ‘Core-Self ’. In order to achieve this goal, and also for matters of simplicity, I will confront/ compare two proposals regarding the notion of ‘Core-Self’ by adopting different methodologies: Bjorn Merker’s neuro-evolutionary perspective (2007) and Shaun Gallagher’s cognitive/phenomenological stance (2000). I will try to show that, in relation to the aforementioned problem, that the answer consists of a conciliatory move where we can state, without contradiction, that Merker’s and Gallagher’s characterizations both point towards a unitary concept and, at the same time, select different concepts of ‘Core-Self’. My strategy will fundamentally consist in redefining these two proposals within my own taxonomical framework for Behavioural Neuroscience and reformulating the notion of ‘CoreSelf’ as a Theoretical Concept in Behavioural Neuroscience within that same framework.

Empirical and conceptual clarifications

167

II. Fundamentals of a Model-Theoretical Framework for Behavioural Neuroscience I will adopt a model/semantic approach to neuroscientific practice (specifically, in the present context, that part of neuroscientific practice dealing with the explanation of behaviours, hence I will call this kind of activity behavioural neuroscience – or BN for short). The semantic view I adopt here has some connections with Ronald Giere’s approach to physics (Giere, 1988, 1999). The semantic approach to science endorsed by Giere focuses on scientific activity as a practice and how this practice is achieved and carried out by scientists as human cognisers. Likewise other model/semantic approaches to science are an explicit reaction against the Deductive-Nomological conception. In particular, Giere is quite sensitive to the way mainstream physics is academically communicated through textbooks (Giere, 1988). According to him, the main media of information displayed by those books are idealised physical entities and systems that do not exist in the world (e.g. frictionless motion). Given this, Giere’s proposal is to consider those objects and systems as abstract entities that constitute theoretical models of Classical Mechanics. These models have the status of abstract cognitive entities capable of being represented in several ways: equations, linguistic descriptions, graphic representations deployed by physicists. As Giere states: “[T]hey function as ‘representation’ in one of the more general senses now current in cognitive psychology. Theoretical models are the means by which scientists represent the world – both to themselves and for others.” (1988, p. 80). In this sense, models are interpreted intentional objects and not just structures as sometimes proposed.1 1

Current literature on the model-theoretical approach to science is vast and diverse. The word ‘model’ can mean very different things depending on the approach adopted. Sometimes the term ‘model’ is used to refer to entities or practices that help in the process of constructing explanation (as in the case of ‘scale models’). Differently from this perspective, my usage of the term ‘BN models’ refers to the explanations themselves. In this view, animal experimental models are part of the evidence in constructing BN models conceived qua explanations and so are not covered by my intended usage of the term ‘BN model’.

168 

João Fonseca

Just like models of classical mechanics, models of behavioural neuroscience (BN models) are intentional objects, cognitive representations used by scientists in several ways (here too, they can be deployed verbally, graphically or in other ways). More precisely, BN models state general explanations of behavioural phenomena. These models are achieved by inductive abstraction from previous neuroscientific empirical results. Usually, these intentional models are generalizations from animal experimental findings. For instance, results from protocol experiments of spatial memory in mice serve as a basis for constructing the explanatory content of a BN model for spatial memory covering all the animals that display this mental state, assuming the conservation across species of the relevant structures (in this case, homologues of the CA1 area of the hippocampus in vertebrates). This important topic will be addressed in more detail in section III. In structural terms, a BN Model (Mbn) is a triple: Mbn = , where: ‘B’ is a target behaviour or cluster of behaviours (e.g. the acquisition of different behaviours that different animals adopt in order to avoid a threat). B can group several behavioural phenomena (different – and exhibited by different species – but probably similar under a certain explanatory perspective) under a unique general concept – that fills ‘B’. For example, ‘The Acquisition of Avoidance Behaviour’. ‘f ’ states the functional profile of the explanation (imposing constraints on M). f states that there must be a neural mechanism M such that M explains B by stating the causal relation ‘M causes B’. Other constraints stated on f directly or indirectly qualify the nature of this causal relation. For instance, and depending on the explanatory target described in B: M is domain specific on B, M is multimodal, etc. In general, the causal claim, in conjunction with other constraints, qualifies the causal relation ‘M causes B’ as being necessary and sufficient (within the explanatory goal determined by the model itself). For instance, taking B as ‘The Acquisition of Avoidance Behaviour’, f could state that M is a neural structure such that M causes the acquisition of avoidance behaviour, M is domain specific concerning avoidance behaviour acquisition (i.e., its physical manipulation would only affect avoidance behaviour acquisition – at least in the intended, and therefore, relevant

Empirical and conceptual clarifications

169

experimental protocols)2, M is input multimodal (there is one structure for all kinds of avoidance acquisition, independent of the sensory modality involved in each particular case) , M is a memory/learning structure, etc. This set of constraints states the functional profile of the psychological concept ‘Fear Conditioning’ that explains ‘The Acquisition of Avoidance Behaviour’. Assuming the principle according to which ‘mental states cause/explain behaviours’ we can, more or less safely, establish that f states the mental/psychological concept that supposedly causally explains B. ‘M’ is a description of a neuronal mechanism satisfying the constraints expressed in f. Differently from f, M states a componential and not just functional explanation of B. In more formal terms, a mechanistic explanation M of a phenomenon B is a description of the (relevant aspects of the) physical mechanism that produces/causes B. More specifically (and adopting Carl Craver’s terminology used in several places. e.g. 2007) this explanation depicts a (finite) set of components (Φs) and activities (Ωs) organised in such a way as to produce the role or effect (Ψ). For example, concerning the functional profile of ‘Fear Conditioning’, a good candidate for a neural mechanist structure satisfying that profile is the Central Amygdala, since a causal connection between this structure and ‘the Acquisition of Avoidance Behaviour’ has been experimentally established and it also satisfies the other constraints (i.e. it is multimodal and domain specific). It should be noted, however, that mechanistic explanation M can assume different kinds of abstraction depending on the available knowledge we have regarding a specific explanation. In this case, and following Darden’s suggestion, we have different mechanism schemas which are “truncated abstract descriptions of a mechanism that can be filled with more specific descriptions of component entities and activities” (Darden, 2002, p. 356). Darden calls this process ‘Schema Instantiation’ and calls attention to the fact that these schema can come in different degrees of abstraction. 2

This ‘Domain-Specificity Constraint’ is required in order to avoid attribution of functions to any structure capable of affecting the target behaviour. For instance, it would be an error to attribute the function of ‘fear conditioning’ to the eyes, even if is true that a blind mice would not be capable of acquiring the avoidance behaviour based on visual cues.

170 

João Fonseca

Extensionally, Mbn range over a domain D of terrestrial multicellular animals i.e. the objects of Behavioural Neuroscience research are not counterfactually conceived extra-terrestrial or ‘possible world’ forms of life or even actual artefacts such as robots or any kind of artificial intelligent ‘device’. As an important ‘final statement’, BN Models reveal the structure of Theoretical Concepts of Behaviour Neuroscience, i.e. Psychological/ Mental Concepts. Psychological Concepts are Theoretical Concepts of Behavioural Neuroscience because they refer to the putative underlying/neuronal explanation/cause of B, where B is a Phenomenological Concept of Behavioural Neuroscience, grouping several observable behavioural phenomena. We can summarize this by stating that: for any Psychological Concept/Theoretical Concept of BN there is one corresponding BN Model (with the general form ‘’). BN Models simultaneously reveal the structure of Theoretical Concepts of BN and the meaning (and reference) of the corresponding theoretical terms. Henceforth I will use the terms ‘BN Models’ and ‘BN Concepts’ interchangeably.

III. Hierarchical taxonomy of psychological concepts: introducing ‘Nested Concepts’ A general and trivial feature of the family of concepts about a certain topic or cognitive domain concerns the fact that the concepts can, in a very intuitive and non-technical sense, include the cognitive content of another concept. Take the non-problematic example of the concepts ‘Dog’ and ‘Animal’ respectively within a zoology framework. It is obvious, in this case, that all the informational or cognitive content of Animal is included in the concept ‘Dog’ and the same with the latter and the concept ‘Labrador’. The extensional counterpart of this intensional relation consists in the fact that all objects satisfying the open sentence ‘x is a dog’ also satisfy ‘x is an animal’ (that is: the set of dogs is a proper subset of the set of animals). In this section I will claim that something similar is also true for Psychological/BN Concepts.

Empirical and conceptual clarifications

171

Consider, for instance, what we could call the Psychological Domain Memory Consolidation. A model/concept ‘Declarative Memory Consolidation’ includes all the information of the concept/model ‘Memory Consolidation Simpliciter’. Accordingly, the set of creatures satisfying ‘Declarative Memory Consolidation’ is a proper subset of those satisfying ‘Memory Consolidation Simpliciter’. One interesting consequence of this is a suggestion of a taxonomic hierarchy of psychological concepts varying in their intended extensional scope and informational content. Sometimes and in certain circumstances, psychological/BN theoretic concepts are ‘nested’ in each other satisfying the intentional and extensional conditions just suggested. That is the case, I would suggest, with the concepts ‘Memory Consolidation Simpliciter’ and ‘Declarative Memory Consolidation’: the first one is nested in the second since its informational content is included in the latter. Being so, Memory Consolidation can be considered as an example of a Psychological Domain determined by the nested relation between the concepts ‘Memory Consolidation Simpliciter’ and ‘Declarative Memory Consolidation’ (and perhaps other concepts). All these notions of ‘informational content inclusiveness’, ‘subset extensional relation’ and ‘Psychological Domain’ are clearly in need of a more technical refinement. In what follows I will explore the idea of ‘Nested Concept’ and related notions in greater detail. I should start with the fundamental claim that the proposed model/conceptual framework for Behavioural Neuroscience is constrained by evolutionary considerations. Two of the most important are: (i) The conservation of phenotype traces across taxa through homology and (ii) (stated verbally) The ‘fact’ that Evolution does not start from scratch (in other words: adaptive – if not other – requirements of transformation, modification or ‘specializations’ of certain biological functions are achieved by ‘working upon’ the already existing phenotype structures responsible for that function). (i) and (ii) are closely connected (in particular, (ii) assumes (i)). By assuming these principles we should be able to answer the question: How do we establish that two different BN concepts satisfy the ‘nested relation’? Firstly, we should know how to settle the extension of a certain BN concept. Regarding this question, neuroscientists (implicitly or explicitly) assume principle (i) as a rationale in their induction from targeting

172 

João Fonseca

experimental animals models to class generalisations. Commenting on an inference proposed by Squire concerning the neural structures underlying ‘declarative memory’, John Bickle states that: This approach forges a connection between human neuropsychological data and experimental mammalian research. The ‘particular structures and connections’ namely the hippocampus proper, entorhinal cortex, perirhinal cortex, and perihippocampal gyrus, have homologs across the mammalian class. Since declarative (or explicit) memory is coextensive with hippocampal-requiring memory, the term is applicable to memory research on humans, other primates and rodents. (2003, p. 78)

In evolutionary theory, the term ‘Homologue’ means precisely a phenotypic trait conserved through species that share a common ancestor. The answer for our question is that scientists extensionally induce f (in particular, the claim that ‘M causes B’, by both assuming structural homology (e.g. the hippocampus) and functional homology (e.g., Declarative memory) through a class of species belonging to a single evolutionary clade (i.e., sharing the same common ancestor). In the case mentioned in the quotation, the function Declarative Memory is inductively presumed to exist in all the animals (mammals) that share the Hippocampus as a common phenotypic trace conserved in all the members of the clade. Taking this into consideration we should now ask: what principles govern the hierarchical relations comprising Nested Concepts? Taking again the examples of ‘Memory Consolidation simpliciter’ and ‘Declarative Memory Consolidation’, consider the following quotation from Squire and Kandel that adopts evolutionary principle (ii) as an explicit rationale: The simplest memory capabilities, and those that seem to have appeared earliest in evolution, seem to be nondeclarative memories related to survival, feeding, mating, defense and escape. As a variety of additional types of nondeclarative memory and then declarative memory evolved, the new memory processes retained not simply a set of genes and proteins, but entire programs for switching on and stabilizing synapse connection. Moreover, these common mechanisms have also been conserved through the evolutionary history of species: they are found in both simple invertebrates such as Drosophila and Aplysia and complex mammals such as mice. (1999, p. 155)

Two important and related questions arise here: the process by which the information is shared (in a nested/inclusive relation) among BN concepts,

Empirical and conceptual clarifications

173

and how that relation is ‘translated’ in extensional terms. The sharing of information in an inclusive way among models follows the answer to the question of how the extension of a concept is inferred (conserved phenotype and functional traces through species) and adds to that which is stated in principle (ii): modifications (‘specializations’) of a certain function are achieved by modifying the already existing structures responsible for that function. This means that a certain structure responsible for a certain function is conserved in the more specific functional modifications (this process is well attested in the above quotation). Regarding again the above example concerning ‘Memory Consolidation simpliciter’ and ‘Declarative Memory Consolidation’, we can propose (in a simplistic way) the current proposed neural mechanisms corresponding to each of the two different types of memory as follows: ‘Memory Consolidation simpliciter’ = ‘cAMP-PKA-CREB Molecular Pathway’ (see, for example, Bickle, 2003); ‘Declarative Memory Consolidation’= Early to Late-Long Term Potentiation switch in the Synapses of Hippocampus Cell’s. Following Squire and Kandel (1999), ‘Declarative Memory Consolidation’ is a function that was ‘built upon’ the evolutionarily conserved structures corresponding to the more general and basic form of ‘Memory Consolidation simpliciter’, which underlies all kinds of Memory Consolidation (be them ‘declarative’, ‘non declarative’, or any other specification on what kind of memory is being consolidated – more on this below). Therefore, the mechanistic abstraction M of relative to ‘Declarative Memory Consolidation’ (i.e. E-LTP to L-LTP Switch at the level of hippocampal synapses) has as one of its ones components the ‘cAMP-PKA-CREB Molecular Pathway’ (specifically, the latter corresponds to the ‘E-LTP to L-LTP Switch’ ‘part’ of the more complex mechanism of ‘Declarative Memory Consolidation’). Simply stated, the Mechanistic Schema of ‘Memory Consolidation simpliciter’ ‘is part of’ or ‘is filled in by’ the (in comparison) less abstract Mechanistic Schema of ‘Declarative Memory Consolidation’, i.e., the information stated in the model of the former is included in that of the latter. Regarding the behavioural component B of for ‘Memory Consolidation simpliciter’ it abstracts consolidation of behaviours regardless of being declarative or non-declarative. The B for ‘Declarative Memory Consolidation’, on the other hand, retains the fundamental information from the former and adds to it more informational specifications.

174 

João Fonseca

As a matter of clarification, it should be emphasized that the Psychological Concept/BN Model ‘Memory Consolidation Simpliciter’ does not correspond to a particular very simple kind of memory consolidation deployed by a very ancient and simple creature. On the contrary, the model/concept corresponds to an abstraction (a very abstract Schema) that is satisfied by many different specific kinds of memory consolidation (e.g.: Declarative) deployed by a large variety of species. As Bickle says regarding the molecular mechanisms corresponding to ‘Memory Consolidation Simpliciter’: [T]he molecular and gene composition and the intracellular pathway interactions are shared across species. These shared features obtain despite vast differences in brain size, organization, site of principal effect (presynaptic or postsynaptic), behavioural repertoire, and even ‘cognitive logic’ of the distinct types of memory being consolidated (declarative versus nondeclarative). (2003, p. 148)

The concept ‘Memory Consolidation Simpliciter’ is abstract to the point that it can encompass a vast array of target behaviours and does not specify if the proposed mechanism ‘cAMP-PKA-CREB Molecular Pathway’ occurs pre or postsynaptically. This information can be added to and made more fine-grained when we specify the particular kind of memory consolidation we are seeking to understand (e.g.: Declarative Memory Consolidation). Thus, we can state something like a general intensional condition (GIC) regarding BN Nested Concepts: GIC: if a concept C1 is nested in concept C2 and C1= and C2 = , then all the information stated in M and B is included in M* and B*. It is not difficult to translate the above line of reasoning to the question of the extensional relation between nested models/concepts of BN. From what was stated above, we can easily envision that the class of species which are true of the open sentence ‘x has Declarative Memory Consolidation’ is a sub-set of the class of species which are true of the open sentence ‘x has Memory Consolidation Simpliciter’. Sustaining this inference is the evolutionary condition stating that a class A of living creatures is a subset of another class B, if the evolutionary clade corresponding to class B is Basal in relation to the evolutionary clade corresponding to class A.

Empirical and conceptual clarifications

175

A clade is basal in relation to another clade if it contains that other clade as a branch within it (both clades share a common ancestor). For instance, whereas ‘Memory Consolidation Simpliciter’ is shared by a vast number of species including invertebrates and vertebrates – probably all multicellular animals (let us assume for the sake of the argument), – ‘Declarative Memory Consolidation’ is possessed only by mammals (and presumably other non-mammal vertebrates). In more formal terms, the clade corresponding to ‘Memory Consolidation simpliciter’ (i.e., all species possessing this trait are descendent from a common ancestor also possessing the trait) is basal in relation to that corresponding to ‘Declarative Memory Consolidation’ and, therefore, the class of Mammals is a sub-set of the more general and comprehensive class of Multicellular Animals. We are now in position to state something like a general extensional condition (GEC) regarding BN Nested Concepts: GEC: if a concept C1 is nested in concept C2, then the class of objects of the domain satisfying ‘x is C2’ is a proper subset of the class of objects of the domain satisfying ‘x is C1’. From GIC and GEC we can now introduce the notion of ‘Psychological Domain’ that mixes extensional and intensional parameters regarding nested BN concepts. A Psychological Domain refers to the sets of actual neural mechanisms and behaviours referred to by the abstractions stated in B and M. A general definition for ‘Psychological Domain’ states: Def.: A Psychological Domain PD is a set consisting of the union of a set of actual neural mechanisms AM and a set of actual behaviours AB such that, if C1= and C2 = , and C1 is nested in C2, then AM is the extension of M, AB is the extension of B, AM* is the extension of M* and AB* is the extension of B* (where AM* is a proper subset of AM and AB* is a proper subset of AB – given GEC). Departing from the definition of a Psychological Domain (PD) the General Intensional Condition (GIC) and the General Extensional Condition (GEC), we are now in a position to formulate 3 structural conditions regarding the nested relation between BN concepts. These conditions

176 

João Fonseca

are structural in the sense that they imply empirical claims regarding the causal structure of reality (in particular those parts of reality denoted by the sets AM, AB, AM* and AB*). This means that for a particular case of two BN concepts being nested there are empirical claims those concepts have to be capable of verifying (at least in principle). If C1 and C2 are BN concepts, if C1= and C2 = < B*,f*,M*> and C1 is nested in C2 defining an extensional Psychological Domain PD, then C1 and C2 must conjunctively fulfil the following three structural conditions: 1. Necessary Structural Condition: M (of C1) states necessary structural (neural-mechanistic) conditions for AB (i.e., all the behaviours comprising PD). Empirical claim (regarding 1): if (by means of an ideal intervention) all elements of neural mechanisms AM are physically inhibited, then all the elements of the set of behaviours AB of PD will be compromised. Example: Assuming that the concept ‘Memory Consolidation Simpliciter’ (MCS) is nested within the concept ‘Declarative Memory Consolidation’ (DMC), then (by (1) and its empirical claim) if the M of MCS (‘cAMP-PKA-CREB Molecular Pathway’) is inhibited, then all behaviours of the Psychological Domain Memory Consolidation will be compromised. In fact, several studies have shown that this is the case. Take, for instance, the important and influential work by Bourtcholadze et al. (1994) where two distinct kinds of learning (dependent on declarative/hippocampus and non-declarative/amygdala memory respectively) were tested regarding CREB knockout mice. John Bickle summarizes their results as showing that “the loss of CREB function disrupts long-term memory for both contextual (declarative) and cued (nondeclarative) conditioning.” (2003, p. 90) 2. Sufficient Structural Condition: M* (of C2) states sufficient structural (neural-mechanistic) conditions for the set difference AB\AB* (i.e., all elements/behaviours of AB that are not members of AB*). Empirical claim (regarding 2): if (by means of an ideal intervention) all elements of neural mechanisms AM* are physically inhibited, then all the elements of the set of behaviours from the set difference AB\AB* of PD will be compromised.

Empirical and conceptual clarifications

177

Example: Assuming again the MCS and DMC nested relation. This condition states that if the M* of DMC (i.e., ‘E-LTP to L-LTP Switch at the level of hippocampal synapses’) is disrupted, then only a sub-set of behaviours from the Memory Consolidation psychological domain will be compromised. Considering the general role of the hippocampus in memory (assuming here for simplification that the consolidation process is included in this more general picture) it is well known, since the seminal observations of Patient H.M., that the hippocampus role in memory is selective to what was then called Declarative (or Explicit) Memory. As Squire (1992) puts it: “in the absence of the hippocampus several other kinds of learning can still be accomplished, including the learning of skills and habits, simple conditioning, and the phenomenon of priming” (Squire, 1992, p. 196). 3. Unitary Mechanism Condition: Given GIC, all the information stated in M (of C1) must be included in M* (of C2) (i.e., M must be a submechanism of M*). As a matter of causal mechanistic explanation, if M is a submechanism of M*, then all members of AM* have to be physically/causally linked to some members of AM described by a mechanism schema M** in a way not stated in M or M*. Empirical claim: we should be able to empirically detect the physical components and activities regarding the neural mechanism M**. Example: in the Memory Consolidation example this condition states that there must be a physical causal link between the M of MCS (‘cAMP-PKA-CREB Molecular Pathway’) and the M* of DMC (‘E-LTP to L-LTP Switch at the level of hippocampal synapses’). The nature and structure of such a link has been profusely studied as a matter of explaining what are the molecular and sub-cellular mechanisms underlying the E-LTP L-LTP link. It is more or less well established how the cAMP-PKA-CREB mechanism affects the nucleus of pyramidal cells (via phosphorylation of CREB) and produces proteins that ultimately affect the structure of the cell at the level of synapses. A detailed description of such a mechanism lies beyond our purposes here. All we need to acknowledge is the existence of such a physical link between M of MCS and M* of DMC as predicted by the application of this third condition. We are now in a position to state the conditions two BN concepts must satisfy in order to establish between them a nested relation and,

178 

João Fonseca

henceforth, determine a Psychological Domain PD. A BN concept C1 is nested in another BN concept C2 (i.e., they satisfy both the extensional and intensional conditions stated in GEC and GIC and define a Psychological Domain PD) only if they satisfy the three structural conditions stated above (Necessary Structural Condition, Sufficient Structural Condition and Unitary Mechanism Condition).

IV.  R  edefining Merker’s and Gallagher’s proposals for Core-self Having presented the fundamentals of the framework, my purpose now is to consider both Merker and Gallagher’s proposals for ‘Core-Self ’. Hence, I will rephrase both proposals, considering Core-Self as a Theoretical Concept in Behavioural Neuroscience using the basic tools summarized in the previous sections. The basic aim of this conceptual clarification is to answer the question mentioned in the introduction, posed by Gallagher & Zahavi (2008), according to which: “[o]ne problem to be posed in this light is whether different characterizations of self signify diverse aspects of a unitary concept of selfhood, or whether they pick different and unrelated concepts”. IV.1 ‘Core-Self’ as a Theoretical Concept in Behavioural Neuroscience According to the framework, one could try to understand the notion of ‘Core-Self’ as a Theoretical Concept in Behavioural Neuroscience. Therefore, we can assume initially that ‘Core-Self’ corresponds to a BN Model assuming the general structure: . Given the particularities of both Merker and Gallagher’s proposals, I will start by addressing the functional description f in the model of Core-Self. Following on from what has been stated above, f expresses the functional profile of the psychological concept in question. Besides the ‘mandatory clause’ stating that ‘M causes B’, other more specific clauses

Empirical and conceptual clarifications

179

are introduced. The most important property of Core-Self emphatically stated by both Merker and Gallagher concerns the cognitive ability of basic ‘Decision Making’. The most important trait of Core-Self, according to these authors, consists of a basic subjective substratum capable of discriminating objects and events in the world and judging them accordingly in terms of the actions to take on each occasion. Important for this is what both authors call ‘sense of agency’ as a fundamental characteristic of Core-Self; Merker says that “[Core-self] enters consciousness only in the form of motivated agency” (2005, p. 96), whereas Gallagher & Zhavi equate Core (or Minimal)-Self with an “experiential sense of agency” (2008, p. 160). This fundamental property of CoreSelf is shared by other authors and is encapsulated in the following quotation from Panksepp and Northoff: “We envision the core-self to allow organisms to be active agents” (2009, p. 160). We will describe later how both Merker and Gallagher approach this idea in a more detailed way. For now, we can try to define the functional profile f of ‘Core-Self’ as stating the following clause (in addition to the ‘mandatory’ clause stating: ‘M causes B’): ‘M allows purposive voluntary non-reflexive action through a subjective sense of agency (M is the substratum for basic – immediate – decision making)’. Given the task of ‘building’ a BN Model for ‘Core-Self’ based upon Merker and Gallagher’s proposals, I have to risk proposing what the ‘bare’ target behaviour B should be, since neither Merker nor Gallagher are explicit concerning this point. Assuming that (as stated above) the basic functional trait of ‘Core-Self’ is the capacity for basic decision-making (through a subjective sense of agency), I risk suggesting that the target behaviour B of the model for Core-Self consists of ‘Operant/instrumental conditioned behaviour’. The following quotation commenting on the nature of ‘Operant Conditioning’ is clear enough in establishing the link between the fundamental functional trait of coreSelf (decision making) and this kind of behaviour: The distinction between Pavlovian and operant conditioning therefore rests on whether the animal only observes the relationships between events in the world (in Pavlovian conditioning), or whether it also has some control over their occurrence (in operant conditioning). Operationally, in the latter outcomes such as food or shocks are contingent on the animal’s behaviour, whereas in the former these occur regardless of the animal’s actions […] The scientific study of operant conditioning

180 

João Fonseca

is thus an inquiry into perhaps the most fundamental form of decision-making. It is this capacity to select actions that influence the environment to one’s subjective benefit, that marks intelligent organisms. (Staddon & Niv, 2008, Emphasis added)

So, it is safe enough to suggest that, in the BN Model for ‘Core-Self ’, the respective target behaviour B states: ‘General (i.e., all kinds of) operant/instrumental conditioned behaviours’. The functional profile of this model states that these behaviours are caused by a neural mechanism M that allows purposive voluntary non-reflexive action through a subjective sense of agency. As such, Merker and Gallagher’s proposals agree (by assumption) both on the target behaviour B and the functional profile f of the putative model for Core-Self. Where they disagree sharply is on the third member of the ordered triple of the model for Core-Self. That is, they propose very distinct neural mechanisms M for implementing what is stated in f. A closer look at each proposal is now necessary with regard to the neural implementation of Core-self. IV.2  Merker’s upper brainstem proposal What follows is an over-simplification of Merker’s detailed proposals and suggestions. The aim of this section is to briefly present Merker’s perspective concerning the neural mechanism implementing ‘Core-Self’. Merker’s view regarding Core-Self is an adaptative-evolutionary one. In his view, the emergence of consciousness (and Self) was nature’s response to two different (although deeply related) evolutionary pressures: 1- The Stability Problem and, 2- The Decision Making Problem. In what follows I will briefly present each pressure separately and show what evolution’s solution to each one has been. Let us focus on the first evolutionary pressure The Stability Problem. This problem is well illustrated by what Merker calls The ‘Earthworm Dilemma’: Earthworms display a swift withdrawal reflex to cutaneous touch. […] Consider the worm’s initiation of a crawling movement. Such a movement will produce sudden stimulation of numerous cutaneous receptors, yet no withdrawal reflex is released to abort the movement. (2005, p. 90)

Empirical and conceptual clarifications

181

The solution to this dilemma (how to distinguish self-produced from other-produced sensory inputs) has a very simple solution in the case of worms by simply distinguishing excitatory vs. inhibitory synapses related to self-produced movement. The solution to this problem has, of course a high survival value. The real problem is not with worms but rather concerns how more complex bodies solve this same problem/ dilemma. Here is the problem as stated by Merker: The evolution of higher animals leads not only to increased complexity of single sensory and motor systems but produces a diversification of such systems in the equipments of a given species. Vision, hearing, touch, smell, taste, entroception, proprioception and vestibular systems are some of those on the sensory side, while a great variety of locomotors, orienting, grasping, and manipulatory appendages – often paired in sets and with multiple, independently moving joints – proliferate on the motor side. Such diversity brings special problems in its train. On the sensory side, for example, the receptor arrays of different modalities are often disposed in different parts of the body. They are therefore, not affected in the same way by self-motion and so cannot be subject to the same compensatory remedies, nor be integrated directly. (2005, p. 92)

In a nutshell, there is a complexity problem at the level of sensorimotor integration in vertebrates given the several distinct sensorial modalities, the myriad of different possible body actions or the fact that the different sensorial transducers are located in different parts of the body, therefore being affected differently by the same movements (the ‘orientation problem’). There appears to be too many scattered sensorial-motor variables that, if kept separately, would prevent the distinction between self-produced from other-produced sensory inputs. What is needed here is an Integration (for action) of different sensorial and motor ‘local solutions’. We need a single coordinate reference system in order to frame all this proliferation and diversification. So, what was evolution’s ‘solution’ to this problem? According to Merker, nature’s structural answer relates to the phylogenetically very old and widely conserved structures corresponding to the basic Vertebrate Brain Plan, namely the roof/tectum of the midbrain/Superior Colliculus (S.C.). Merker stresses the peculiar anatomical and functional roles of this structure:

182 

João Fonseca

[The superior colliculus] is the only site in the brain in which the spatial senses are topographically superposed in laminar fashion within a common, premotor, framework for multieffector control of orienting. Its functional role appears to center on convergent integration of diverse sources of information bearing on spatially triggered replacement of one behavioural target by another (2007, p. 67).

The S.C. laminar structure is such that some information registered in one layer (say a mapping of auditory space) translates directly to another layer (say, a mapping of oculomotor orientation) by means of coordinate transformation. This is why when we hear a sound, our eyes ‘know’ immediately where to look in the search for the source of the sound. What this means is that the S.C. laminar structure in fact brings the various local sensory-motor solutions under a common referential system framework. The end result is that when someone moves his or her own body (the eyes for instance) the world seems to remain stable. Consider now what happens if you slightly tap your eyeball. You will notice that the world seems to be moving independently. This is so because your eyes move slightly without you intentionally doing this. The layer in the SC responsible for a retinotopic map of stimulus registers change whereas the layer responsible for the oculomotor map does not. There is a mismatch in the coordinate transformation of information from one layer to another. When there is a coordinate transformation from a motor map layer to a sensorial map layer, the result is a sense of an external stabilized world. According to Merker this is, therefore, nature’s solution to the evolutionary pressure of the stabilization problem in more complex (vertebrate) bodies. Let us turn now to the second selection problem faced by creatures with a complex body and movement; The Decision Making Problem. According to Merker, the evolution of real time decision making is the result of the merging of three distinct biological selection domains: 1Needs/Action Selection: ‘time axis’ – what action should be taken now and next; 2- Opportunities/Target Selection: ‘space axis’ – which targets in the world should be ‘chosen’; 3- Motivational Ranking: some targets should be avoided while others should be approached. Again, how did evolution ‘solve’ the problem of integrating these three selection domains into one single decision space? According to Merker, it was by answering this question that consciousness arose

Empirical and conceptual clarifications

183

in biological systems, since it provides an optimal solution to the decision-making problem. The question, of course, is ‘how’. The answer is somewhat complex. Therefore, in order to explain Merker’s views more easily I will divide them into three different elements. The first element for the answer relates directly to the previous solution to the first evolutionary pressure, i.e., the multimodal sensory-motor common reference framework at the level of SC. According to Merker, as we have just shown, this provides a phenomenological contrast between an external stabilized world and my own intended actions. But this is an Abstract/Formal contrast. It should be stressed that this ‘Common Reference System Framework’ does not depend on particular content – it is, so to speak, a built-in formal framework. n a Kantian vein, we can conceive this as providing a kind of pure intuition of sensibility, i.e., a spatio-temporal formal aspect of conscious experience. However, as Kant taught us, forms without content are empty. So, if the first element deals with the formal aspect of consciousness, the second element regards the content aspect of that formal frame. More particularly, this is the domain of conscious ‘representations’ lying within the formal frame. Two major domains of representation are of interest here: the representation of the external world in terms of its objects and events (what Merker calls an ‘analogue representation of the world’) corresponding to the target selection axis of the decision problem. The other pole to take into consideration regarding conscious content representations is a particular and special object in the world: the body (what Merker calls the ‘analogue representation of the body’) that corresponds to the action selection side of the decision problem. Finally, the third element relates more directly to the phenomenological aspect of consciousness. It consists of sensations/feelings/emotions signalling the motivational ranking of targets in the world, serving as bias in the decision process. We have, therefore, the merging, through consciousness, of the three selection domains: 1 action selection – perceived (analogue) body; 2- target selection – perceived (analogue) world: and 3- motivational ranking- emotional feelings, within a single frame conceived for optimal real time decision-making. Considering now the neural implementation of this functionally described solution to the Decision Problem, Merker’s suggestion is

184 

João Fonseca

that biologically evolved structures at the level of the upper brainstem constitute the structural ‘side’ of this solution: close interconnection between 1- Superficial layers of SC (the multimodal shared integrated spatial coordinate system – ‘space-time sensory world’); 2- Substancia Nigra/Basal Ganglia (motor commands – Body); 3- PAG/Hypothalamus (motivation – feelings/emotions). These three poles: the multi-modal/ reality, the hedonic/motivational and the motor commands, functionally and structurally interconnected, provide the basics for a (phenomenic) conscious decision making process. According to Merker, this complex structure is capable of implementing a ‘virtual’ Core-Self in the form of an implicit ‘ego-center’ subjective ‘decision-maker’: There is reason to believe that the implicit “ego-center” origin of this coordinate space is the position we ourselves occupy when we are conscious […] The ego-center places the conscious subject in an inherently “perspectival”, viewpoint-based, relation to the contents of sensory consciousness. (2007, p. 72)

However, Merker suggests a possible specific neural implementation of (Core) Self, turning it, so to speak, more ‘real’. Merker’s hypothesis concerns the possible role of the Zona Incerta (or Thalamus Ventralis in non-mammals) in the resolution of remaining decision making problems not ‘completely solved’ by the ‘superior culliculus – PAG / hypothalamus – Substancia Nigra/Basal Ganglia’ complex: [a] sophisticated control system needs some means for resolving residual conflict among alternatives left unsettled by routine mechanisms because of stochastic happenstance in a complex multicomponent system or because of exceptional combinations of contingencies encountered in a lively and unpredictable world. […] At the level of the core control system the Zona Incerta […] seems optimally placed to play such a role (2005, p. 103).

The Zona Incerta is also an (sub-cortical) upper-brainstem structure massively connected with the other structures (sensorial and emotional) of the upper brainstem. According to Merker, this structure and its above mentioned putative functions could ’supply consciousness with that subjective presence of a tacit “witness” to its contents which is at the heart of the very concept of consciousness. These characteristics would amount, in other words, to a subjective sense of self, not in the

Empirical and conceptual clarifications

185

sense of self-image – which is a content of consciousness – but in the sense of the nodal monitoring function presupposed by all contents of consciousness’ (2005, pp. 104–105). Independently of the two possibilities; a ‘virtual’ arising from the ‘superior culliculus – PAG /hypothalamus – Substancia Nigra/Basal Ganglia’ interaction, or a ‘real’ one directly implemented in the Zona Incerta, what matters for the present purposes is that Merker’s proposal concerning the neural implementation of the ‘Core-self’ points to an upper-brainstem structure, evolutionarily conserved across vertebrates, predating the emergence of neo-cortical structures in evolutionary more recent creatures (namely mammals). Therefore, with regard to the BN Model for Core-Self and according to Merker’s proposal, the neural mechanism ‘M’ could be stated, generally and loosely, as consisting of the Upper-brainstem complex (more specifically in the ‘superior culliculus – PAG /hypothalamus – Substancia Nigra/Basal Ganglia – (and possibly) zona incerta’ complex. IV.3  Gallagher’s cortical motor forward model proposal Shaun Gallagher’s departure concerning the neural implementation of core-self rests on his phenomenological definition of this concept. For Gallagher, the notion of Core-self or, as he prefers, the minimal-self, is defined firstly, at a more general level, as “consciousness of oneself as an immediate subject of experience” (2000, p. 15) and secondly, at a more specific level as the conjunction of two phenomenic experiences: the sense of agency and the sense of ownership. Both ‘senses’ (of agency and ownership) are conjunctively necessary conditions for Intentional Action and Decision Making. More specifically, the sense of agency corresponds to the experience that I am in control (I am responsible) of my action; this makes the action in question to be considered a voluntary and intentional one. On the other hand, the sense of ownership corresponds to the experience that a certain movement is mine, be it voluntary or not: even in a pure reflex movement I can report it as being mine. Gallagher deploys a control engineering schematic approach currently used in contemporary cognitive neuroscience (see, for instance Grush, 2004; Wolpert, Mial and Kawato, 1998). The phenomenon under

186 

João Fonseca

consideration is motor control and Gallagher’s aim is to locate the roles of the senses of ownership and agency in the schema. Although complex, the fundamentals of this system are encapsulated in the following: “[A] comparator mechanism operates as a part of a non-conscious premotor or ‘forward model’ that compares efference copy of motor commands with motor intentions and allows for rapid, automatic error corrections […] This comparator process anticipates the sensory feedback from movement” (Gallagher, 2000, p.16). Two different control models make up the overall mechanism: the forward model and the feedback model. This distinction is important since Gallagher suggests that the sense of agency corresponds to the forward model – by means of a sense of anticipation of the movements comprising an intentional action – and the sense of ownership corresponds to the feedback model – the feedback from actual movements allows the agent to identify it as movements from his or her own body. Given this, the question arises what about the neural implementation of the mechanism responsible for intentional action and, in particular, regarding the two component models (forward and feedback)? Departing from cognitive neuroscience empirical data, Gallagher suggests that the main areas relating to the forward model and hence, to the sense of agency, correspond to the Motor, Pre-Motor and Prefrontal areas of the neo-cortex. On the other hand, the feedback model (responsible for the sense of ownership) seems to be implemented in the Cerebellum. Gallagher illustrates this hypothesis with empirical studies concerning schizophrenia: Schizophrenic patients who suffer from thought insertion and delusions of control also have problems with this forward, pre-action monitoring of movement, but not with motor control based on a comparison of intended movement and sensory feedback. The control based on sensory feedback is thought to involve the cerebellum. By contrast, problems with forward monitoring are consistent with studies of schizophrenia that show abnormal pre-movement brain potentials associated with elements of a neural network involving supplementary motor, premotor and prefrontal cortexes. Problems with these mechanism might therefore result in the lack of a sense of agency that is characteristic of these kinds of schizophrenic experience. (2000, pp. 16–17)

In fact, schizophrenics report that their actions are not under their command and will; they seem to lack a sense of agency. The phenomenological

Empirical and conceptual clarifications

187

failure in feeling their actions as caused by themselves promotes delusional reasoning concerning the real element ‘responsible’ for their actions: ‘I’m not causing my movements, someone else is’. Since the sense of agency is a necessary condition for a ‘normal’ core or minimal self to arise, this illness can be considered, under these current assumptions, as resulting from a ‘truncated’ core/minimal-self. Thus, if according to Gallagher the core/minimal-self can be identifiable with the conjunction of the sense of agency with the sense of ownership and if we accept the proposed suggestion for the neural identification of each one of these phenomenic experiences, we can identify the ‘M’ in the BN Model for ‘Core-Self’ as: ‘Motor, Pre-Motor and Prefrontal areas of the neo-cortex plus Cerebellum’.

V. Conciliating Merker and Gallagher’s proposals: ‘Core-Self’ as a Theoretical Nested Concept in Behavioural Neuroscience. V.1  The ‘Explanatory Dilemma’ We have just testified Merker and Gallagher’s disagreement concerning the proposed neural implementation of Core-Self. If we start by assuming that both proposals are empirically correct, then, as I claim, we face a kind of ‘Explanatory Dilemma’: either 1. the two proposals select the same concept and the claim of Multiple Realization of ‘Core-Self ’ is vindicated, or 2. the two proposals select different and unrelated concepts. The explanatory ‘dilemma’ (so to speak) consists of the fact that either of the two possibilities implies a fragmentation in our explanatory practices where some kind of unity would be preferable. In this section, I will address this problem and show how it is possible to suppress this. In particular, I will try to show how, in the present context, the two proposals relate to each other in a way not predicted by the rationale implicit in the formulation of the dilemma.

188 

João Fonseca

The first step in addressing the dilemma consists of challenging the Multiple Realization (MR) claim. The MR claim assumes that the two proposals have exactly the same functional profile as stated (within the framework adopted here) by the ordered pair . The fact that both proposals (assuming that they are both empirically correct) only disagree in the neural realization of the same functionally described profile of ‘Core-Self’ seems to support the MR claim that this particular concept is differently realized in different species or individuals. I will try to show that this claim, in fact, is not vindicated; while maintaining the assumption that both proposals are empirically correct, I will show that the two proposals differ in their functional profile and, as a natural consequence, they select different concepts. The question that naturally arises concerns in What and How the two proposals differ in their functional claims. In order to answer this crucial question we must look closely at what both authors claim concerning the functional properties of their respective proposals. As we described in section III, both proposals address the mechanisms (cognitive and neural) underlying basic purposive/voluntary motor actions. In this particular, and at this general level, both proposals seem strikingly similar. It is my contention that the key for detecting a fundamental difference between the two proposals consists in looking at their proposed solutions rather than their (apparently similar) problems. Recalling Merker’s proposal, we have seen that, fundamentally, his suggestion consists of the integration of various sensory modalities, body representation and emotional values within a common coordinate system at the upper brainstem. At a more abstract/functional description, Merker’s solution consists of the evolved common state space integrating all the other domains (body, world and emotions) otherwise kept separate. In other terms, and assuming the risk of great simplification, his proposal can be understood as a dynamic system where some variables change their values while others remain constant when some motor action takes place. According to some authors, this constant interaction of values is sufficient for the emergence of a subjective sense of an external world distinct from a self responsible for body agency in that world (see, for instance, Philipona et al, 2003). In its fundamentals, Merker’s solution is computationally very simple – for instance, the ‘sensorial multimodal – body’ interaction taking place at the roof of

Empirical and conceptual clarifications

189

the superior culliculus assumes the form of a simple direct mapping look-up table resulting in a pre-motor framework for (among other functions) body orientation in space. If we take Merker’s proposed computational solution as a reference point of departure, we can testify that Gallagher’s proposal is, from a computational/control theory point of view, much more complex. Whereas Merker’s suggestion is not, at its core, much more complex than a simple look-up table mapping input-output, Gallagher introduces a more complicated scenario with looping and re-entrance information-processing culminating with the idea of a motor forward-model. Basically, the role of a motor forward-model (FM) consists of the refinement of voluntary movements, in particular rapid ones. At its essence, the FM compares the relation between the intended state of the body (e.g. The position of my arm in relation to a certain target) with its actual state. ‘Running’ an FM allows error correction before the actual execution of the movement. The end result is a smooth and precise rapid voluntary movement. The resulting ‘sense of agency’ is a by-product of this FM. There are other cognitive features of the FM worth mentioning not addressed previously but important within the current context. A very important cognitive property of systems with the architecture of an FM, as several authors have noticed, consists of the fact that, when running off-line (independently of actual motor movements) the FM is capable of produce motor imagery (i.e., the capacity of imagining movements without actually making them) (see Churchland, 2002; Grush, 2004). This imaginary capacity is, nevertheless, important for actual behaviour in the environment since “[o]ff-line imagining yields an advanced peak at the likely consequences of pondered actions, which permits undesirable consequences of contemplated actions to be foreseen and avoided” (Churchland, 2002, p. 81). How do these cognitive features and properties of an FM relate to Merker’s upper-brainstem proposal? An important clue is given in a note in Merker’s article for Behaviour and Brain Sciences. In that note Merker states the following: To avoid possible misunderstandings of this key point, note that the analogue “reality simulation” proposed here has nothing to do with a facility for simulating things such as alternate course of action by, say, letting them unfold “in imagination” or any other versions of an “inner world,” “subjective thought,” “fantasy”

190 

João Fonseca

or the like. Such capacities are derivative ones, dependent upon additional neural structures whose operations presuppose those described here. (2007, p. 81)

This remark is crucial in the present context since it establishes a fundamental cognitive difference between the two proposals. From what is stated in this quotation, it becomes clear that Merker’s proposal is, at a cognitive level, simpler than Gallagher’s. Gallagher proposes a cognitive scenario, presupposed by the FM, where ‘imagery’, ‘off-line planning’ and other cognitive properties are the fundamental traits. These traits are, as we have described, absent in Merker’s perspective, which is much simpler, and provides the fundamentals for these more complex cognitive tasks to arise. If we take this cognitive difference seriously, then we can conclude, after close scrutiny, that the cognitive profile of both proposals is not the same after all. Things being so, I feel comfortable enough to address the MR claim as initially proposed. It is clear, or so I hope, that the MR claim regarding the notion of ‘Core-Self’ is thus seriously called into question. Recall that, in addition to the assumption that both proposals are empirically correct, the MR claim rests on the supposition that those two proposals have exactly the same functional profile as stated by the ordered pair . This is precisely the conclusion I have shown to be false. So, the MR claim is not vindicated. However, another consequence of this conclusion is that each proposal picks a different concept of ‘Core-Self’. But this leads us to the second horn of the ‘dilemma’ since this ‘solution’, just like the MR one, has the unwelcome consequence of an explanatory fragmentation regarding the notion of ‘Core-Self’. The splitting into two different concepts does not seem to help us. We are back to the original problem (this is what constitutes the ‘explanatory dilemma’ after all). V.2 Core-Self as a ‘Psychological Domain’ of BN. Introducing the ‘Nested Concept Hypothesis’ My suggestion in order to avoid the hindrance described above (what I have dubbed, for want of a better term, the ‘Explanatory Dilemma’) consists of proposing the following hypothesis: Merker’s and Gallagher’s proposals correspond to two different concepts of ‘Core-Self’ but

Empirical and conceptual clarifications

191

related to each other in a ‘nested ‘way that corresponds to a single Psychological Domain. This hypothesis, if confirmed, would solve the ‘explanatory dilemma’ since it would grant us, at the same time, two different concepts of ‘Core-Self’ but inter-related in a way which is capable of providing the desired explanatory unification. Having stated the hypothesis, we are faced now with the challenge of testing it. Following what was established in section II, there are several characteristic properties and constraints that two BN concepts have to satisfy in order to be related in a single nested way. These several principles and constraints are encapsulated in the three structural conditions stated at the end of section II. As a start, we have to assume that the two concepts of Core-Self, corresponding to each of the two proposals, are related in such a way that one of the concepts (call it ‘Core-Self1’ provisionally) is nested in the other (‘Core-Self2’). What we do not know is the following: which proposal (Merker’s or Gallagher’s) corresponds to which concept (‘CoreSelf1’ and ‘Core-Self2’). As we have seen in section II, there are two basic conditions two BN concepts must satisfy in order to be considered nested in each other: the General Intensional Condition (GIC) and the General Extensional Condition (GEC). For several technical reasons (not addressed here), an extensional difference is easier to determine and quantify than an intensional one. Therefore, we should explore the required extensional difference between the two proposals first. Let us recall what was stated in the general extensional condition (GEC) regarding BN Nested Concepts: GEC: if a concept C1 is nested in concept C2, then the class of objects of the domain satisfying ‘x is C2’ is a proper subset of the class of objects of the domain satisfying ‘x is C1’. In this respect, it is not very difficult to establish and justify the extensional difference. Although Gallagher is not very explicit about the real intended extensional scope of his proposal, nevertheless, taking into consideration the fact that the neural implementation implies a relatively developed neo-cortex (including pre-frontal areas) we could, reasonably, establish the scope of his proposal as varying from: only humans at minimum, or: all mammals at maximum. On the other hand, Merker is very explicit concerning his proposed domain of ‘application’: all

192 

João Fonseca

vertebrates. As he says in relation to the neural structures implementing his proposal of core-Self: “[t]hese structures together are proposed to form the essential scaffolding of the mechanism of consciousness in vertebrates.” (2005, p. 106) Merker justifies this statement by relying on evolutionary structural conservation (homologies) across taxa. The upper brainstem complex comprising Merker’s proposal, like other neural structures, belongs to a common ‘brain plan’ homologous across vertebrates (2005, p. 68). Things being so, the set corresponding to the intended scope within domain D of Merker’s proposal is wider than Gallagher’s; even assuming the maximal perspective, presumably the set of all mammals is a proper subset of all vertebrates. Let us recall the evolutionary criterion for subset relation among animal classes: a class A of living creatures is a subset of another class B if the evolutionary clade corresponding to class B is Basal in relation to the evolutionary clade corresponding to class A. In accordance with this rule, the clade corresponding to the extensional domain of CoreSelf1 should correspond to a clade basal in relation to the clade corresponding to the extensional domain of Core-Self2. Things being so, we can establish that Core-Self1 corresponds to Merker’s proposal whereas Core-Self2 corresponds to Gallagher’s, given that Vertebrates correspond to a basal clade in relation to Mammals. What is more, therefore, Merker’s concept/proposal is (in principle) nested in Gallagher’s. For taxonomic and explanatory reasons we have to classify each proposal/concept by ‘naming’ them. Taking into consideration that Merker’s proposal (‘Core-Self1’) appears to address some more fundamental aspects of Core-Self ranging over a wide class of vertebrate species, I suggest naming his proposal as ‘Core-Self Simpliciter’. On the other hand, since we have seen that Gallagher’s proposal (Core-Self2), based on the implementation of a forward-model, sustains a series of more elaborate cognitive abilities, my personal suggestion is to call it ‘Cognitive Core-Self’. So, for now, the nested concept hypothesis gains some strength since ‘Core-self Simpliciter’ and ‘Cognitive Core-Self’ satisfy the GEC condition for being considered related in a ‘nested’ way. It remains to be shown that the Three Structural Conditions are also satisfied. Starting with the Necessary Structural Condition, it should be recalled that it stated that the M (in this case, of ‘Core-Self Simpliciter’)

Empirical and conceptual clarifications

193

comprises necessary structural (neural-mechanistic) conditions for the set of all the behaviours of the related Psychological Domain (in this case of the Core-Self domain). In the present context this means that, if ‘Core-Self Simpliciter’ is nested in ‘Cognitive Core-Self ’, the neural structures corresponding to the former necessarily underlie the later. In fact, we were given just a glimpse of this situation when we were searching for a cognitive difference between Merker’s and Gallagher’s proposals in sub-section IV.1 above. There, a quote from Merker’s article form Behavioural and Brain Sciences, announced that his views were cognitively different and simpler than certain cognitive competencies such as ‘imagery’ and ‘off-line planning’ supposedly sustained by a more complex Forward-model. But, in addition to the cognitive/functional difference, Merker stated that: “[s]uch capacities are derivative ones, dependent upon additional neural structures whose operations presuppose those described here.” (2007, p. 81, emphasis added) According to the empirical consequence associated with this condition, if the neural structure corresponding to the concept which is nested in another is disrupted, all the behaviours within the psychological domain are completely disabled. There is disparate empirical evidence that an analogous situation is true in the present context of ‘Core-Self’. A particular line of evidence is important here: the situation where the upper-brainstem (the neural structure corresponding to ‘Core-Self Simpliciter’) is physically impaired. An important contribution favouring this line of evidence is given by the neuroscientists Jaak Panksepp and Georg Northoff when presenting a proposal of (Core)Self very similar to the one suggested by Merker. Panksepp and Northoff (2009) depart from the seminal experiments conducted by Bailey and Davies in the early nineteen forties to stress how fundamental the role of some structures of the upper-brainstem complex is, namely the PAG, on sustaining any purposive or intentional behaviour. This conclusion is driven by the results of Bailey and Davies showing that: complete damage of the lowest midbrain integrator for emotionality, namely the PAG, compromises all world directed action tendencies in animals. Such animals lose all intentionality, and seem to be barely conscious; this highlights how severely compromised are both agency and ownership of experience when the most fundamental emotional-affective SELF-generating functions of the brain are destroyed. (2009, p. 204, emphasis added)

194 

João Fonseca

Merker arrives at the same conclusion, although through different empirical sources: Cases of human collicular damage are rare, but in one such case whose damage extended into the midbrain tegmentum (Denny-Brown, 1962) the symptoms were severe, consisting of profound apathy with “no evidence of recognition of people or any event in her surroundings.” This points to the potential importance of the extended midbrain damage in determining outcomes – a point also underscored by the findings on macaques in the same report – and to the significance of spontaneous, self-initiated behaviour and reactivity in this regard. (2005, p. 101)

Generally, as Parvizi and Damasio point out: “damage to the upperbrainstem is a known cause of coma and persistent vegetative state” (2000, p. 1351, emphasis added). What can be concluded from these several sources is that a selective impairment in the upper-brainstem complex (sometimes at specific structures such as the PAG or the Superior Culliculus), and maintaining the cortical apparatus intact, seem to disable any kind of purposive behaviour (the behavioural target of ‘Core-Self’. Panksepp and Northoff even mention explicitly that, in particular, the senses of agency and ownership, the main phenomenological characteristics of ‘Cognitive Core-self’, are disabled when those more fundamental structures comprising ‘Core-Self Simpliciter’ are physically impaired. All these data and results point towards a very probable vindication of the Necessary Structural Condition regarding the supposed nested relation between the concepts ‘Core-Self Simpliciter’ and ‘Cognitive Core-Self’. In accordance with this, it has been shown that the set corresponding to all behaviours of the psychological domain Core-Self (i.e., all purposive behaviours) are compromised if the set of all the neural mechanisms corresponding to ‘Core-Self Simpliciter’ (i.e., the upper brainstem) is physically disabled. Regarding the Sufficient Structural Condition, in the present context its empirical claim basically states that if the neural structures corresponding to ‘Cognitive Core-Self’ (namely the cortical areas implementing a motor forward model) are physically inhibited, then just a subset of the set of all behaviours of the psychological domain Core-Self (the elements of the set of behaviours from the set difference AB\AB* of CoreSelf) will be compromised. It is time now to explore the line of evidence resulting from damage to the cortex while sparing the upper-brainstem.

Empirical and conceptual clarifications

195

We could start with Gallagher’s own point regarding situations where the cortical implementation of a forward-model is impaired. Following Gallagher, in these cases what is observed is that the phenomenological ‘sense of agency’ is disabled and, as a consequence, the subjects report that it is as if ‘alien’ or ‘outside agents’ were the causes of their own voluntary actions. This kind of phenomenological report is typical of subjects diagnosed with schizophrenia. Nevertheless, in these cases, what we are describing is not a complete loss of the self but a sort of a truncated or distorted experience of it. There is a self which is disabled in some of its properties, not a complete displacement (this is well attested in some verbal reports by schizophrenic patients: “I’m not causing my movements, someone else is”). Even more impressive are the various empirical results showing a total functional independence of the upper-brainstem from the neo-cortex. In this respect, Merker notes that: All of the behaviors just mentioned have also been exhibited by experimental animals after their cerebral cortex has been removed surgically, either in adulthood or neonatally. Best studied in this regard are rodents (Wishaw 1990; Woods 1964). After recovery, decorticate rats show no gross abnormalities in behavior that would allow a casual observer to identify them as impaired in an ordinary captive housing situation, although an experienced observer would be able to do so on the basis of cues in posture, movement, and appearance (Whishaw 1990, on which what follows relies, supplemented by additional sources as indicated). They stand, rear, climb, hang from bars, and sleep with normal postures (Vanderwolf et al. 1978). They groom, play (Panksepp et al. 1994; Pellis et al. 1992), swim, eat, and defend themselves (Vanderwolf et al. 1978) in ways that differ in some details from those of intact animals, but not in outline. Either sex is capable of mating successfully when paired with normal cage mates (Carter et al. 1982; Whishaw & Kolb1985), though some behavioral components of normal mating are missing and some are abnormally executed. Neonatally decorticated rats as adults show the essentials of maternal behavior, which, though deficient in some respects, allows them to raise pups to maturity. Some, but not all, aspects of skilled movements survive decortication (Whishaw & Kolb 1988), and decorticate rats perform as readily as controls on a number of learning tests (Oakley 1983). Much of what is observed in rats (including mating and maternal behavior) is also true of cats with cortical removal in infancy. They move purposefully, orient themselves to their surroundings by vision and touch (as do the rodents), and are capable of solving a visual discrimination task in a T-maze. (2007, p. 74)

The above extensive citation (justified by the profusion of important examples and reference sources) suggests that only a subset of all

196 

João Fonseca

purposive behaviours (i.e., a subset of the set of behaviours corresponding to the psychological domain Core-Self) are disabled by the physical impairment of the neural structures corresponding to ‘Cognitive Core-Self ’ (the cortical areas implementing a motor forward model). Just like the necessary structural condition, the sufficient structural condition seems to be vindicated in this particular case. Finally, the Unitary Mechanism Condition, applied to the present context, states that there should be a neural mechanism neither stated in ‘CoreSelf Simpliciter’ (upper brainstem complex) nor in ‘Cognitive Core-Self’ (cortical implementation of a Forward Model) that causally links both. We have just described above how the upper-brainstem complex (the neural mechanism corresponding to ‘Core-Self Simpliciter’) seems to consist a structural necessary condition for more specific and finegrained concepts/models of ‘Core-Self’ (namely ‘Cognitive CoreSelf’). It remains, however, to address the question concerning How the neural mechanisms of ‘Core Self Simpliciter’ and ‘Cognitive Core-Self ’ relate to each other in a physical way. Panksepp and Northoff (2009), although departing from a slightly different perspective from the one adopted here, address this point directly. They do this by introducing additional cortical structures into the picture, namely: the cortical midline structures (such as the posterior cingulated cortex, the dorsomedial prefrontal cortex, or the orbitomedial prefrontal cortex). Regarding Gallagher’s proposal of the implementation of a motor forward-model in the motor, pre-motor and prefrontal areas of the cortex, they note (based on several empirical findings) that these cortical areas “are closely connected with midline regions” (2009, p. 204). According to Panksepp and Northoff, the forward-model cortical areas are physically connected with the sub-cortical upper-brainstem region via these midline structures (2009, p. 204). Therefore, according to Panksepp and Northoff’s suggestion, the neural mechanism M* described in the model/concept ‘Cognitive Core-Self’ would consist of the ‘filling’ of the schematic M deployed in the model for ‘Core-Self Simpliciter’ (upper-brainstem complex) via the inclusion of the information stated in the mechanism schema M** (projections to certain cortical midline structures plus its connections to the areas of motor and pre-motor cortex implementing the forwardmodel). This means that the Unitary Mechanism Condition is satisfied.

Empirical and conceptual clarifications

197

At the beginning of this section I suggested that the concepts of Core-Self involved (Merker’s and Gallagher’s) could relate in a nested way and, therefore, define a single Psychological Domain in order to overcome the Explanatory Dilemma described in the previous section. In order to vindicate this relation the two concepts had to conjunctively satisfy the three structural conditions presented earlier. At this point, therefore, the two concepts can be said to be related in a nested way (Core-Self Simpliciter is nested in Cognitive Core-Self) since they satisfy the three structural conditions. If they are indeed related in a nested way, they define a single Psychological Domain, then we could simply classify this as the Core-Self domain. It remains to show how this nested relation between the two concepts is capable of solving the so-called Explanatory Dilemma. V.3  Explanatory Unification and the Overcoming of the Dilemma Before directly addressing the solution of the explanatory dilemma, it is important to consider the important claim concerning the promised ‘explanatory unification’ of the ‘Nested’ concept of Core-Self, since, as stated above, the satisfaction of this claim is instrumental in the overcoming of the dilemma. Here I will just scratch the surface of this topic and advance some suggestions for further empirical research. In the previous section it was shown that ‘Core-Self Simpliciter’ and ‘Cognitive Core-Self’ satisfy the three structural conditions in order to be able to be considered as nested concepts. At the same time, the notion of a single Psychological Domain covering Core-Self was also vindicated. This means that a certain level of unification was gained, in particular a structural/neuronal explanatory unification. This unification, among other things, provides us with a clearer view regarding the neural structure of the notion of Core-Self as a whole (i.e., as a single Psychological Domain). In parallel to this unification at the structural/neuronal level, one at the functional/cognitive level is also suggested. The very functional characterization of both notions is in need of further clarification along with their mutual dependence (as predicted by the ‘nested hypothesis’). We have already shown in section IV.1 that Merker’s proposal (renamed

198 

João Fonseca

‘Core-Self Simpliciter’) and Gallagher’s (‘Cognitive Core-Self ’) differ in some of their basic cognitive properties. In particular, it was established that whereas Gallagher’s proposal, based on the implementation of a forward-model, is capable of more complex cognitive features such as imagery, Merker’s is simpler and lacks that specific kind of sophistication. What remains is to understand these differences in more detail within a context that frames both. The above three structural conditions are not only useful in clarifying the neural mechanistic relation between the two concepts of Core-Self. A more clear and fine-grained determination of what both concepts are really about (i.e., what precisely their cognitive content is) also comes about as an important consequence brought about by the nested conceptual relation between the two. For instance, by applying the Sufficient Structural Condition we can state with greater precision which behaviours (B*) are supposed to be explained by ‘Core-Self Simpliciter’ () and ‘Cognitive Core-Self’ () respectively. According to the Sufficient Structural Condition there is a difference between B and B* (formally captured by the set difference AB\AB* of the Psychological Domain Core-Self). The question is: what exactly is this difference? If the target behaviour of the initially proposed unique model of ‘Core-Self’ corresponded to ‘purposive behaviour’, then the B of ‘Core-Self Simpliciter’ is supposed to correspond to that same target of ‘purposive behaviour’ generally (without further specification). On the other hand, ‘Cognitive Core-Self’ is expected to causally explain only a particular sub-set (B*) of the more inclusive set of general/ non-specified purposive behaviour (B). It is hard to state firmly what specific behaviour ‘Cognitive Core-Self’ is supposed to explain since Gallagher is far from clear regarding this issue. Nevertheless, recasting the idea of ‘explanatory unification’, some progress is legitimately to be expected. In fact, Panksepp and Northoff once again provide some interesting clues within the context of explanatory unification. In the following quotation, they start by addressing the functional difference and dependence and end by extrapolating to the behavioural level: Functionally, higher cortical regions might represent the functions that are primarily represented subcortically in a more detailed, specific and refined way. [...] Though both complex sensory and motor functions are already represented in subcortical regions like the PAG, basal ganglia, mesencephalic locomotor system,

Empirical and conceptual clarifications

199

those primitive sensory-motor functions are again represented in cortical regions, e.g., somatosensory and motor cortices. This may allow behavioural functions to be elaborated in more detail, with specific types of flexibility, especially with much greater regulation of and inhibition over instinctual outputs, than it is possible on the subcortical level. (2009, p. 201, emphasis added)

So, an obvious dependence relation between the behavioural/cognitive profiles of ‘Core-Self Simpliciter’ and ‘Cognitive Core-Self ’ is again suggested and vindicated. At the same time, that relation is inferred and clarified by linking the supposed behavioural differences to both cognitive and neuronal differences, interactions and dependencies. A unified research programme into motor functions, as suggested by Panksepp and Northoff, could clarify in greater detail, which specific target behaviours (B and B*) correspond to ‘Core-Self Simpliciter’ and ‘Cognitive Core-Self’. This clarification, within a unified framework, would also reformulate both the functional profiles (f and f*) and the neural mechanisms (M and M*) of both concepts. These considerations show, at the very least, how explanatorily fruitful it is to consider ‘Core-Self ’ as a single Psychological Domain where the concept ‘Core-Self Simpliciter’ is considered to be nested within ‘Cognitive Core-Self ’. Given the above considerations we can address, and easily solve, the initial ‘Explanatory Dilemma’. Recall the initial formulation of the ‘dilemma’: assuming both proposals for the neural implementation of ‘Core-Self’ (Merker’s and Gallagher’s) there are two distinct and incompatible possibilities: 1. The two proposals choose the same concept and Multiple Realization (MR) is vindicated, or 2. The two proposals pick different and unrelated concepts. The ‘explanatory dilemma’ consisted in the fact that, independently of which possibility one adopts, the end result would consist of an explanatory fragmentation over a desired unification. The nested concept solution for ‘Core-Self’ avoids this dilemma by, at the same time, overcoming MR and radical fragmentation implicit in solution 2. The ‘nested solution’ overcomes MR since it claims that, contrary to the assumption in the argument for MR, the two proposals select two different concepts (with different grains of explanation and different extensions in the domain D). On the other hand, it avoids radical explanatory fragmentation since the two concepts are inter-related and partially overlap each other both intensionally and extensionally (they are related in the sense that they belong to the same Psychological Domain

200 

João Fonseca

Core-Self). This explanatory unification between the two concepts was clearly shown in the preceding paragraphs of this section. Therefore, by adopting the nested concept solution, we overcome the explanatory dilemma and achieve explanatory unification and clarification regarding the notion of ‘Core-Self’ by gathering the two initial proposals under a common (and empirically vindicated) explanatory framework.

Conclusion At the beginning of this paper we showed the methodological fragmentation brought about by the current profusion of scientific approaches concerning ‘Self’/‘Core-Self’ (neuroscience, psychology, philosophy, among others). The main concern was that a methodological fragmentation could lead to (and sometimes does) a related conceptual fragmentation. As Gallagher and Zahavi state: “One problem to be posed in this light is whether different characterizations of self signify diverse aspects of a unitary concept of selfhood, or whether they pick different and unrelated concepts.” (2008, p. 198). I tried to address this question directly by focusing on two different proposals: Merker’s evolutionary perspective and Gallagher’s phenomenological one. By deploying my proposed model-theoretic framework for theoretical concepts of behavioural neuroscience, I tried to uncover some of the underlying principles sustaining both proposals. By using the instrumental and fundamental notion of ‘Nested Concept’ I tried to show how the two proposals could relate to each other. Finally, I suggested a possible answer to Gallagher and Zahavi’s anxiety: Merker’s and Gallagher’s proposals point, simultaneously and without contradiction, towards a unitary concept of CoreSelf and to two different concepts. This ‘nested solution’ proved to be both empirically and conceptual vindicated, and explanatorily fruitful (even suggesting avenues for further empirical research). I hope I have shown that, for explanatory and conceptual reasons, it is useful to address the notion of ‘Core-Self’ as a theoretical concept of behavioural neuroscience. By assuming such an approach we gain conceptual

Empirical and conceptual clarifications

201

and taxonomic clarification, explanatory richness, and bridges, both conceptual and empirical, between different scientific disciplines and practices.

References Bard, P. & Rioch, D. M. (1937). A study of four cats deprived of neocortex and additional portions of the forebrain. Bulletin of the Johns Hopkins Hospital, 60, 73–148. Bickle, J. (2003). Philosophy and Neuroscience. Dordrecht: Kluwer Academic Publishers. Bickle, J. (2006). Reducing mind to molecular pathways: explicating the reductionism implicit in current cellular and molecular neuroscience. Synthese, 151, 411–434. Bjursten, L.-M., Norrsell, K. & Norrsell, U. (1976). Behavioral repertory of cats without cerebral cortex from infancy. Experimental Brain Research, 25, 115–30. Bourtcholadze, R., Frenguelli, F, Blendy, J., Cioffi, D., Shultz, G., & Silva, A. (1994). Deficient long-term memory in mice with a targeted mutation of the Camp-responsive element binding protein. Cell, 79, 59–68. Carter, C. S., Witt, D. M., Kolb, B. & Whishaw, I. Q. (1982). Neonatal decortication and adult female sexual behaviour. Physiology and Behavior, 29, 763–66. Churchland, P. S. (2002). Brain Wise, Cambridge MA: MIT Press. Craver, C. (2007). Explaining the Brain. Oxford: Oxford University Press. Damasio, A. (1999). The Feeling of What Happens: Body and Emotion in the Making of Consciousness. New York: Harcourt Brace. Darden, L. (2002). Strategies for discovering mechanisms: schema instantiation, modular subassembly, forward/backward chaining. Philosophy of Science, 69, S354–S365. Dennett, D. (1991). Consciousness Explained, Boston MA: Back Bay Books. Denny-Brown, D. (1962). The midbrain and motor integration. Proceedings of the Royal Society.

202 

João Fonseca

Gallagher, S. (2000). Philosophical conceptions of the self: Implications for cognitive science. Trends in Cognitive Science, 4, 14–21. Gallagher, S. & Zahavi, D. (2008). The Phenomenological Mind: An Introduction to Philosophy of Mind and Cognitive Science. New York: Routledge. Giere, R. (1988). Explaining Science. Chicago IL: University of Chicago Press. Giere, R. (1999). Science Without Laws. Chicago IL: University of Chicago Press. Grush, R. (2004). The emulation theory of representation: motor control, imagery, and perception. Behavioral and Brain Sciences, 27, 377–442. Kogan, JH., Frankland, PW. & Silva, AJ (2000). Long-term memory underlying hippocampus-dependent social recognition in mice. Hippocampus, 10, 47–56. Merker, B. (2007). Consciousness without a cerebral cortex: a challenge for neuroscience and medicine. Behavioral and Brain Sciences, 30(1), 63–81. Merker, B. (2005). The liabilities of mobility: A selection pressure for the transition to cortex in animal evolution. Consciousness and Cognition, 14, 89–114. Metzinger, T. (2003). Being No One: The Self-Model Theory of Subjectivity. Cambridge MA: MIT Press. Metzinger, T. (2009). The Ego Tunnel: The Science of the Mind and the Myth of the Self. New York: Basic Books. Northoff, G. & Bermpohl, F. (2004). Cortical midline structures and the self. Trends in Cognitive Sciences, 8, 102–107. Oakley, D. A. (1983). The varieties of memory: A phylogenetic approach. In Mayes (ed.) Memory in Animals and Humans, 20–82. Van Nostrand Reinhold. Panksepp, J., Normansell, L., Cox, J. F. & Siviy, S. M. (1994). Effects of neonatal decortication on the social play of juvenile rats. Physiology and Behavior, 56, 429–43. Pankseep, J. & Northoff, G. (2009). The trans-species core SELF: The emergence of active cultural and neuro-ecological agents through self-related processing within subcortical-cortical midline networks. Consciousness and Cognition, 18, 193–215.

Empirical and conceptual clarifications

203

Parvizi, J. & Damasio, A. (2001). Consciousness and the brainstem. Cognition, 79, 135–159. Pellis, S. M., Pellis, V. C. & Whishaw, I. Q. (1992). The role of the cortex in play fighting by rats: Developmental and evolutionary implications. Brain, Behavior and Evolution, 39, 270–84. Philipona, J. K. & O’Regan, J. P. (2003). Is there something out there? Inferring space from sensorimotor dependencies. Neural Computation, Vol. 15, No. 9, 2029–2049. Praetorius, N. (2009). The phenomenological underpinning of the notion of a minimal core self: a psychological perspective. Conscious Cognition, 18(1), 325–38. Squire, L. & Kandel, E. R. (1999). Memory: From Mind to Molecules, New York: Scientific American Library. Staddon, J. & Niv, Y. (2008). Operant Conditioning, Scholarpedia, 3(9), 2318. . Vanderwolf, C. H., Kolb, B. & Cooley, R. K. (1978). Behavior of the rat after removal of the neocortex and hippocampal formation. Journal of Comparative and Physiological Psychology, 92, 156–75. Whishaw, I. Q. (1990). The decorticate rat. In B. Kolb & R. C. Tees (eds.), The Cerebral Cortex of the Rat, 239–67. Cambridge MA: MIT Press. Whishaw, I. Q. & Kolb, B. (1985). The mating movements of male decorticate rats: Evidence for subcortically generated movements by the male but regulation of approaches by the female. Behavioural Brain Research, 17, 171–91. Wolpert, DM., Mial, RC. & Kawato, M (1998). Internal Models in the Cerebellum. Trends in Cognitive Science, vol. 2. Woods, J. W. (1964). Behavior of chronic decerebrate rats. Journal of Neurophysiology, 27, 635–44. Zahavi, D (2009). Is the self a social construct? Inquiry, 52(6), 551–573.

Part IV Ontology and Taxonomy

Core Self and the Problem of the Self Jorge Gonçalves1

The concepts of consciousness and self have been central in contemporary philosophy of the mind. Inevitably, this has led to the recuperation of a few conceptions from classical Phenomenology, starting with Husserl. This is the case with the concept of “pre-reflective self-consciousness”. The approach of these philosophers (Zahavi, 2007; Gallagher, 2007; Parnas, 2010) is not existential, but what could be called “biological” in the sense that they considered consciousness and self as natural phenomena, explained scientifically. One of the problems that these philosophers intended to resolve is the renowned problem of the self that was initially formulated by David Hume and more recently by Metzinger, among others. In this lecture, I intend to argue, using the facts regarding the origin of the self in the early stages of life as a foundation for the argument, that the concept of pre-reflective self-consciousness does not resolve this problem. In spite of the facts not being conclusive, I will affirm that there is good reason to reject the idea that the entire form of phenomenal consciousness assumes a feeling, no matter how little, of self. According to these authors, all consciousness is simultaneously self-consciousness. For example, when I look at a landscape, in the moment I am looking there is always a feeling that it is I who is looking. This feeling could be reflective, when I carry out an introspection of my mental states with the view that I am studying consciousness. However, for this feeling of self to exist, it is not necessary to be in a thematic reflective state. Even without it being reflective, consciousness always implies a sense of self. We could succinctly divide phenomenal consciousness into two dimensions. One of them is known as “what is it to be like”, a term 1

Fellowship funded by the Portuguese Foundation for Science and Technology.

208 

Jorge Gonçalves

that was popularized which comes from a celebrated article by Thomas Nagel. We can roughly say that it describes the states that imply to feel.2 The other dimension is of the first person. The phenomenological states always imply, according to the referred authors, a subject who is minimally conscious. There are no phenomenological states that can be felt being disconnected from a feeling of the self. An example referred to by Sartre, is of a person who is concentrated on reading. She is not thematically conscious that she has a self, because she is absorbed in the object she is reading. However, if she is asked what she is doing, she immediately responds that she is reading. This fact demonstrates, according to the defenders of phenomenal consciousness, that she was always conscious of herself reading, but in a pre-reflective way, the proof of this being in the immediacy with which she responds to the question. This minimal feeling of self has been designated as the core-self or minimal self. In Gallagher’s words, “Minimal [core] self: phenomenologically, that is, in terms of how one experiences it, a consciousness of oneself as an immediate subject of experience, unextended in time. The minimal self almost certainly depends on brain processes and an ecologically embedded body, but one does not have to know or be aware of this to have an experience that still counts as a self-experience.” (Gallagher, 2000). This core self is always elusive, it never really becomes an object, and is never thematic. The Buddhists use metaphors to describe it: it is like a flame that illuminates objects, but never illuminates itself. The authors maintain it is impossible to present the self before ourselves as an object, that could be observed. However, they also claim that the I/self is not a pure empty place, as Kant conceived it to be. The core self is endowed with a certain mass of a “phenomenological material”. This “phenomenological material” is invisible, meaning that it is not observable using the methods of the third person. One of the philosophical problems the authors wish to see resolved with the notions of self-reflective pre-consciousness and core-self is, as 2

For example, “What is it to be like on the road and getting a new bed almost every night?” 

Core Self and the Problem of the Self

209

we referred to earlier, the so-called problem of the self, as was famously presented by David Hume. But self or person is not any one impression, but that to which our several impressions and ideas are supposed to have a reference. If any impression gives rise to the idea of self, that impression must continue invariably the same, through the whole course of our lives; since self is supposed to exist after that manner. But there is no impression constant and invariable. Pain and pleasure, grief and joy, passions and sensations succeed each other, and never all exist at the same time. It cannot, therefore, be from any of these impressions, or from any other, that the idea of self is derived; and consequently there is no such idea.3

When I show my own self to myself, I represent it as if it were an entity with a given unity and continuance in time. This representation raises the problem of knowing what kind of entity it is. For many authors the self does not exist; it is nothing more than an illusion. Metzinger (2009) recently elaborated a theory of the self where he argues that we are an illusion, an illusion for no one, in fact, the “we” of the previous phrase does not exist. There is no entity with these described characteristics. “We” are no one. Despite these statements seeming counter-intuitive, Metzinger defends them in a convincing way combined with philosophical perspectives and scientific data. He does not argue that the feeling of the self does not exist. He considers it as a phenomenological datum. What he defends is that phenomenology itself is an illusion, just like the other illusions he explains with his model of the mind. It the case of the rubber-hand illusion (Metzinger, 2009). The subject is put in a position such that his hand is hidden and he looks at an artificial hand as though it were his real one. His real hand is touched at the same time as the artificial one. In a few seconds, this produces the illusion for the subject that his real hand is the artificial one. Phenomenologically the subject feels that his hand is the artificial one, but this is an illusion because he has a real hand of his own flesh and blood. In the same way, the experience that a self exists, an independent entity capable of existing on its own, having an essence, or rather a set of invariable intrinsic properties, and is provided with individuality, is an illusion.

3 In A Treatise of Human Nature, by David Hume (1739), Book I, Part 4.

210 

Jorge Gonçalves

The difference is that in the case of the partial illusions we are conscious, while in the case of the self we are, ourselves, the illusion, and in this sense we do not have the possibility of perceiving that we are illusions. Metzinger explains this situation by stating that the self has a transparent property, in the phenomenal sense. The self understands itself as a reality because it does not know the processes from which it originates; it does not know the reality where it is an illusion itself. In order to understand this better, we can think about another case of transparency: visual perception. When I look at a pen, I think I am looking directly at an object just as it is in the objective exterior world. I do not comprehend this object as a result of representational and neuronal mechanisms. If I do not give in to a reflective process, I am in the position of naïve realism or the belief that reality is just as it appears to me. According to Metzinger, organisms are equipped with representational mechanisms that result from the course of biological evolution due to the diverse advantages that they bring. Some of these representational mechanisms represent the organism to itself. Metzinger develops an entire complex theory, from which here I will present only the conclusions. What interests me here refers to what the phenomenological self, what we feel, is the momentary content of an auto-representational mechanism of the human body. This auto-representation has evolutionary advantages and is not, because of this, an epiphenomenon. To be conscious of our own self, we are completely unaware of the representational mechanisms from which we originate. We are transparent in this sense: we comprehend ourselves directly without paying attention to the mediatory mechanisms. Metzinger draws out the following results from this biological situation: we are no one, we think that we exist as a reality but in truth we do not exist because we are like shadows in Plato’s cave. Moreover, Metzinger himself uses this thousand-year-old image4. Neurology is reality, the self is a shadow of this reality that we take to be reality. The ontological status of the self is that of the illusion. From the scientific point of view, the self is a concept to be eliminated because it does not notice the reality it is trying to describe. The experience that I have of the self is not the real thing, and thus I live in a state of auto-illusion. 4

Metzinger, T., Being No One. The Self-Model Theory of Subjectivity, MIT Press, Cambridge, MA, 2003 p. 547.

Core Self and the Problem of the Self

211

Metzinger’s ontological conclusion that the self does not exist has been criticized in the field of the philosophy of mind by various authors5 who assert the Phenomenology tradition of Husserl and other philosophers, his followers such as Heidegger, Sartre, Merleau-Ponty, and Michel Henry. This new Phenomenology has a naturalist character, different from that of the aforementioned authors whose problems were existential more than anything else. The basic critical idea is that Metzinger, stating that the self does not exist, has a specific conception of self. Despite Metzinger reporting the most recent scientific discoveries in neuroscience, he continues to hold a traditional conception of the self, which is conceived of as an image of material things, as being something equipped with spatiality and permanence. Another conception defended by these authors is that of calling the self something more basic and not differentiating it from the flow in phenomenal consciousness. In this conception, the self is reduced to its minimal level. According to Gallagher’s definition, “phenomenologically, that is, in terms of how one experiences it, a consciousness of oneself as an immediate subject of experience, unextended in time.”6 The concept of self as we just defined it is broadened in time. However, these authors maintain that if we take away all of the non-essential characteristics of the self, we will still remain with a nuclear self, a minimal self. This pure consciousness of the self exists only in the so-called “specious present”, where there are controversies regarding its real duration. It deals with a minimum point in time when we are conscious. The core self is connected to the flow of consciousness as it is an integral part of it. Whatever experience – pain, for example, the perception of a color, a sudden rage, a brilliant idea, is always felt by a subject as something happening to him and not to another. The experience is always given in a certain way and felt as a happening “to me”, as if I were its possessor. The phenomenality of the experience and its first person character are not separable. You cannot speak of an experience in an impersonal way. The experience always happens, by definition, to somebody.

5 6

Zahavi, 2005; Zahavi, D.,  Grünbaum, T. &  Parnas, J. 2004; Gallagher, 2000, 2006. Gallagher, S.  (2000). Philosophical Conceptions of the Self. Trends in Cognitive Sciences, 4(1), pp. 14–21, p. 15.

212 

Jorge Gonçalves

In this way, when we speak of the self it is unnecessary to conceive it as a separated entity. The experience itself has a personal character. The self is not an essential center that groups experiences around itself. It is implicitly present in any experience. This idea that the self is implicitly present is very important for the phenomenologists’ argument. In order for the self to exist, a conscious auto-reflection is unnecessary. It is unnecessary that I am thinking about myself in order for the feeling of mine-ness to arise. One of the ways of verifying that this pre-reflective consciousness of self is always present consists of, for example, asking a person who is absorbed in reading what he or she is doing. Without hesitation he will respond, “I’m reading”. In the moment that he is absorbed in reading, the subject is not conscious of himself explicitly. He is completely absorbed in that which he is reading. Nevertheless, why he responds without hesitation to a question shows (according to the phenomenologists) that his consciousness of himself was always tacitly present. This consciousness of self is not purely empty; there is a specific depth that constitutes the core self. In this way, the self is always implicitly present in all of the stream of consciousness and exists not just in a few privileged moments of reflection. What implications does this have for the ontological thesis that the self is an illusion? The self will only be an illusion when we see it in analogy with physical objects, with the objects of the world. If we allow ourselves to think of the self in this way and we see it as an auto-referential process, the problem of knowing whether the self exists or not becomes different. In this case, questioning the existence of a core self will be like questioning the existence of consciousness and even of the world itself, as the consciousness in Phenomenology is not opposed to the objective world. Consciousness is where the world is given. The concept of the core self does not stop the phenomenologists from having abandoned the concept of the broader or narrative self. Merely that the core self is assumed in these most diverse selves and that once it is admitted that the core self is part of the stream of consciousness, it does not make more sense to say that “we are nobody” as Metzinger and others claim, unless they wish to deny the existence of consciousness itself. The phenomenologists thus claim to reduce the problem of the existence of the self to the problem of the existence of consciousness. If there exists a mind-body problem, it includes the problem of the self

Core Self and the Problem of the Self

213

because once consciousness is explained there is no longer an additional problem of the self. Of course empirically the formation of the broader self from the core self remains to be explained, but this would not be a hard problem. The so-called “hard problem” is a problem of understanding how the connection between materia and consciousness is possible. It is a problem that many consider to be insoluble. The problem of the relation of the nuclear self with the narrative self is not a hard problem; it is an easy problem because there isn’t a difference in nature, there is no explanatory gap. Pre-reflective self-consciousness is not a consciousness of a preexisting self, but a consciousness that contains a minimal sense of self, a core self, as an integral part within itself. If there is experience, there is a core self. This assumes that any being to which we can attribute phenomenal and sensorial states will always have a minimum sense of self. Now, an objection to this train of thought is that the self has not existed since the beginning of life, even though there are experiences, such as for example, sensations of pleasure and pain. If babies have experiences but no feeling of self, the thesis of the self as an integral part of consciousness is questioned. There can be consciousness without a feeling of self, therefore they are two different things although intimately related. This question is difficult to investigate because, remember, what interests us here is the phenomenology of the child and not only his observable behavior. We can infer the former from the latter, however this generates different interpretations and it is difficult to produce evidence for the theory to be proven. The most traditional researchers (Henri Wallon7, Merleau-Ponty8, Lacan9) have defended the thesis that the first months of human existence are selfless. Nevertheless, the most recent research has claimed to demonstrate that that is not the case. Gallagher and Meltzoff give an account of this research10. Both the traditional researchers and the most recent ones agree regarding the proprioceptive system of the body schema as responsible for the feeling of self and of others, during the first months of life. Gallagher and Meltzoff confirm meanwhile that there has been a confusion regarding 7 8 9 10

Wallon, H. (1934). Mearleau-Ponty M. (1990). Lacan, J. (1966). Gallagher, N. & Meltzoff, A. (1996).

214 

Jorge Gonçalves

the body schema and the body image. We can define the body schema as a system that is unconscious of its processes (motor capacities, abilities, habits) and that constantly regulates posture and movement. The body image is a system that can be conscious of perceptions, attitudes, and beliefs directed at the body itself. The body schema is connected to doing, moving, while the body image is the perception, analysis or control of these movements. The ambiguous use of these two terms complicates the understanding of the theory of traditional authors, because they are not conscious of the fact that they are speaking about different things. Everyone accepts, in a pacificatory fashion, that a child has an innate body schema; the same does not apply to that of a body image. However, the body image is what gives the feeling of self in childhood. When a child acquires a body image, it is able to feel experiences as its own and differentiate them from the experiences of others. According to traditional authors, this does not happen before recognizing one’s own image in the mirror. This does not mean that the mirror, as a physical object, is the cause of this mental evolution, but the mirror translates it and measures it. Before this there is a phase designated by Merleau-Ponty as pre-communication and by Wallon as social syncretism where there is no differentiation between the I and others: as Merleau-Ponty says, “There is no one individual over against another but rather an anonymous collectivity, an undifferentiated group life”11. In early stages of life, there is a “chaos in which I am submerged” (Merleau-Ponty), a “blooming, buzzing confusion” (William James). A child does not have a unified vision of its body. This unified vision of the body depends, in part, on learning. A child’s perception starts as being interoceptive, since a newborn does not have the capacity to understand exterior information, as exterior. There is no distinction between self and environment. When a child recognizes the image in the mirror as his own this signifies that he can distinguish the other because it looks at him in the same way as another. His body appears to him as unified (in anticipated imagination because in reality it isn’t yet, according to Lacan) and like one among many others. This conquest is not yet total and definitive because it is here that the sketch of the self begins, which did not exist before. The evolution of the no-self phase to the self phase is described 11 Mearleau-Ponty, M. op. cit. p. 83 (my own translation).

Core Self and the Problem of the Self

215

by Wallon and Merleau-Ponty as an evolution of an interior that exteriorizes itself. Lacan is more radical in indicating that this pertains to an identification process in the psychoanalytical sense. There is, in the identification of the image of another, a process from outside to inside. Without the other, the child cannot construct the self, which is wholly realized in the image of this other. Whatever the explanation, what is important here is that these authors make a clear distinction between a period in which the self does not exist and one where it begins to exist. This has been criticized by more recent conceptions that attribute more capacities to newborns than those authors considered as traditional. According to the new investigators, a newborn already has an innate feeling of self that begins working right away. A newborn already has the feeling of its own experiences – for example, the sensations of pleasure and pain, that are his own and not another’s. Of course it would be a minimal feeling, but justifiably if these authors are right, their data will serve as support for the thesis that there is a core self where there is consciousness. The most important empirical fact that these authors have established is the observation of imitation by newborns. Merleau-Ponty wrote in “The Child’s Relations with Others” that “The child executes a gesture to the image that he sees made by another: he smiles because it smiled at him. It is necessary that the perceived image is translated into motor image; since he cannot have an image of himself smiling nor the motor feeling of another. The transfer of the other to him is impossible by analogy.”12 He shows a model here of translation that states perception and motor skills are different by nature considering that it is necessary to transform visual stimulus into motor stimulus. However, a child does not have an image of itself nor the feeling of the other before 6 months of age and because of this it cannot imitate. The integration of perception and motor skills will be progressive through maturity and learning. Gallagher and Meltzoff argue against this, based on various research, that there is imitation among children between 0 and 6 years of age, the main empirical fact being in reference to tongue protrusion. In agreement with their interpretation of the facts, a child recognizes the adult’s gesture of sticking out a tongue at him and rapidly transforms this perception into 12 Mearleau-Ponty, M. op. cit. p. 70 (my own translation).

216 

Jorge Gonçalves

a motor perception. To explain this fact, they argue that a supramodal system/device exists where the visual and motor systems speak, from the beginning, in the same language. A child does not have any difficulty in relating the actions he sees in others, that are in fact others to him, with the invisible actions within himself because innately, without requiring any training, he already has a body image that makes this connection. Obviously training will play an important role in his development, but Gallagher and Meltzoff argue that the child already has the cognitive baggage necessary in order to carry out elementary imitation. If imitation is possible, then we have evidence that a minimal sense of self and of others exists from birth. However, this interpretation of the empirical data has been opposed by researchers who carried out other experiments and offered alternative interpretations. Talia Welsh refers to these results13 and as a philosopher assumes the freedom to propose a global interpretation of the other results. According to her perspective, intelligent and communicative behavior does not necessarily imply a self-consciousness. If it did then we would also have to consider that bees have consciousness of self, which does not seem to be the case. Another interpretation of the facts related to sticking out the tongue is possible. The child sticking out his tongue is not imitating an older child. His behavior seems identical, but the phenomenology is different. In an older child’s imitation, there is, as we have seen, a mental picture, consciousness of himself and the other. Assuming that the same behavior exhibited by the newborn has the same meaning is adapting the facts to the theory. Talia Welsh argues this is valid both in the strong imitation thesis according to which a newborn does imitate, and in the weaker thesis according to which a newborn’s imitation is not properly imitation but is a preform whose development will result in imitation. The second possibility seems more plausible, but different experiments seem to deny it. The alternative theories that Talia Welsh refers to basically claim that the initial behavior does not reveal self-consciousness. It is merely exploratory reflex and auto-regulation, that exists to avoid discomfort. It is because of adult influence – the parents – that this exploratory behavior and auto-regulation transforms into a self-consciousness. I think that we could say that if we were left alone to ourselves (believing this were possible), the exploratory behavior and 13

Welsh, T. (2006).

Core Self and the Problem of the Self

217

auto-regulation would not make us auto-conscious. In the case of tongue protrusion, it is an exploratory behavior just like any other, which is reinforced by an adult and is repeated as such. Sticking the tongue out does not signify that the child understands the behavior of another as an other and then reproduces it in himself. This implies knowing how to move an invisible part of the body, knowing that it is his tongue that is moving and not the other’s. Nevertheless, various experiments, that I cannot enumerate here, go in the direction of showing that tongue protrusion is an exploratory behavior just like any other and does not have a posterior sense of imitation. It cannot even be said that it deals with a proto-imitation in the sense of a very basic mimicry that will be developed. It does not deal with any type of imitation, but with a different behavior, an exploratory reflex like any other. Although it is not proven that there are no feelings of self in the initial stages of life, even of a core-self, I think that there is good reason to speculatively concede that it does not exist. If it were true, what would the consequences be for the problem we are studying, the problem of the self? If the feeling of self is not present from the start, will there be a period of time in which the feeling of self and simple phenomenal consciousness are in different states? There will be phenomenal consciousness without the feeling of self. Thus the concept of pre-reflective self-consciousness cannot resolve the problem of the self. The feeling of self is a phenomenological fact and its reality depends on the reality of phenomenology. Metzinger considers that all phenomenology is an illusion, and in this manner the self does not exist. He seems to share the idea that phenomenology implies a self, but here I consider that phenomenology can exist without a self, thus the two problems are distinct from one another. If we accept that phenomenology is not a hallucination but a reality, we will find ourselves with yet another problem of knowing from where a self forms from no self. Through what mechanisms does the state of chaos and indistinctness becomes the feeling of being one, of continuance in time, of being a separate entity. Through simple nervous maturity? Through social interaction? From the phenomenological point of view, as a process of integration of the various states in one unified center? I do not intend to explore these questions, but only to demonstrate that the notion of core-self does not resolve the problem of the self. In

218 

Jorge Gonçalves

conclusion, given that one can phenomenologically exist without a minimum feeling of self, the problem of its nature and origin remains unsolved.

References Gallagher, S. (2007). The phenomenological Mind: An Introduction to Philosophy of Mind and Cognitive Science, New York: Routledge. Gallagher, S. (2000). Philosophical conceptions of the Self: implications for cognitive science. Trends of New Science, vol. 4, Nº 1, January. Gallagher, S. (2006). How the Body Shapes the Mind, Oxford: Oxford University Press. Gallagher, S. (2000). Philosophical Conceptions of the Self. Trends in Cognitive Sciences, 4(1): 14–21. Gallagher, N. & Meltzoff, A. (1996). The Earliest Sense of Self and Others: Merleau-Ponty and Recent Developmental Studies. Philosophical Psychology, 9(2), 211–233, 1. Lacan, J. (1966). Le Stade du Mirroir comme Formateur de la Fonction du Je, Écrits, Paris: Seuil. Mearleau-Ponty M. (1990). As relações com o outro na criança. In Sorbonne: Resumo de Cursos, Papirus, S. Paulo: Éditions Cynara. Metzinger, T. (2009). The Ego Tunnel – The Science of the Mind and the Myth of the Self, New York: Basic Books. Metzinger, T. (2003). Being No One. The Self-Model Theory of Subjectivity, Cambridge, MA: MIT Press. Parnas, J. & Sass, L. (2010). The Structure of Self-Consciousness in Schizophrenia. In Gallagher, S. (ed.) Oxford Handbook of the Self, Oxford: Oxford University Press, 521–546. Wallon, H. (1973). Les origines du caractère chez l’enfant. Les préludes du sentiment de pesonnalité, Paris: Boisvin. Welsh, T. (2006). Do Neonates Display Innate Self-Awareness? Why Neonatal Imitation Fails to Provide Sufficient Grounds for Innate Self and Other-Awareness. Philosophical Psychology, 19(2), 221–238.

Core Self and the Problem of the Self

219

Zahavi, D. (2005). Subjectivity and Selfhood: Investigating the firstperson perspective, Cambridge, MA: The MIT Press. Zahavi, D. (2005). Being Someone. Psyche, 11(5), 1–20. Zahavi, D. (2007). Subjectivity and Selfhood: Investigating the FirstPerson Perspective. Cambridge, MA: The MIT Press. Zahavi, D., Grünbaum, T. & Parnas, J. (2004). The structure and development of self-consciousness John Benjamins.

The Reality of the Virtual Self as Interface to the Social World Robert Clowes

Introduction This article argues for a virtual notion of self built around the idea of social self- positions. It begins by analysing several virtual-reality models of mind and noting the different ways the ideas of virtuality, virtual reality, and presence, can work in explaining mental phenomena. It looks particularly at Metzinger’s idea that the self is a form of virtual body representation. It then explores his contention that the self, if virtual in this way, is strictly speaking inexistent and should therefore be eliminated from our scientific conceptual vocabulary. It finds this conclusion is premature and ill-motivated. Metzinger’s notion of phenomenal self models (PSMs) opens the way to make the case that selves are actualizing virtualities. These are projective virtual entities that play a central role in organising our actions and constituting us as beings that act, take decisions and realize ourselves in consideration of an ongoing sense of ourselves. These particular actualizing virtualities can be viewed as instruments that we use to help maintain and realize ourselves as coherent beings. The paper then changes tack to look at one area where a notion of the self appears to be indispensible, namely, psychiatric science and practice and specifically the theorisation of schizophrenia. This examination reveals that selves are necessary in much psychiatric theorisation perhaps especially the Ipseity Disturbance Hypothesis (Cermolacce, Naudin & Parnas, 2007; Parnas, 2003; Sass & Parnas, 2003). I find that there are strong empirical implications that we will need to explain to

222 

Robert Clowes

understand how self can become distorted in pre-reflective experience in order to achieve better theorisations and explanations of certain aspects of schizophrenia. I then turn to a related, phenomenologically led, analysis of schizophrenia: the dialogic theory of self (Lysaker & Lysaker, 2008) which leads me to conclude that the virtual notion of self plays an important role in reconstructing our notion of pre-reflective self experience. One implication of this theory is that, in important ways, the self cannot be identified with the body, real or virtual. Rather self, even the pre-reflective self, has an irreducibly social and interpersonal aspect, albeit based in an extended form of body representation. Another implication is that we may not be able to fully explain “disorders of the self ” such as schizophrenia purely in terms of the embodied self, but may require a more extended and richer idea of self and selfexperience. Having made this leap, I return to, and attempt to reconstruct the idea of the virtual self, that, I argue, needs to be refocused around the social roles we inhabit, namely Lysaker & Lysaker’s idea of self-positions. We look at a reconstructed version of the virtualist representation of the self, arguing that while the self may be based in bodily representation, this representation is not targeted on the body. It should rather be understood as a sort of body-based projection into the social world, managing the more or less coherent projection of appropriate self-positions. This gives us the basis to contend that while the self may be virtual, it is not necessarily illusory. Rather the virtuality of self plays a central role in constituting us as the sorts of beings we are. Somewhat reconceived, the self concept nevertheless continues to do an important job both in our scientific and folk-psychological theorizing and potential explanation. Indeed, achieving a proper understanding of how the self may be grounded both in neuroscience and the social world may be needed if we are ever to make progress in understanding conditions such as schizo­phrenia and what it means to be a coherent human being.

The Reality of the Virtual Self as Interface to the Social World 

223

Virtual Reality and the Experience of Presence The idea that virtual reality might be a good metaphor for the architecture of our minds persistent in the literature on mind from the early 1990s. Several philosophers of cognitive science (Clark, 2008; Metzinger, 2004a; Noë, 2004; Revonsuo, 2006) have taken up the idea seriously although there has been little systematic attempt to look at the variety of sometimes conflicting ways that the metaphor is used1. Metaphors are important in scientific and philosophical model building (Lakoff & Johnson, 1999) although there is always a danger that they can be misleading if not examined properly, so this section will discuss some of the ways in which the ideas of virtual reality, and virtuality have been used in recent theoretical cognitive science before we return to its potential application to the self. Ancestors of the idea of virtual reality can be found in philosophy as far back as Plato’s cave and, in early modern thought, Descartes idea of a malicious demon. An influential recent version of these ideas was Revonsuo’s analysis of consciousness, dreaming and especially the then nascent work in virtual reality around the concept of presence. This led him to claim that perception, and our sense of reality more generally, could be considered a sort of total simulation (Revonsuo, 1995). Revonsuo thus uses virtual reality and telepresence technologies as a metaphor for normal or everyday experience. Today the concept of some virtualistic working out of perception has become rather widespread among philosophers of cognitive science, who ostensively hold very different ideas about what perception is, how it works and its broader place in the mind (Clark, 2008; Metzinger, 2004b; Noë, 2004; Revonsuo, 2006)2. One common way of theorising what virtual reality and telepresence technologies do is to hold that they project presence (Lombard & Ditton, 1997). That is to say they make us feel present in a scene, a 1 2

However the job of working out some of the possible positions was begun in the recent paper Virtualist Representation (Clowes & Chrisley, 2012). A detailed discussion of the strengths and weaknesses of a number of these views and how they relate to a broader virtualist picture can be found in (Clowes & Chrisley, 2012).

224 

Robert Clowes

locale, a place which may or may not be the one our bodies are actually located in. In virtual reality our bodies are typically located at a computer screen or, more exotically, using VR goggles or some other piece of technology. Yet we feel present, that is to say spatially and temporally situated and embodied in some projected scene. Behind any perceptual seeming is typically a computer simulation. The sense of presence – at least in virtual reality scenarios – always appears to depend on the mediation of an interface. An interface connects some actual or simulated world via technological device (such as a remote video-camera controlled by a joystick, or a head-mounted video display into a virtual world) to our sensory apparatus (our eyes, our ears). It is through the interface that we view the perceptual world and are able to act in it. All interfaces both make available and constrain certain possibilities for action. It is through the interface that the sense of presence is conveyed. When presence is spoken of as the sense of being there, or being now there in such scenarios it is relatively clear what is meant. We feel – to a greater or lesser extent – as if we were actually located in the virtual world being projected by the computer, or (in telepresence scenarios) in the actual location of the video camera we are controlling, rather than wherever our bodies are physically located. Some hold that the concept of presence as understood in virtual reality research can be understood to have a deeper significance than the ways in which we interact with computers. It has been argued that presence should really be understood as a central aspect of phenomenal experience itself (Seth, Suzuki, & Critchley, 2011). It has been linked to certain psychiatric conditions – e.g. depersonalisation disorder – supposedly marked by a deficit in this sense of presence (Radovic & Radovic, 2002; Sierra, Baker, Medford, & David, 2005). Another possibility is to link it to a wider set of existential feelings (Ratcliffe et al., 2008). Using the concept of presence – or the sense of presence – to refer to the something more basic about our apprehension of the world outside of VR scenarios, stretches the use of VR as a metaphor. The idea is that our basic cognitive architecture itself projects a sense of presence3. Anti Revonsuo (2006) has perhaps gone further than anyone in developing ideas around virtual reality as a sort of total metaphor both for 3

I return to this idea later in this article in the discussion of ipseity and the ipseity disturbance hypothesis about schizophrenia.

The Reality of the Virtual Self as Interface to the Social World 

225

mind and our knowledge of the world. Revonsuo uses the term “world simulation” to convey the idea than the world as experienced is itself a sort of simulation run by our brains. He writes: in world simulation, the brain creates for us only virtual presence or telepresence in the surrounding world, but we ourselves, and our consciousness never in actual fact escape the confines of the brain. (Revonsuo, 2006, p. 109)

On this analysis the use of telepresence or virtual reality technologies can be understood as promoting a perceptual illusion that we are somewhere – out in the world – when we are not. Our intuition that we are truly in touch with the world, on this analysis, is a sort of naïve-realistic illusion created by the brain as part of its representational activities. The idea that perceptual experience can be understood as a form of virtual reality which projects a world has become widespread. Indeed it can be taken as the new way of working out traditional philosophical ideas about indirect perception. For instance Barry Dainton recently describes the projectivist approach to perception: We take ourselves to be directly perceiving the world; we do not (while in the natural attitude) detect the presence of an experiential medium which lies between ourselves and the things in the world we see, hear and touch. Yet if projectivism is true, there is a sense in which we are all enclosed in spheres of virtual reality, phenomenal spheres somehow produced by activities within our brains: all we are directly aware of are contents within these spheres. But this is not how it seems. Even when I am dwelling on the absurdities of naive realism I do not seem to be enclosed in a virtual world. I seem to be surrounded by ordinary material things, tables, chairs, walls not experience. (Dainton, 2006)

Yet one does not have to be a projectivist or anti-realist in the way that Dainton discusses here in order to consider that the metaphor of virtual reality has some use. Implications of the virtual reality metaphor for mind are not in fact univocal. They can and have been used in ways to conceptualise the mind without implying that we are out of contact with the world beyond our perceptual processes. In fact, the idea of virtuality has also been used to theorize the way that our (especially) perceptual systems are in constant contact with, even including, the proximal world. On these views our perceptual system does not need to reproduce information internally because our cognitive systems operate as if absences were not there. (See Noë’s idea of enactive perception discussed below).

226 

Robert Clowes

One early and influential development of such ideas can be found in Chapter 11 of Dennett’s (1991) Consciousness Explained. This is developed around his critique of the then popular idea of filling in. Dennett’s discussion began from the close analysis of a number of phenomena about visual perception. His starting point was the well-known blind-spot in our visual field. It is well established that the part of our retina, where it connects to the optic nerve, does not in fact have light receptors. Yet as Dennett (and others) pointed out we do not notice any gap in our visual field. The question became – in much neuroscience discussion of the time – how is the gap filled in? Dennett believed the question to be misguided. On the contrary he argued the brain does not need to fill-in, it simply ignores the lack of information. In a related example Dennett imagines a viewer entering a gallery and coming across a wall of Marilyn Monroes in the style of Andy Warhol (ibid, pp. 354–355). Dennett notes that the art lover experiences a seemingly immediate impression of the visual similitude of all of those Marilyns and yet her visual system does not have enough time to pick-up and represent internally all of the information that would seem to be necessary for her to enjoy the apparent rich visual detail of the scene. Dennett’s explanation of this apparently rich phenomenology again turns on the brain not having to fill in detail but, on the contrary, its ability to selectively ignore the lack of certain apparently needed information to complete the visual field. Dennett imagines what our art lover’s brain might have to do on entering the gallery: “Having identified a single Marilyn, and received no information to the effect that the other blobs are not Marilyns, it jumps to the conclusion that the rest are Marilyns, and labels the whole region ‘more Marilyns’ without any further rendering of Marilyns at all.” (ibid, p. 355). According to the filling-in hypothesis the brain must synthesize content for those missing parts of the visual field in the blind-spot or produce extra Marilyns. Dennett argued instead that the brain practiced a benign form of neglect simply ignoring much absence which was only signalled to higher brain areas if some problem was noticed. Dennett formulated the problem of the tradition view and his response like this “the fundamental flaw in the idea of ‘filling in’ is that it suggests that the brain is providing something when in fact the brain is ignoring something” (ibid, p. 356). What the brain is ignoring for Dennett is the absence of a lot of detailed information about the environment.

The Reality of the Virtual Self as Interface to the Social World 

227

Nevertheless the brain does produce a sense of virtual presence. Dennett introduces the idea with a citation from Marvin Minsky: “We have the sense of actuality when every question asked of our visual system is answered so swiftly that it seems as though those answers were already there.”4 For Dennett the Marilyns were virtually present in consciousness in the sense that we had access to highly-detailed visual information when and as needed. Dennett and others have taken this line of thought to indicate that perception is not all that it seems (Dennett, 2001). Some even take it that the richness of the perceptual world itself is a kind of Grand Illusion (O’Regan, 2002). Other responses to the putative Grand Illusion, that hold that we do not misperceive the world. Alva Noë, Kevin O’Regan (2001) and others have developed an enactivist (or Sensory Motor Contingency) view of perception that in many ways builds upon Dennett’s ideas about virtual presence. They contended that the reason we enjoy rich experience of the world is not because of the construction of a detailed internal representation, but because we are constantly in touch with the world and able to pull in detail on a need-to-know basis. Noë refers to the idea in his more recent writings where he claims we enjoy virtual presence in perceptual consciousness because we are always able to draw on the actual presence of the world. He writes: We have the impression that the world is represented in full detail in consciousness because, wherever we look, we encounter detail. All the detail is present, but it is only present virtually, for example, in the way that web site’s content is present on your desktop. (Noë, 2004)

For Noë the detail of the visual scene is virtually present in the sense that it can always be effectively drawn upon with an – often unconscious – glance or a movement. However unlike Dennett, Noë argues this does not mean we are mistaken about perception (Noë, 2004). It is our active engagement (not our out-of-touchness) with the objects of our perceptual processes that makes a visual phenomenology come out correct. We believe we are in touch with the world and indeed we are, in virtue of our ongoing contact with it, and action within it. On this view virtual presence is not an illusion but a way of making sense 4

Cited in (Dennett, 1991) p. 359, from (Minsky, 1985) p. 257.

228 

Robert Clowes

of both phenomenology and the findings of perceptual psychology. For Noë, it is not that we are persistently out of touch with the world, but rather our cognitive architecture continually factors in the presence of that world. It is because we are continually drawing on our contact with it, that it is (virtually) present to us. On Noë’s interpretation we are not operating under any illusion that we are experiencing the rich detail of the world, because we are indeed experiencing that detail in the only way we could. Not because of the construction of a detailed internal representation, but because of our ongoing contact with the world. This article seeks to extend Noë’s general approach to the use of the virtualist metaphor, and specifically his critique of the idea of a Grand Illusion to dealing with the problem of self. Just as Noë argues we are in continual contact with the world and therefore not experiencing an illusion, I will argue we are similarly always in contact with the bases of self. For now it is enough to say that endorsing and using the virtual reality metaphor, in some form, need not commit us to the sorts of global anti-realism that at least Metzinger and Revonsuo think follow from it. On the contrary, it may be that a form of virtualism is actually one of the best ways of rebuilding a concept of veridical perception that does justice to psychological findings. As I will attempt to show, it can help rebuild a concept of the self that might allow us to ground some of the more important roles self plays in our current conceptual schemes. In order to do this, however, we need to do some conceptual groundwork. The virtual reality metaphor is easily misused such as when it is suggested that we are ‘imprisoned in the brain’. However, I shall attempt to show here that it has real purchase in helping us understand some of the ways in which situated agents like us experience our social presence in the world. It offers new possibilities for developing a refined analysis of that sense of presence and even how having a sense of presence links up with some important themes in contemporary psychiatry. This essay will aim at the reconstruction of the virtuality metaphor in such a way that it preserves some of its important theoretical advantages while some of its more paradoxical implications are overcome.

The Reality of the Virtual Self as Interface to the Social World 

229

The Virtual Self and Pre-Reflective Self Experience The idea that the self is in some way virtual can be found in several places in the literature including Dennett’s idea that selves are Centres of Narrative Gravity (Dennett, 1991). For Dennett, selves are narrative constructs created by organisms like us to make sense of themselves, but their apparent coherence is greater than its reality. We spin selves, and they do a job for us, but they are rather fictional: The best stories we can come up with, rather than a true story. It is Thomas Metzinger, however, who has done most to develop the idea of a virtual self in the recent discussion (2000; 2004a; 2009). Unlike Dennett, Metzinger’s notion of the virtual self is not a high-level narrative construct but is grounded in the way organisms represent and control their bodies. Metzinger’s idea of virtual selves can be seen as a way of resolving a major problem implied by Revonsuo’s (internalist) account of virtual presence, namely that our inner (Cartesian) selves are out-of-touch with the world beyond its representational media. But – it is natural to ask – if we concede that we are never in touch with reality itself, only our own representational media, what makes the self so special? Might not the self itself just be another form of representational content? The boldness of Metzinger’s (2004a) account is to do away with the confined or imprisoned self as posited by Revonsuo by proposing that the self is itself virtual. What Metzinger appears to primarily mean by this is that the self is a form of simulation produced by our brains in order to situate ourselves in the world. The benefit here is that with both the self and the world considered as artefacts of brain simulation, the radical disconnect is closed. On one level this is a nice way out of the cage that Revonsuo has built for the self. If there is no real self to begin with it makes no sense to think of it as being imprisoned in the brain. The cost is that, by Metzinger’s lights, the self as typically understood becomes itself a type of illusion. Whether it is worth bearing this cost, and indeed whether a well-developed theory of the virtual self requires us to bear this cost is one of the central questions of this paper. Representational systems on Metzinger’s analysis – in terms which derive from the mental models theory of classical cognitive science (see Johnson-Laird, 1983) – are just a certain class of brain models.

230 

Robert Clowes

Metzinger’s basic notion of the Phenomenal Self Model (or PSM) is a brain model that an organism builds of itself in order to plan its actions and allow the coordination of its parts. PSMs are necessary insofar as systems (organisms) like us require an approximately total coherent sense of themselves in order to organise their activities as a whole system. The evolutionary value of PSMs accrues in virtue of the fact that, for complex multi-limbed organisms like us, much everyday and planned action requires not just a response to environmental stimulus, but a nuanced, whole-body response that requires the overall coordination of body-parts with many degrees of freedom. Nevertheless, Metzinger frequently emphasizes how the self/world distinction is merely a somewhat arbitrary boundary in representational space; as Metzinger says at various points, it is a partition (e.g., Metzinger, 2004a, p. 87). One part of representational space is coded as the organism itself, the rest its environment. The sense that we have a self arises from this basic (internal and fallible) partition our brains make between self and world. The sense that this partition is in some sense arbitrary appears to be another reason Metzinger wants to say it is virtual, i.e., the self partition does not respect the existence of real entities in the world; it is only a form of projection that the organism finds useful. Metzinger claims that the self itself is an artefact of this sort of internal informational organisation. Selves arise in organisms that need to plan, reason, and act upon the world in sophisticated ways. In large part the job of our brains is not just to produce models of the environment, but, by using a variety of motor models, to simulate, model and situate our place in the world. According to Metzinger complex organisms like us build models of not just their environments but also of themselves, they (we) are self-simulators. The root of Metzinger’s argument that the self is an illusion hinges on his idea that the self is really a sort of artefact produced by representational systems like us in the act of self-control. This arrangement, for Metzinger, simply falls out of the nature of our basic computational architecture. Crucially the Phenomenal Self Model (PSM) is not merely a way of navigating the environment but the basic means by which we become aware of ourselves as unified beings. PSMs are conscious self models. PSMs thus unite two important aspects of the sorts

The Reality of the Virtual Self as Interface to the Social World 

231

of beings we are. They are simultaneously a sort of content produced by certain complex agents for self-control, but they are also the means by which those agents become aware of themselves. As Metzinger writes: Conscious experience, too, is an interface, an invisible, perfect internal medium allowing an organism to interact flexibly with itself. (Metzinger, 2009, p. 104)

PSMs are also thus, in terminology we have already discussed, to be regarded as interfaces. They exist in order to mediate certain types of action which require the coordination of the organism as a whole. Insofar as these models are interfaces they present particular modes of action and possibilities and obscure others. Those that are presented are ultimately oriented to the organism’s self-control. They do not however present, as Metzinger often notes, some deep ontological truth about the body or its relationship with the world. Rather they are tools with which to produce coherent action and ways of regarding and interacting with oneself. For Metzinger the PSM is a model that complex organisms build of themselves in order to produce and integrate coherent action. The content of a PSM is the organism’s current best guess about how it is concurrently poised and organised. This is another sense in which Metzinger holds that selves are virtual, i.e., they do not reflect any objective truth about the world, but, like the red arrow on a subway map, they allow us to find ourselves in the world and orient and act within it. Metzinger writes: This self-model is the little red arrow that a human brain uses to orient itself within the internal simulation of reality it has generated. The most important difference between the little red arrow on the subway map and the little red arrow in our neurophenomenological troglodyte’s brain is that the external red arrow is opaque. (Metzinger, 2004a, p. 552)

This metaphor of the little red arrow is important to Metzinger because it establishes a central phenomenological role for the PSM. The PSM is the means by which organisms are able to transparently and pre-conceptually register their presence in the world. But the PSM, rather than being a content of experience, is given transparently in experience. When Metzinger writes the red arrow is opaque, the notion of opacity being used is a phenomenological one to be contrasted with phenomenological

232 

Robert Clowes

transparency. The red arrow on the subway map is a perceptual object that allows us to establish our position in the world. But the PSM is not encountered as a content or object of perception but as part of the way the world is given to us in experience. Put another way, through the PSM, the self is encountered, phenomenologically speaking, not as a content but as an aspect of experience. In this way – and perhaps unexpectedly5 – Metzinger’s ideas connect up those neo-phenomenological writers trying to explicate the basic structure of experience; especially with those writers who contend that there is minimal sense of self to be found pre-reflectively in experience (Gallagher, 2005a; Legrand, 2007; Sass & Parnas, 2003; Zahavi, 2005b). Metzinger’s notion of a transparent PSM thus endorses much recent and classical phenomenological discussion which finds a basic apprehension of self in the structure of phenomenal experience itself. What Metzinger refers to (citing Frank, 1991) as pre-reflexive self-intimacy6 appears to be the very same concept that Zahavi and others call pre-reflective self intimacy. What more ontologically realistic writers are inclined to call the minimal self is thus very closely connected to what Metzinger calls a PSM. The way that the PSM operates within the organism is supposed to explain certain aspects of phenomenology, and crucially for our discussion, why it is misrepresentational. It is important to remember that Metzinger claims that the pre-reflective self experience presents us with an illusion. Metzinger thinks that insofar as the basic structure of subjective experience presents us as having a form of pre-reflective self-consciousness, it is also presenting us with a distorted picture of ourselves and our place in the world. For Metzinger our pre-reflective sense of self is merely the projection of a virtual reality producing system, that is, a system manufacturing a sense of presence by continually attempting to model its ongoing interactions with the environment in the most coherent way possible. 5 6

As Metzinger is often taken as being anti-realist about phenomenology. Metzinger writes the following, in terms that seem to patently echo what others call pre-reflective self-consciousness: “There seems to be a primitive and prereflexive form of phenomenal self-consciousness underlying all higher-order and conceptually mediated forms of self-consciousness and this nonconceptual form of selfhood constitutes the origin of the first-person perspective.” (Metzinger, 2004a, p, 158).

The Reality of the Virtual Self as Interface to the Social World 

233

What we experience then is not ourselves or the world but a complex simulation. Thus Metzinger argues that “phenomenal first-person experience and the emergence of a conscious self are complex forms of virtual reality” (Metzinger, 2009, p. 106). Metzinger argues that in a deep and metaphysically necessary sense, PSMs only present a bestguess about how things really are. At any given time, the content of bodily experience is the best hypothesis that the system has about its current body state. The brain’s job is to simulate the body for the body and to predict the consequences of the body’s movements, and the instrument it uses is the self-model. (Metzinger, 2009, p. 114)

This sense in which PSMs are only best-guesses is for Metzinger another, indeed the central reason that they should be considered virtual. To recap then, for Metzinger, selves are virtual, or are analogous to virtual reality devices, in at least five distinct senses. 1. Selves are virtual because they are simulations; or more precisely, simulational content. 2. Selves are virtual in that they are constructs arising from partitions in representational space that do not necessarily respect any divisions in reality. 3. Selves (PSMs) are virtual – by analogy with virtual reality systems – because they can be regarded as control interfaces that make available (and obscure) certain possibilities for action. 4. Like the interfaces in virtual reality devices, selves (PSMs) convey a sense of presence. 5. The contents of PSMs are also virtual because they tend not to correspond to, or depict, actual states of affairs in the world, but best-guesses or mere possibilities about the state of the body. Insofar as PSMs show up for us as pre-reflective and transparent aspects of our experience, they also tend to foster certain misapprehensions or ‘illusions’ about what we are and our place in the world. As Metzinger likes to remark: “there is no thing like “the self.” Nobody ever had or was a self.” (Metzinger interviewed in Taft, 2012). While such conclusions are clarifyingly provocative, I do not believe they follow from anything we have discussed so far. It is possible

234 

Robert Clowes

to endorse most, or indeed all, of what Metzinger has to say about the virtuality of self, without drawing his conclusions about the self ’s inexistence or unreality. In order to better understand how Metzinger comes to the conclusions he does, and why they do not follow, we need to narrow our focus to his ideas about the body and body representation.

The Virtual Body and the Minimal Self Metzinger develops several examples to help illustrate his case that bodily representations, or more properly representations of a self grounded in the body, are virtual and thus illusory. One rather exotic case study comes from the observations of some problems of astronauts during the course of space-travel (Metzinger, 2004b). While in orbit or otherwise placed in a condition of weightlessness, it appears that human beings can develop a confused sense of presence, i.e. the sense of how their body fits into a wider spatial frame of reference. Alongside this, they sometimes undergo quite bizarre illusions about the current state of their bodies. This condition, known as “Space Adaptation Syndrome”, can also cause severe nausea and appears to affect around 50% of those who participate in space flight. Interestingly it is not merely the sense of spatial orientation which is affected but also the disposition and integration of one’s body. Here are the words of one astronaut: The first night in space when I was drifting off to sleep,” recalled one Apollo astronaut, “I suddenly realized that I had lost track of [...] my arms and legs. For all my mind could tell, my limbs were not there. However, with a conscious command for an arm or leg to move, it instantly reappeared – only to disappear again when I relaxed. (Barry & Phillips, 2001)

This condition seems to be created by the lack of certain contingent stimulations or sensory information that would that appear to help our brains construct a fused image of the body, along with its current disposition, composition and spatial configuration within a reference frame. The particularly common aspect of the deficit appears to be a disruption of the subject’s ongoing sense of where the ground is and hence how up

The Reality of the Virtual Self as Interface to the Social World 

235

and down should be organised in egocentric space. Metzinger (2004b) writes that in such cases other astronauts can “help their buddy” by striking the soul of his or her foot. This seems to help with restabilising a more normal reference frame and with it a coherent representational fusion for the body. It is perhaps surprising that a coherent and stable body image turns out to depend on such apparently contingent matters as the impact of one’s feet on the ground. However, it is not clear why one should think that, because body-image and the sense of presence can be disturbed by the absence of a normal gravitational field, they are systematically deficient or misrepresenting. What we might admit, however, is that insofar as the systems that the brain uses to represent and control the body are considered to give a fully contemporaneous and veridical snapshot of the body, they are not fully successful. One further hypothesis, in the case of the astronaut, is that it is not so much that the egocentric body image is impoverished in itself, but that some bodily representations are poorly integrated into one’s overarching subjective experience. This line of thought fits nicely with some theoretical accounts that claim that the brain systems that underlie our sense of body schema and peripersonal space are the same system (Cardinali et al., 2009). In space sickness the sense of how the body is embedded or situated in a wider space becomes, for the organism, confused and alongside it the occurent sense of the body. On this analysis the partition in representational space appears to be intact but its embedding in a more global space can no longer be adequately fused. The condition is in part interesting because it reveals how under standard conditions, the representational systems of our body do invisibly perform this integration. Space sickness is the phenomenological correlate of the inability to fuse a stable sense of the body embedded in space. The discussion of space sickness also helps us understand what Metzinger means by a virtual model. Regarding the astronaut example he notes, “[t]his shows that the self-model of human beings is a virtual model which, if underdetermined by internally generated input (i.e., from gravitational acceleration affecting the maculae utriculi and sacculi in the vestibular organ), is highly context dependent: Its content is a possibility and not a reality.” (Metzinger, 2000, p. 290) This helps convey part of what might make the PSM a virtual model in Metzinger’s sense, and

236 

Robert Clowes

presumably in some sense ontologically dubious. First the model goes beyond the data, insofar as it is made to fit into some pre-existing structure. Second, the model does not convey and probably does not seek to convey the actual current information from the body, but is rather a sort of most-coherent best-guess about that body’s actual state as embedded in a spatial reference frame. But why should being under-determined by the data in this way make body representation illusory? Insofar as the brain’s model of the body goes beyond the actual state of the body by, for instance, creating various forward models to predict action, this can be argued to be – far from necessarily problematic – the best way to actualize an overall and coherent view of the body. Indeed this picture fits nicely into some current attempts to understand representation along Bayesian lines (Clark, 2012) or in other ways that see mental representation grounded in anticipatory systems (Grush, 2004; Pezzulo, 2008). Is there some other – non-problematic – way that we could grasp the state of our body, say by creating some instantaneous internal picture? Is the fidelity on an internal ‘snapshot’ of the body really a good success criterion for a felicitous representational architecture? Who or what would be looking at such a model anyway? Whether our sense of ourselves is compromised insofar as it relies on such models is surely something which needs to be shown. Nevertheless, I would like to agree that Metzinger is importantly on the right track about the way in which our representational processes might be considered virtual. But whether or not this is inherently problematic rather depends on what we think the desiderata might be for a proper or realistic sense of self and indeed successful representation. A second set of illustrative examples, Metzinger uses, involves phantom limbs. It is well known that when a human being suffers the loss of a limb, a side-effect of the corporeal damage is often the appearance – for the patient – of a phantom limb (Ramachandran & Blakeslee, 1998). Phantom limbs are experienced by some patients as highly present and real even to the point that they will appear to forget they do not really have the phantom, by falling over as they attempt to stand on a missing leg, or attempting to reach for an object with an arm that doesn’t exist. On reflection however subjects with phantom limbs are

The Reality of the Virtual Self as Interface to the Social World 

237

generally aware that the phantom is unreal although they cannot shake off its felt presence. Phantoms thus appear to be relatively immune from higher-level knowledge or standard rational inference. For Metzinger, this is another illustration of how the body is virtual. One hypothesis about such cases is that the body’s best guess about itself is constrained by certain pre-suppositions about the form the body must take. This causes the brain to fairly systematically miss-represent the current reality of the body.

Figure 1: Taken from Brugger, P., S. Kollias, et al. (2000).

Strong evidence for this hypothesis is the fact that subjects who have never possessed their actual organic counterparts can experience phantom limbs. Such aplasic phantoms are demonstrated in the well-known case by AZ: a middle aged woman born without forearms and legs, who nevertheless consistently and continuously experienced phantoms of these ‘missing’ limbs. The diagram (from Brugger et al., 2000)7 7

This diagram is reproduced with kind permission of the authors.

238 

Robert Clowes

shows her rating of the reality of her limbs on a seven point scale; with seven signifying a bodily part feels as real as actual parts of her body. As can be seen, parts of AZ’s virtual body appear to her as almost fully realised. The systems that produce this sense of bodily presence appear sometimes limited in their ability to reconfigure to meet unusual or abnormal body configurations, or simply unable to track changes to the plan of the body after injury. While explanations of this phenomenon are tentative8, it seems that body image does have a sort of default topography. But the pre-reflective sense of the presence, disposition and organic composition of our bodies seem difficult and sometimes impossible to shake up by rational inference. When these systems go wrong there can be jarring inconsistencies between what we see, believe and what we feel. Even insofar as we know at some level we do not have certain body parts they can still be felt as present, indeed as almost fully real. The sub-personal mechanisms of the astronaut with space sickness cannot seem to integrate a stable body configuration in relation to the immediate environment (a body-world relation). The sub-personal mechanisms that present the body of the phantom limb patients have trouble squaring residual (or expected) image with the body’s current configuration. From these examples it is easy to see why Metzinger claims that body representation is virtual in the sense that it is not a straight-forward mirroring of reality but a sub-personally produced best-guess about the body’s current state and its location in a spatial frame. But it is less clear why he thinks this implies that our conscious experiences are of a body that is in some sense unreal. Take Metzinger’s remark that: “Strictly speaking, and on the level of conscious experience alone, you live your life in a virtual body not a real one.” (Metzinger, 2009, p. 114). On the face of it, given Metzinger’s ultimate argumentative strategy, this comment is difficult to interpret. Who is this you that lives its life in a virtual body? Metzinger seems to be simultaneously denying the validity of a concept while trading on it. Yet there is a deeper problem with what is being said here. Is the 8

There is an interesting discussion of whether the findings about aplasic phantom limbs should be taken to indicate body-image is innate (2005a). The question goes beyond the scope of this paper but Gallagher still appears to think it an open one, awaiting more compelling empirical research.

The Reality of the Virtual Self as Interface to the Social World 

239

virtual not part of the real? It has been remarked that counter posing the real with the virtual is an ontological mistake, a better opposition may be between the virtual and the actual (Deleuze, 1991). In the case of the experience of presence it is not so much that what we experience is unreal but that there is a virtual element to our experience of the actual. The real question is then, what role do virtual constructs play in our cognitive lives? Statements that imply we ‘live our lives’ in unreal bodies elicit more questions than they answer. To see why, consider how the PSM is a production of bio-representational systems in the act of producing coherent action. In the case of body-representation there is a something, an “image” or model, that is produced as part of the process of an organism taking control of itself. But this certainly does not make either the processes that produce this representational content or the content itself unreal. Insofar as a subjective sense of presence is the correlate of this fusion of an adequately coherent representation of our bodies in space, this also is not obviously illusory. Insofar as PSMs are used as part of a bodily control system they are as real as all of the neural structures and bodily structures that realize them. Insofar as PSMs also make available our capacities for pre-reflective self-awareness, they are as real and non-illusory as any other aspect of phenomenology. In terms of content they are a real reflection of a set of processes that seek to achieve coherent self-control through projecting an overarching self-unity. Body image may be virtual in Metzinger’s sense: that it is the brain’s best guess about the state of the body, yet it is not clear why this should make body-image, or other self-representational properties, illusory in any general sense. It merely indicates that they operate within certain tolerances. Yet while phantom limb and the astronaut examples are striking evidence of the variability and sometimes non-veridical character of body image, any implications this has for our understanding of self requires further analysis. It is clear that our brains operate given certain assumptions, not all of which are fulfilled in all circumstances. The lack of gravity and actual limbs may be the limits of the condition under which our good representational system can fuse a coherent and reasonably veridical sense of the body. It is also clear that they have a predisposition to present

240 

Robert Clowes

a coherent sense of body, even under circumstances of great difficulty in integrating that information into a coherent whole. One way of developing this is to draw an analogy with Noë’s response to the Grand Illusion. For Noë we are mislead about the nature of our perception only if we practice bad phenomenology, e.g. by assuming that the way the world is presented to us is like some photographic snapshot. But the world is not presented in this way and our apparent richly detailed access to it is not similarly an illusion (Noë, 2002). By analogy we might argue that if we require for selves to exist that we need to have a snapshot-like impression of some underlying reality, then there is an illusion. However, if we relax the constraint and merely require we are in contact with the bodily basis of self, as and when needed, then there is no illusion. It is just the way that the brain fuses a body image, a sense of presence and the bodily basis of self – is by virtual means. But none of this implies our bodily sense of self is typically illusory.

In What Sense is the Self Illusory? Metzinger’s notion of a PSM and his use of the metaphors of virtuality has not especially drawn positive comment from those working in the contemporary phenomenological tradition (Gallagher, 2005b; Zahavi, 2005a)9. But it is important to recognise that Metzinger’s theorisation of the subjective structure of experience has many points of contact with that tradition and perhaps especially the contemporary approaches to neuroscience-inflected phenomenology being developed by Gallagher and Zahavi (2012). In fact, Metzinger’s PSM is supposed to offer an explanation of pre-reflective self-consciousness by grounding its analysis in terms of functional and neuroscientific analyses. But this can make the basic point about the self being illusory or unreal all the more perplexing. If the self is grounded in the body and 9

On the subject of Metzinger’s virtuality metaphors, Shaun Gallagher wonders whether “Metzinger has become tangled in a matrix of his own self-generated metaphors” (Gallagher, 2005b).

The Reality of the Virtual Self as Interface to the Social World 

241

virtual, why should this imply the scientific elimination of the concept of the self? In fact, the ‘partition’ in representational space that makes possible the PSM reflects the way our body is a relatively distinct part of the world, an organism in fact. The idea that embodied self is unreal merely because it depends on virtualistic body representation seems, in this light, bizarre. Nevertheless it is clear that Metzinger thinks the various cases of body illusion – and therefore in his view, the virtuality of the body – are supposed to convince us that we should eliminate the self. I have found it difficult to fully reconstruct the argument in Metzinger’s work on this point as it often seems that he just takes it that if the self can be shown to be virtual, in the several senses previously detailed, this just implies it is ripe for scientific elimination10. Nevertheless, I detect the following lines of reasoning all of which Metzinger takes to imply the elimination of the self in theoretically interesting contexts that seek to explain subjectivity and how our minds work. I. As PSMs are virtualist in the sense they are only best-guesses or hypotheses about actual states of affairs (5th sense of virtual above), they are strictly misrepresentations. II. As PSMs are optimised to be control devices (virtualist in sense 3) they are not just simplifications or reductions of reality, but tend to systematically misrepresent reality in terms of an agent’s action possibilities. III. As PSMs are pre-reflective modes of accessing reality they tend also to pre-dispose us to trust the way they represent reality and just take their presentations as actual. These presentations are relatively immune from higher level knowledge. Which Metzinger takes to imply IV. As PSMs are our basic mode of encountering ourselves, and because they necessarily misrepresent what we really are, we operate under incorrigible and systematic illusions about ourselves. Let us examine these points in a little more detail. 10 In fact Metzinger is very committed to the “no-self ” theory on independent grounds. See Metzinger (2011).

242 

Robert Clowes

Even some of Metzinger’s favourite examples seem to work against his case for I. The cases of phantom limbs and the astronaut, rather than demonstrating why we are deluded about the body, seem to show merely that body representation is contingent on some unexpected but lawful characteristics of how the body is in fact situated in the world or internally configured, and the way the mind brings structure to those representations. The astronaut gets space sickness not because the connection between bodily content and the body is random but rather because the relation is contingent on worldly interactions such as the sensation we normally have of our feet hitting the ground, and indeed a host of other sensory impressions that are created merely by moving our bodies around in gravity. When this normal information is absent it may or may not be surprising that bodily representation cannot be fused in the usual way, but this hardly indicates we are usually misled. Similarly in cases of phantom limbs there are number of explanations but all seem to turn on the usual or expected information the brain would receive from the body being interrupted and what remains being misinterpreted11. Phantoms are produced when organic bodily damage (or congenital conditions) produce bodily configurations that cannot in some sense be integrated veridically into the brain’s pre-conceptual sense of its body. These findings actually undermine hyper-representational and internalist pictures of mental representation that picture us experiencing only “world simulation” or “living our lives in a virtual body”. They also indicate problems inherent in internalist ideas of presence where simulations of the body float entirely free of the actual body’s worldly interactions with its immediate environment. They rather serve to demonstrate that these simulations are being fed by contact with the extra-dermal world and a constant attempt to make the best sense possible of this ongoing contact. If the experienced body is virtual in the sense that it is a best guess, isn’t this a reflection that all such knowledge can never be more than 11 Clearly this point is tricky in the case of aplasic phantoms. Nevertheless if we assume that at least the core representational systems involved in the representation of the body are developed to deal with a four-limbed creature, the absence of limbs, even if congenital, can still be regarded as a case where the absence of information which would have been normal in the evolution of the system.

The Reality of the Virtual Self as Interface to the Social World 

243

provisional? The claim that our representational equipment is based on best guesses elicits questions about what the successful conditions for our perceptual apparatus are supposed to be: Godlike knowledge? What an examination of body representation does suggest – contra some influential claims that representational fusion is unnecessary for intelligence (Brooks, 1991) – is that the various component systems that represent the body generally are unified in some way. Nevertheless the various component systems can come into conflict in a variety of ways and these can result in odd perceptual configurations and senses of bodily self. None of this obviously implies the systems that produce a bodily sense of self, or underly our experience of presence, generally misrepresent the way things actually are with our bodies. If our pre-conceptual sense of presence, including the configuration of our body, the body’s organic composition, and the way our body fits into a wider reference frame, appear fallible, this is no different from the fallibility of any other sort of perceptual content. Metzinger’s first argument (I) turns on the series of examples around malfunctioning body representation that are supposed to indicate the PSM’s non-veridical contents. While, the examples of the astronaut and phantom limbs may indeed convince us that body representation is in various ways fallible and often has non-veridical contents, it is unclear how this is supposed to indicate such representational content is typically misrepresenting. A second line of argument (II) is based around the claim that because our sense of self is really given by an instrument the primary function of which is self-control, it misrepresents the way things really are with us. This line of thought is particularly evident in his recent book The Ego-Tunnel (Metzinger, 2009), indeed the name of the book reflects the way in which Metzinger holds that our representational architecture does not present a veridical picture of reality but a literally self-centred construct: An Ego Tunnel. Sometimes Metzinger runs this argument around the more general claim that all representation is a sort of simplification and reduction of reality beyond our perceptual and representational apparatus. On this analysis representational contents misrepresent insofar as they necessarily reduce and limit the true complexity of their worldly objects. The problem is that this is not just something that faces the representation

244 

Robert Clowes

of self, but all noumena more generally. The representational process is always in a certain sense a reduction of reality, or so Metzinger claims. But then, what is special about self-representation? Aren’t the targets of all presentations similarly approximate? Our grip on them only ever partial and reductive? All presentations are just ‘tunnels’, i.e. a kind of collapsing of the world into a reduced form that is mentally graspable. In fact, as we have seen, there is an important sense in which PSMs are indeed special, i.e., in the third sense of virtuality noted above: because they are control systems. The actual content produced by the PSM is optimised for certain specific types of self-control. It is not an objective representation, especially in the “view from nowhere” sense of objectivity. But, this glass can just as easily be portrayed as half-full as half-empty. Rather than aiming to reconstruct a snapshot of external reality or a god’s eye perspective on ourselves, it is better to recognise that, at least at the pre-reflective level, our brains produce a sense of ourselves and our place within our immediate surroundings – essentially a sense of presence – which is optimised towards the production of a coherent scene and time-frame in which action is coordinated. This physical and temporal space of presence can be viewed as much a way of going beyond the merely actual, as it is a reduction of reality. To take just one example let’s take the apprehension of “thick” moments of time in consciousness. What James called “the specious present”, and Husserl described in terms of the protentive and retentive aspects of consciousness, both gesture toward how our subjectivity creates or projects a sort of extended now. The example often used to illustrate this is how we listen to music. When listening to a tune we do not merely hear the note being played at some exact time-slice of that instant but have a sense of it being the continuation of a melody from a few seconds ago, and tending toward a future now being realized. For Husserl (and James) retention was not memory and protention not foresight but rather – phenomenologically speaking – part of the moment of consciousness itself. In fact Metzinger’s notion of presentationality12 is exactly a notion of how consciousness presents the world as necessarily temporally extended. This extended now is but one way that 12 Metzinger discusses what he call “the presentationality constraint” extensively in Being No-One as a necessary aspect of conscious representation, see especially sections 3.2.3 and 6.2.7 (Metzinger, 2004a).

The Reality of the Virtual Self as Interface to the Social World 

245

the representational process can be considered an enhancement, rather than a reduction of reality. Other forms of representational enhancement make it possible to form impressions not just of how the world is, but how it might be and how we might affect the world in order to make it more like we would might like it to be. It is thus very one-sided to claim that the conscious representational process is simply one of diminishment as it is just as easy to characterize it as one of enhancement. More problematic still is that Metzinger offers us no standard as to what would be an accurate and veridical process of representation. For these reasons it is difficult to conclude that either I or II have demonstrated that we necessarily misrepresent ourselves in consciousness. Third, Metzinger argues the PSM forces us to confound ourselves – presumably the real organismic us – with a type of content produced by a simulation process in our brains. Or, put differently, from our experience we take ourselves to be an entity in the world, but this we cannot be, for all we can experience is a form of content produced by the brain when involved in certain modelling activities. As Metzinger writes: In other, more metaphorical, words, the central claim of this book is that as you read these lines you constantly confuse yourself with the content of the self-model currently activated by your brain (Metzinger, 2004a, p. 1).

But are we really ‘confusing’ a type of representational content with an actually existing being. It is odd to view representational content as things that can be mistaken for worldly entities. If the content of our internal representations is – qua Metzinger’s hyper-representationalism – the only things we apprehend, what is the comparison class with which we are mistakenly comparing them? The natural response here is to say that we apprehend the world through our representational media. Metzinger has not shown us at that this way of thinking is a mistake. Metzinger’s critique of our supposed naïve realism appears to have a more general problem, however. If we are making a mistake about the self, aren’t we making similar mistakes about everything else we perceive? But if this is so, then selves are no more problematic than all of the other mediumsized targets of our representata. We do not just mistake representations for ourselves, but also representations for trees, rocks and automobiles. But Metzinger does not call for the elimination of all of these. Either we

246 

Robert Clowes

are global anti-realists or moderate realists about the putative targets of representation. There is nothing special about selves here. Metzinger does flirt with a more global anti-realism in The Ego-Tunnel, but this does not much help underwrite the supposed special unreality of the self. More­ over if, as I have argued, Metzinger’s arguments I and II fail to demonstrate that the PSM does misrepresent reality, then argument III does not follow in any case. For, if PSMs do not systematically misrepresent, then it has not been demonstrated we are systematically operating under illusions. On these grounds at least, there is then no reason to conclude IV: that we are operating under systematic illusions about ourselves.

Presence, the Minimal Self, and Ipseity Disturbance One further problem with Metzinger’s apparently radical goal of eliminating the self is, in the field of psychiatry, and clinical psychology, it is kicking at an open door. Parnas and Handest, writing on schizophrenia and arguing for the necessity of self: “The notion of a self is deleted from the terminology of DSM-IV and ICD-10. It is rarely used in psychiatric literature, and usually in a colloquial or psychoanalytic sense” (Parnas & Handest, 2003, p. 122)13. As the Diagnostic and Statistical Manual of Mental Disorders (DSM) and the International Classification of Diseases Manual (ICD) represent the most important and widely recognised diagnostic manuals used by psychiatrists, clinical psychologist and theoreticians alike it is important to realise here that the self as theoretical category – at least in psychiatric practice – is a category under threat. Yet there has been a far from universal positive reception of these moves. DSM’s focus on describing symptoms and presentations of illness 13 In fact it is not strictly true that DSM IV and ICD-10 were entirely successful in deleting the self. It may not show up in the diagnostic criteria for schizophrenia, but, as we shall see in a moment, it still appeared necessary in to use the concept in the diagnostic criteria for both Depersonalisation Disorder (DPD) and Dissociative Identity Disorder (DID); (the latter once known as Multiple Personality Disorder). Some residual of the self as a theoretical category it would seem is not so easily annihilated.

The Reality of the Virtual Self as Interface to the Social World 

247

rather than providing underlying explanation may be one reason the category of the self can apparently be discarded. Parnas, Sass and their colleagues have been at the forefront of those arguing that we need, as a matter of urgency, an adequate rethinking of the concept of self that would allow us to do justice to the reality of several types of mental illness, (Parnas, 2000, 2003, 2012; Parnas et al., 2005; Sass, 2003). Historically speaking, some notion of the distortion of self, self experience, and the unity of consciousness was central to the very first formulations of the disorder undertaken by Bleuler and Kraepelin, along with related themes such as the unity of consciousness. Commenting on this absence of the self in the contemporary diagnostic manuals, Louis Sass and Josef Parnas write: It has long been recognized that schizophrenia involves profound transformations of the self. Eugen Bleuler (1911, p. 143) noted that the patient’s ego tends to undergo “the most manifold alterations,” including splitting of the self and loss of the feeling of activity or the ability to direct thoughts. Kraepelin (1986) considered “loss of inner unity” of consciousness (“orchestra without a conductor”) to be a core feature of schizophrenia. (Sass & Parnas, 2003, p. 427)

Thus since its clinical recognition, schizophrenia has often been understood as a disorder of the self but this has arguably been somewhat obscured by the clinical concentration on other aspects of the disorder, especially on the set of positive (hallucinations and delusions of control) and negative symptoms (inexpressive faces, monosyllabic or absent speech, lack of interest in the world) codified by the DSM IV. Nevertheless, several groups of contemporary researchers have continued to develop the theoretical links between a series of refined notions of self and self-experience, and how they can be compromised in a variety of psychiatric illnesses. Arguably, an adequate conceptualization of several types of altered mental state and psychiatric illness appears to be virtually impossible to even characterise without making some use of the notion of the self. Even in the super-positivistic new version of DSM (DSM-5) it is still apparently necessary to make use of the concept of self in describing the symptoms of Dissociative Identity Disorder (DID)14. 14 The draft of DSM 5 includes the following explanation as part of the definition of DID: “Disruption of identity characterized by two or more distinct personality states or an experience of possession. This involves marked discontinuity in sense

248 

Robert Clowes

Another important disorder in which reference to some notion of the self appears to be a conceptual necessity – and indeed the word self is still used in both of the main standard diagnostic manuals is15 – Depersonalisation Disorder (DPD). Patients with DPD often find their way to a specialist complaining of difficult to articulate “feelings of unreality”. Patients find it hard to explain or put into words their strange experiences they encounter but they often fall back on one or both of two metaphors: A first kind makes reference to a sense of being cut-off, alienated from oneself and surroundings. For example, patients would often talk about ‘being in a bubble’, or being ‘separated from the world by an invisible barrier such as a pane of glass, a fog, or a veil’. A second group of metaphors emphasise instead a qualitative change in the state of consciousness, and the feeling as if in ‘a dream’…‘stoned’, ‘not awake’ or an indescribable ‘muzzy feeling’, etc. (Sierra & David, 2011, pp. 99–100).

Two signature features of DPD are depersonalisation: The disturbing feeling of being separate from oneself, observing oneself from the outside, or feeling like an automaton or robot; and derealisation: the threatening sense of the unreality of the environment, a perceptual sense of unreality, and sense that others are actors in a play. But in fact depersonalisation and derealisation look to be closely interrelated in that it is the problematic embedding of body (or bodily self) in the world which appears to be at the root of the disorder: Depersonalisation disorder involves an unpleasant, chronic and disabling alteration in the experience of self and environment [my emphasis]. In addition to of self [their italics] and sense of agency, accompanied by related alterations in affect, behavior, consciousness, memory, perception, cognition, and/or sensory-motor functioning. These signs and symptoms may be observed by others or reported by the individual.” . 15 DSM IV characterises depersonalisation as ‘alteration in the perception or experience of the self so that one feels detached from and as if one is an outside observer of one’s mental processes or body’ (American Psychiatric Association, 1994: pp. 488–490). ICD-10 characterises depersonalisation-derealisation syndrome as ‘in which the sufferer complains that his or her mental activity, body, and/or surroundings are changed in their quality, so as to be unreal, remote, or automatized’ (World Health Organization, 1992: p. 171). Both cited in (Medford, Sierra, Baker, & David, 2005).

The Reality of the Virtual Self as Interface to the Social World 

249

these classic features of depersonalisation and derealisation, symptoms may also encompass alterations in bodily sensation and a loss of emotional reactivity.16 (Medford et al., 2005)

Depersonalisation and derealization appear to turn, for their conceptual coherence, on the problematisation of the self/world boundary or, to put this another way, the pre-reflective experience of how we are embedded and embodied in the world. This is often referred to as the sense of presence. Despite DSM and ICD, today several initiatives are underway to attempt to understand the onset and underlying aetiology of schizophrenia that ties the disorder directly to the basic structure of consciousness, presence and especially the categories of pre-reflective or minimal self. Much of this work starts from a recognition of the importance of the phenomenology of schizophrenia (Parnas, 2003; Parnas & Handest, 2003). An important line of this contemporary work represents the attempt to seriously take account of the subjective experience of patients with schizophrenia in understanding the disorder; hence the phenomenological orientation. One influential set of studies has focused especially on the preonset or prodromal stages – that is the pre-psychotic stages – of the disorder where patients who will go on to develop fullblown schizophrenia often report a variety of disturbed and altered experiences of self, presence, ability to think, and agency (Parnas, 2000, 2003; Parnas & Handest, 2003). Some of the most important of these current theorisations explore the link between anomalous self-experience, or self/world boundaries in the prodromal stages of schizophrenia (de Vries et al., 2013; Henriksen & Parnas, 2012; Sass & Parnas, 2003). Much of this work depends on theorizing schizophrenia, or at least the prodromal stages of schizophrenia, as a disturbance of the pre-reflective sense of self, (Parnas, 2003; Parnas, Bovet, & Zahavi, 2002; Sass, 2003; Sass & Parnas, 2003), or, what has been termed ipseity (Zahavi, 2005b). The term ipseity originally comes from the phenomenological tradition and was especially developed by the French phenomenologist

16 The experience of distorted presence here clearly recall the phenomenology of space-sickness discussed above.

250 

Robert Clowes

Michel Henry (1963)17. Parnas clarifies the term: “We may speak of ipseity or prereflective self-awareness whenever we are directly, noninferentially or nonreflectively conscious of our own ongoing thoughts, perceptions, feelings or pains; these always appear in a first-person mode of presentation that immediately tags them as our own, i.e. it entails an automatic self-reference” (Parnas, 2003, p. 224). Parnas continues that it is this dimension of ipseity that make experiences subjective. Thus, when ipseity is compromised in our experiences, thoughts and sensations, they can take on a variety of morbid symptoms which can be characterised as a distorted or diminished unity in our thoughts, body or sense of presence. It is sometimes also emphasized, that it is the self pole of presence that is particularly disturbed, or that the pre-experience of self (ipseity) fails to saturate experience (Parnas, 2003, p. 255). The notion of ipseity is sometimes used more or less interchangeably with the idea of a minimal (or core) self. The minimal self concept was first developed, or at any rate codified, in Gallagher’s (2000) paper although it was significantly influenced by Strawson’s ‘The Self’ (1999). Strawson had attempted to theorize the most basic sense of self conceivable, thus the minimal self. The minimal self – in Gallagher’s sense – is to be understood as hiatus-free, sense of self that is immediate and non-observational. Importantly, it shouldn’t involve “a perceptual or reflective act of consciousness” (2000, p. 15) and thus be definition is pre-reflective. For this reason the term “minimal self ” is often understood as a putative explanation of, or just treated as synonym of pre-reflective self consciousness. As will become clear I do not agree with this analysis in that it seems to be possible and indeed likely that more extended senses of self can also be treated as pre-reflective dimensions of experience. According to Cermolacce, Naudin & Parnas (2007) this minimal self is no mere conceptual entity but it also confers our basic sense of self or ipseity, and indeed it is claimed that it is this sense of self – perhaps more properly of self-world boundaries – which is interrupted in the early preonset (or prodromal) stages of schizophrenia. In light of this, schizophrenia is understood as a disorder where this basic sense of self, the role it plays in structuring subjectivity and in unifying mental life, becomes 17 For an excellent discussion of terms development, usage and application to the understanding of schizophrenia see (Zahavi, 2005b).

The Reality of the Virtual Self as Interface to the Social World 

251

altered in one of several ways. It is partly because it is the self pole, or the self / world relation that is generally thought to be affected in pro-dromal schizophrenia, that this approach is referred to as the ipseity disturbance (Sass, 2003), or self disorder, hypothesis (Sass, 2013, In Press)18. In fact there is now a great deal of research that shows that prodromal patients seem to have very marked and particular symptoms that up until now have been largely ignored by DSM and ICD, as indeed has the more general attempt to achieve a phenomenological characterisation of schizophrenia (Parnas, 2003). Prodromal patients often report a variety of altered and distorted experiences – especially self experience – from feelings that body parts are changing, to the feeling that aspects of the outside world are entering into oneself, from feelings of lack of agency, to a lack of bodily presence, to feelings that one’s thought is doubled, or somehow delayed (de Vries et al., 2013; Parnas, 2003; Parnas & Handest, 2003).19 These feelings and reports have marked similarities with the feelings of depersonalisation and derealisation that we have already discussed around DPD. However in addition to these alterations in the sense of presence, there are variety of other altered experience experiences including: altered sense of the body or its parts feeling heavier or lighter, smaller or larger, longer or shorter; alterations to the stream of consciousness especially where one’s thoughts appear to take on quasiautonomous character, or one’s thoughts, especially inner speech take on an object-like character; other phenomena include feelings of dissolution of the self or that the self is somehow doubled20; another problem 18 It should be noted however that as the disorder is theorized as routed in an experiential problem of the self world boundary, it may be better to think of it as a disorder of presence rather than of ipseity per se. See a related discussion in (Sass, 2013, In Press). 19 A recent attempt to catalogue and distinguish the different types of anomalous self-experience in prodromal schizophrenia and related conditions resulted in the Examination of Anomalous Experience or EASE scale, “a symptom checklist for semi-structured, phenomenological exploration of experiential or subjective anomalies that may be considered as disorders of basic or ‘minimal’ self-awareness.” (Parnas et al., 2005, p. 236). 20 “The most prominent feature of altered presence in the preonset stages of schizophrenia is disturbed ipseity, a disturbance in which the sense of self no longer saturates the experience. For instance, the sense of myness of experience may become

252 

Robert Clowes

is the apparent increased subjectivisation of the world, i.e. that reality is increasingly mind dependent; lack of volition and asociality (Parnas, 2003). One way of generalizing many of these experience is to say that “The most prominent feature of altered presence in the preonset stages of schizophrenia is disturbed ipseity, a disturbance in which the sense of self no longer saturates the experience.” (my emphasis: Parnas, 2003, p. 225) or at least those aspects of experience that should be saturated. One way of interpreting the problem is as a lack of coherence or simply a lack of – as Metzinger might say – a proper partion of self and world. That the disorder is typically theorized around the ipseity pole of experience perhaps reflects the fact that subjects are left feeling that aspects of their mental life are not really their own: the typical problem then appears to be one of under saturation. However, prodromal patients maintain a sense that these are ‘as if’ experiences, that is to say, their reality testing is relatively intact and they do not believe – as patients with chronic schizophrenia in full blown psychosis – that they “are the world”, or that “someone else is causing my thoughts”. Apart from these various disturbed senses of presence it is clear that other aspects of subjectivity, whether by direct implication, or as forms of compensation, also are transformed. One of the most interesting of these is what was generally referred to as hyperreflexivity (Sass, 2003, 2013, In Press; Sass & Parnas, 2003); and sometimes hyperreflectivity (Parnas, 2012)21. This is a sort of exaggerated self consciousness, or intense attention to one’s inner life, thoughts, or inner speech that interferes with one’s ability to think coherently and act in a timely manner. However, it should not be understood as merely an excessive form of introspection or reflection on one’s own thoughts (this is sometimes referred to as hyperreflectivity), but typically involves a “popping out” of internal phenomena, typically not attended to by healthy subjects (Sass et al., 2013). Patients with hyperreflexivity frequently subtly affected: one of our patients reported that his feeling of his experience as his own experience only ‘appeared a split-second delayed’” (Parnas & Handest, 2003, p. 225). 21 The terminology difference turns on Sass’s insistence that hyperreflexivity is not merely a sort of enhance introspection of self-reflection (Sass, Pienkos, Nelson, & Medford, 2013). For further discussion of the distinction see (Seigel, 2005).

The Reality of the Virtual Self as Interface to the Social World 

253

complain that their normal thinking processes are derailed by this excessive focus. It may be that hyperreflexivity is a sort of compensation for feelings of estrangement from one’s body and mind that are central to ipseity disturbance more generally. One analysis of hyperreflexivity is that typically transparent mechanisms of thought are actually becoming the subject of thinking process in their own right (hence opaque). What these subjective reports in the prodromal stages of schizophrenia seem to indicate is that something of the basic structure of thinking and consciousness is rupturing, or breaking apart. It may be that the problem here then is not simply with the saturation of the relevant parts of experience with ipseity, but with how the mind typically renders parts of its thinking processes as transparent. However we analyse this, hyperreflexivity tends to exacerbate the apparently more basic problem of the disruption in the sense of presence by interrupting and disrupting a patient’s ability to maintain a coherent and directed stream of consciousness. As the patient develops more full-blown schizophrenia, these distorted or altered pre-reflective senses of self “may be, and usually are, followed by the changes in the reflective I awareness and in the social self, thus resulting in the profound disturbances of identity, clinically manifest on all […] levels of selfhood.” (Parnas, 2003, p. 219). The idea that thought, like the body, may increasingly be less “saturated by self ” may offer a way into making sense of the basic phenomenology of schizophrenia, and perhaps can be extended to deal with some of the more social aspects of the disorder (we shall return to this point below). Such findings around the prodromal stages of schizophrenia are now crying out for a refined analysis of pre-reflective self consciousness, the integration and unity of the self and the role of self-consciousness in subjectivity and thought. Further attempts to purge the notion of self from our psychiatric taxonomy seem only likely to harm rather than aid the development of research into conditions like these conditions. Metzinger’s contention that the self concept should be eliminated is of no obvious help to those attempting to study and develop treatments for such conditions, especially when it is “the phenomenology of disturbed self-experience” that appears to be at the centre of what clinicians are trying to treat and theoreticians explain. Further moves to elimination appear far from “useful to a psychiatrist, engaged in describing mental

254 

Robert Clowes

states of his patients” (Parnas & Handest, 2003, p. 122). Conversely, Metzinger’s ideas of phenomenal self models and the virtual self may offer us new resources to productively rethink selves, self experience and their conditions of possibility. The development of refined tools to think about these issues is surely one of the main contributions of Being No-One, which makes Metzinger’s ontological orientation all the more puzzling. Indeed if the ipseity theorists are correct, what we need is a deeper understanding of ipseity, presence, their interrelations, and their role in theorising the structures in which thought and body-alienation can occur and Metzinger’s multi-level analysis give us powerful resources in this regard.

Schizophrenia and Social Self Diminishment Lysaker and Lysaker’s (2008) book Schizophrenia and the Fate of the Self offers an account of schizophrenia and its phenomenology. Their analysis links to the ipseity disturbance hypothesis in that the authors give great importance to the phenomenology of schizophrenia including, although not centring on, the transition from prodromal to the advanced psychotic phases of schizophrenia. They also attempt to understand the phenomenology and aetiology of schizophrenia in terms of disruptions, they use the term ‘diminishment’, of our pre-reflective sense of self, although, crucially not as a minimal sense of self as described above. Finally, they draw heavily upon the European phenomenological tradition in developing their analysis, although their main reference point is Heidegger, rather than Sartre, Henry and Merleau-Ponty22. This said, the account of self-diminishment they propose is quite different from the ipseity account, although as I will explore here, it may be possible to propose a more general account that captures some of the advantages of each. The starting point of the book is an attempt to pay careful attention to schizophrenia from the point of view of how it is experienced by 22 The main reference points of ipseity theorists, see (Zahavi, 2003).

The Reality of the Virtual Self as Interface to the Social World 

255

those who suffer with the condition, but while the phenomenological approach employed has many points of contact with the ipseity approach discussed above, Lysaker and Lysaker focus especially on how our social sense of self is compromised by schizophrenia. They focus on patients that speak of their sense of self being diminished because they find themselves unable to mentally inhabit aspects of their lives in relation to others, especially their inability to mentally inhabit customary social roles such as are necessary at work, in their everyday social interactions and in their intimate relations to others. Lysaker and Lysaker argue this diminishment of the sense of self can at once be understood as a failure to meet and inhabit the social roles that are expected of normally functioning human beings and the accompanying inner diminishment that to go along with this, or possibly precedes as the ipseity disturbance hypothesis holds. On the face of it, this is a very different analysis from disruptions of the self-pole in distortions of presence, or the disruption of inner life as basic mental unity become decomposed. Before discussing this matter further let us look in a little more detail about exactly what Lysaker and Lysaker claim about what they call the dialogic self. The concept of the dialogic self is developed by Lysaker and Lysaker to address one aspect of schizophrenia that afflicts many people in the advanced stages of the disorder, as patients: find themselves no longer able to function among others. Their approach aims to give a phenomenological analysis of changes that take place as the disorder develops from the prodromal phases of schizophrenia into its fullblown psychotic manifestation. They are especially interested in how patients apparently can no longer relate to the social roles and interests they once inhabited. Lysaker and Lysaker theorize these problems as an inability to inhabit what they call self-positions. Self-positions are basically situation or role specific existential orientations, or ways of being in the world in particular social roles. Self-positions simultaneously arise from the social roles we come to inhabit but also are means of disclosing the world through those social roles, or as Lysaker and Lysaker write: “self-positions arise out of the very life that they disclose.” Self-positions, they theorize, come in three types: characterpositions, meta-positions and organism-positions. The first kind character-positions, can be understood as social roles which we inhabit

256 

Robert Clowes

and through which the world is disclosed to us. Examples Lysaker and Lysaker use to illustrate this include: self-as-citizen; self-as-daughter; and self-as-teacher. It is not that all social roles constitute a self-position, but rather that: “A character-position involves a recurring action-orientation characteristic of one’s being in the world, e.g., citizen and lover.” (Lysaker & Lysaker, pp. 48–49). In each character position we inhabit, the sense of self we enact and the world that these positions disclose will be rather different. Character-positions, while open to reflection, generally are manifested pre-reflectively: they are “pre-reflective disclosures concerning action-orientations derived from social roles” (Lysaker & Lysaker, 2008, p. 54). That it is to say they structure our action and interactions with others without the need for us to reflect on them. Different possibilities and constraints of action will be afforded by each particular self-position we pre-reflectively inhabit. As we encounter a situation that implies we inhabit the self-position of self-as-teacher, or self-as-lover, we act and indeed perceive and interpret the world from the stance that particular self-position discloses. As normally functioning human beings we are able to inhabit and shift seamlessly and unconsciously between a variety of self-positions, essentially projecting and inhabiting them as we move between different social contexts. What should I do as a father, as a lover, as a teacher, as a clarinettist? Self-positions often provide us with an answer to such questions before they consciously arise. The particular, cares, joys and action possibilities that are disclosed to me as a father are distinct but they are also distinctly part of me and disclose my presence in the world in one of my various social aspects. Character-positions are situation and role specific pre-reflective ways of encountering the world from those disclosed by my other. But they are not minimal ones in the sense of Gallagher’s minimal self. They imply our worldly roles and they can be as socially rich as the roles which we inhabit. Crucially if Lysaker and Lysaker are right, they are also primary ways in which we pre-reflectively experience our sense of self in the world23.

23 If this line of thinking is correct it implies the ipseity theory as it is currently being developed may have a crucial lacuna. However it opens possibilities for the theory to explain what happens as prodromal stages of schizophrenia move to the more advanced stages. Unfortunately, a full development of the implications of these

The Reality of the Virtual Self as Interface to the Social World 

257

The second type of self-position is the meta-position. These are essentially character-positions plus some evaluative content. Examples would be self-as-good-teacher, self-as-mediocre-lover. Meta-positions also generally imply a reflective dimension so we might think of them as the elevation of our basic ways of modulating social interaction to a level where we can reflect and tune them. A third type of self-position is the organism position. Lysaker & Lysaker link these positions to Damasio’s (2000) core consciousness: “the pre-reflective monitoring of an organism’s being-in-the-world” (Lysaker & Lysaker, 2008, p. 53). It virtue of them being pre-reflective and concerned with the body’s self-monitoring it is natural to make the link here to the minimal self and the sense of ipseity discussed above24. Self-positions are thrown up by typical situations we face in the social world and the problems of action they throw out. They arise out of the striving within those roles but they are not inner essences, rather they “unveil time-slices in a life that is still unfolding” (Lysaker & Lysaker, p. 49). As Lysaker and Lysaker develop their theory it is central that we are not identical with any particular self-position but instead exist at the intersection, or dialogue, of different self-positions. Dialogue for Lysaker and Lysaker is crucial for it is what essentially generates us. Their understanding of self – the dialogic self – is so-named because it is to be understood that we exist and are disclosed as the dialogue between different self-positions. This idea hints at explanations to why we are sometime beset by, and cannot answer such as questions such as “Who am I really?”. According to the self-position theorist, you are the congruence, or dialogue between the different self-positions you are able to inhabit. That such questions are thrown-up will always evidence not just an inner struggle but the problems inherent in being somewhat theoretical formulations for our understanding of schizophrenia, are beyond the scope of this discussion. 24 Another possibility is to think of organism-positions, not as actual self-positions but more as the embodied basis from which other self-positions are projected. Lysaker and Lysaker suggest that “self-as-threatened” is an example of an organism position and why this makes sense as a particular temporary existential orientation it is less clear why it should be consider a self position. We do not need, I think, to posit a type of self for every situation or existential orientation the organism faces.

258 

Robert Clowes

plural beings in this way. We are not defined by an inner essence but by a plurality of sometimes competing self-positions. The central problem for the schizophrenic, according to self-position theory, is not that the self cannot form a stable self-position, but rather that it (typically) has become inflexibly fixed in a single self-position. The theoretical account of Schizophrenia on this analysis helps explain why it is so often been described in terms of a diminishment of self, for if the self is precisely the nexus of dialogues between different self-positions, then to the extent that the self becomes reduced to a single position, dialogue is reduced to monologue and our usual abilities to shift self-positions is curtailed or altogether undermined. Importantly, phenomenological reports of diminishment are explained in the following way. Self-positions endow us with a substantive pre-reflective sense of self that links what we do, and how we behave with who we take ourselves to be. When this everyday ability to inhabit self-positions becomes inhibited for whatever reason, we find ourselves diminished. Lysaker and Lysaker hold that the essential problem faced by the schizophrenic patient is of trying to relate to a variety of social situations for which they cannot find, or more appositely inhabit, appropriate self-positions. This, is according to Lysaker and Lysaker explains the phenomenology of diminishment; they do not feel themselves adequate to be in the situations in which they find themselves. It is also why so many schizophrenics are unable to operate in the social contexts, that prior to the onset of the disorder they faced unproblematically. One of the great advantages of Lysaker and Lysaker’s position is that it helps make sense of a variety of the claims made by schizophrenic patients and the phenomenology of schizophrenia, in particular the widely agreed phenomenology of diminishment which is a central – although under-discussed – aspect of the condition. Dialogic theory is in many ways constructed to, and deals with, the particular problems faced by schizophrenics in the social world. It focuses on how schizophrenia is not only a catastrophic break-down of an inner unity but of the ability to project a sense of self in the world. One might further link the approach to ipseity theory by postulating that just as ipseity has become unstable in other aspects of the subject’s experiential life, so problems

The Reality of the Virtual Self as Interface to the Social World 

259

have developed in the ability to invest self-positions with their typical ipseity25.

Self Positions: An Extended Virtual Self  ? I want now to make an additional claim that while developed within the spirit of Lysaker and Lysaker’s work, also marks an important theoretical departure. Lysaker and Lysaker hold that the phenomenology of diminishment can be explained in terms of the failure of dialogue among self-positions. They more or less assume that healthy subjects just possess self-positions which exist in dialogue with each other. While I certainly would not deny that there can be dialogue between self-positions, it may be our feelings of being a coherent self, or having a coherent sense of self, are not grounded in any particular inner unity that requires dialogue. It may rather be that our everyday experience of the plenitude of self, like the more general plenitude of perceptual consciousness is a rather virtual one. It is grounded in our generally fluid ability to engage and operate within socially adequate self-positions as the need arises in order to inhabit socially adequate roles. Pre-reflective self-experience is however not based on any deeper inner unity. In order to make this case in this and the next section I will draw together ideas about self-positions as just outlined with the ideas of virtual self we discussed in the earlier part of this paper. On this analysis the self-positions we (typically) unconsciously adopt, and the pre-reflective sense of self they engender, is self enough without further unification. While sometimes the potentially contradictory nature of the multiple self-positions we inhabit, may require us to engage in processes of further unification we might call dialogue – whatever that is – the worldly roles we inhabit, are generally enough to provide the needed degree of stability without the need for outright 25 This is just one very schematic attempt to unify self-position theory and ipseity theory. Really developing the points of contact between these theories and whether they are really compatible in explaining schizophrenic phenomenology is beyond the scope of this paper although it would be a highly valuable area to research.

260 

Robert Clowes

unification. We are relatively unified, or unified just enough in order slip between self-positions, but we are not in general very sensitive to the way that this movement implies that we face the world as rather different beings depending on the different self positions we inhabit26. This suggests a way of approaching the self-position account of self in virtualist terms. It is central to Lysaker and Lysaker that some unification of self beyond self-positions take place in dialogue. And yet the theory is rather sketchy about how this happens. One way of dealing with this problem is to argue that the unification of self-positions is itself a rather worldly matter. We seamlessly move between self-positions because the social world we inhabit is generally there to allow and support the transitions. This is not to say that there is no dialogue between self-positions, nor, when problems arise that there is not a need to mediate between our various social roles. However, insofar as we have a pre-reflective sense of self which is engendered by these self-positions it is not necessary that they strictly span any incoherencies, merely that we can continue to employ them as needed. By analogy with the enactive theory of perception that we discussed earlier, it may simply be that we are typically able to unproblematically apply, that is, inhabit the relevant self-positions as and when required and simply do not worry much that all of the roles we are able to inhabit do not overlap in entirely coherent ways. Dialogue between self-positions, insofar as it happens, may occur much more implicitly than the word dialogue seems to imply. That is to say that while we – who are not living through serious psychiatric difficulties – may pre-reflectively feel a sort of plenum of self, but that experience is based on our ability to project and inhabit self-positions as and when required. One theoretical advantage of this position is that we no longer have to explain any further unified self beyond self-positions, nor develop a much fuller account of what dialogue between self-positions actually entails; (something Lysaker and Lysaker’s formulation still appears to owe us). Our pre-reflective self-experience may appear to be a plenum in virtue of our ability inhabit different self positions as and when required, not 26 There are of course important similarities between this proposal and the proposals of Erving Goffman in The Presentation of Self in Everyday Life (Goffman, 1959). Unfortunately there is no space to pursue the various similarities and differences of approach here.

The Reality of the Virtual Self as Interface to the Social World 

261

on the manufacture of inner dialogues27. The apparent coherence of self may be a largely virtual matter depending on our ability to summon and deploy, as and when needed, a coherent self-position. This approach may also offer an important novel insight into some symptoms of schizophrenia. It may be that the development of fullblown schizophrenia happens as patients become increasingly aware of the typically ignored absences and incoherencies between self-positions. This may link to what Sass (2003) calls hyperreflexivity, i.e., a sort of exaggerated focusing on self as an object of attention rather than as a transparent medium through which we experience and engage with the world. Hyperreflexivity on this analysis is engendered precisely because the normally ignored virtuality of the self has become salient and a problem for the subject. This reconstruction of Lysaker and Lysaker’s idea of self-positions in terms of the grounded virtualist theory introduced above, allows us some useful theoretical innovations. Perhaps the most important is that it allows us to link the idea of a pre-reflective sense of self with our social nature. It is often taken (although this is by no means theoretically necessary) that the pre-reflective sense of self is identified with a minimal, typically rather biological sense of self. Such a view perhaps turns on the theoretical treatment that sees minimal selves and narrative selves as two theoretical poles. What I want to propose here is that our pre-reflective sense of self, at least phenomenologically speaking, is typically socially rich and inflected with a sense of who we are in the social world. It may be possible to draw out a conceptually separate minimal pre-reflective sense of self but this is not how experience confronts us. Let us take one example which may give some grounding to this theoretical contention. Imagine you are a teacher who has just made a terrible discovery. Your star student, whose work you have been showing off to colleagues for months, turns out to have plagiarized everything. You find this out after another colleague has raised a doubt and you run an essay through a plagiarism database. As the confirmation comes back you feel a sickness in the pit of your stomach. The feeling here is 27 It must be pointed out that the manner or mode in which dialogue is implemented intra or inter-subjectively is decidedly under theorized in Lysaker and Lysaker’s book.

262 

Robert Clowes

crucial and connects to our pre-reflective sense of self. But the feeling may not come from what Lysaker and Lysaker would call an organism position, but a character or more properly a meta-position: Self-asproud teacher. The feeling here is distinctly embodied – the pit in the stomach feeling – but entirely targeted at and suffused with one’s sense of self in a particular self-position. At the phenomenological level at least, the feelings are entirely fused and it appears as if it is impossible to at least phenomenologically dissociate a minimal pre-reflective sense of self from the wider, socially inflected senses of self which are given in the social roles we inhabit.

Rethinking the Virtual Self: Actualizing Virtualities There are many different types of virtual objects in our minds, implicated in our perception of the world, and even in the way things appear to us. Reflections in mirrors for example can be considered a type of virtual object, albeit in this case, they are mere appearances. However, according to Metzinger’s own proposal, the self, the PSM, is virtual in a special sense that it different from a mere appearance. The PSM operates as an interface that makes available a multimodal sense of the body as an integrated whole. In so doing it allows us to operate as a more coherent entity in space and time. PSMs, and thus the self, are what we might call an actualizing virtuality28. Actualizing virtualities are forms of projection that present coherent possibilities for the agent which can be made actual by taking the actions they imply or simply by inhabiting a stance that makes those action possible. If this is right, and insofar as they play a central role in our control architecture and also in the way we pre-reflectively experience ourselves in the world, they play a crucial role in our existence as agents. But PSMs are also central constituents of the control structure of creatures like us that are self-aware and have a sense of themselves enduring 28 NB., The terminology is mine, not Metzinger’s, although I think the idea is implicit in his concept of the PSM.

The Reality of the Virtual Self as Interface to the Social World 

263

over time. In virtue of this they must be considered a part of the real. The content of PSMs is virtual but the production of this virtual content plays a central role in making us what we are for ourselves, i.e., in conferring ipseity. But this I think is a clear problem, not an argument for Metzinger’s case that the selves are especially unreal, or their content illusory. Rather, PSMs are virtual in a special way. They are, or are at least at the core of, what makes it possible for us to know ourselves and to realize ourselves through our actions in the world. Metzinger himself is uncompromising about what this virtual nature of the PSM means for the existence of selves. He writes: No such things as selves exist in the world. At least their existence does not have to be presupposed in any rational and truly explanatory theory. Metaphysically speaking, what we called “the self ” in the past is neither an individual nor a substance, but the content of a transparent PSM. There is no unchanging essence, but a complex self-representational process that can be interestingly described on many different levels of analysis at the same time. (Metzinger, 2004a, p. 626)

But isn’t this enough? It is really only Metzinger’s prior assumption that the self should be an “unchanging essence” or nothing at all? While there is certainly precedent in the Western philosophical tradition for such a claim, from either the standpoints of scientific or folk psychology there is no real reason the self needs to be considered an unchanging essence in order for it to play the main conceptual roles required of it. Indeed we have seen above how the idea that ipseity, developed in the phenomenological tradition, in many ways play the sorts of conceptual roles that have been required of the self in much philosophical discussion. In making this claim, it is of course open to Metzinger to show that scientific and folk psychology require the self to be an individual, substance, or unchanging essence. But this he has not undertaken. The real question is whether we can plausibly reconstruct folk and scientific psychology around the idea of the virtual self and whether it is worth doing this. It certainly seems premature to abandon the idea altogether. In fact I think there are several plausible ways of reconstructing the concept. One resolution is offered by Marcello Ghin (2005) to the effect that we should identify the self with the organism that projects a PSM. I think this suggestion has much to recommend it, including offering a plausible way of reconstructing folk psychology. In fact Metzinger

264 

Robert Clowes

concedes as much in his (2006) response to Ghin’s critique, and also his responses to (Hobson, 2005). Another plausible (and to me more interesting) way of reconstructing the self is to agree that it is indeed virtual in many of the senses that Metzinger claims, but this is enough for the conceptual purposes we require of the concept. The self then, at its root, is a form of pre-reflective experience, very much as is claimed by those accounts of pre-reflective sense of self developed in neo-phenomenological accounts by Zahavi and Gallagher, but capacious enough to include the forms of pre-reflective self-experience discussed by Lysaker and Lysaker. I want to embrace this analysis while claiming that in important respects our pre-reflective sense of self is indeed virtual. Selves on the revised analysis presented here are a form of content produced by an interface that determines central dimensions of our presence in the world, especially, the social world. On this analysis it is true we are not selves in a strict sense but we possess, or better, are in touch with ourselves at a pre-reflective level. My claim here – contrary to the idea that our pre-reflective sense of self need be minimal – is that it is rather a more expansive and socially inflected sense of being. We do not have to identify our pre-reflective sense of self with the mere actual registration of the body. We can instead think of the pre-reflective self as virtual, projective, and while perhaps based in body representation, going significantly beyond the body. Insofar as self is composed of virtual social (self-position) projections over bodily content, it can link up the idea of a pre-reflective and indeed embodied sense of self, with the idea that our pre-reflective sense of self is not minimal at all, but irreducibly social. The virtualist idea of self as developed here through the prism of self-positions, gives us an alternative which links our social nature to the structure of our experience.

The Virtuality of the Self and its Ontological Status Metzinger’s search for the self is hardly exhaustive. He more or less assumes a very classical – even Cartesian – notion of the self as a sort

The Reality of the Virtual Self as Interface to the Social World 

265

of inner essence. He then finds that this notion is unsustainable in part because the virtual self generated by phenomenal self-models cannot be reduced to such a hypothetical inner essence. But given the account just offered we can note that there are other possibilities. On the virtualist version of self-position theory, selves are not inner essences but rather a pre-reflective capacity to inhabit multiple self-positions. And self-positions are crucially worldly things. They come from our abilities to interact with – in the terminology to interface – with affordances of the social world. Crucially this does not render selves unreal in any sense that should cause a problem for science nor for much of folk-psychology. Rather it gives us a way of making sense of these at both a neuroscientific and social level of important aspects of selfexperience. The self can be identified with certain forms of virtual content; forms which can be understood as projections from an embodied basis and that we typically apprehend pre-reflectively in experience. But while our pre-reflective sense of self may be body based and body inflected it is also substantially composed, in any particular case, of self-positions, i.e., particular social roles that we inhabit as persons. These come with attitudes, responsibilities and are enforced by social expectations. We inhabit these self-positions pre-reflectively and so they can be linked – like our embodied sense of self – to an intrinsic sense of subjectivity. Self-positions are worldly, but they are not minimal. I claim that the self-positions are another form of virtual content, but that Theorizing pre-reflective self-experience in this way allows us to make sense of our worldly experience. That is, not just as bodies attempting to integrate coherent motor action, but as social beings attempting to live coherent lives. The virtual self on this analysis is not best identified with a projected coherent body image, but with projected whole social being or presence which we use to inhabit the social world. When our ability to inhabit this social presence fails, as may happen for a variety of reasons, our ability to function as coherent human beings itself can become deeply undermined. Let us then summarise how we have rethought the idea of a virtual self. To recap Metzinger argued the sense of self was virtualist in the following respects:

266 

Robert Clowes

1. Selves are virtual because they are simulations; or more precisely, simulational content. 2. Selves are virtual in that they are constructs arising from partitions in representational space that do not necessarily respect any divisions in reality. 3. Selves (PSMs) are virtual – by analogy with virtual reality systems – because they can be regarded as control interfaces that make available (and obscure) certain possibilities for action. 4. Like the interfaces in virtual reality devices, selves (PSMs) convey a sense of presence. 5. The contents of PSMs are also virtual because they tend not to correspond to, or depict, actual states of affairs in the world, but best-guesses or mere possibilities about the state of the body. I believe self can be considered virtual but in the following revised senses: 1. Selves are virtual because they are actualized projections into the social world. 2. Selves are virtual in that they are constructs arising from partitions in representational space and arise from respecting the (also virtual) social roles that we inhabit as individuals. 3. Selves (PSMs) are virtual – by analogy with virtual reality systems – because they can be regarded as control interfaces that make available (and obscure) certain possibilities for action. The particular possibilities for action they track, however, are primarily in the social world of interpersonal intercourse. 4. Like the interfaces in virtual reality devices selves (PSMs) convey a sense of social presence disclosed by the self-positions we inhabit. 5. The content of PSMs are also virtual because they tend not to correspond to, or depict, actual states of affairs in the world, but idealisations and projections of the roles we inhabit in the social world. The hypothesis here is that our ability to inhabit and project self-positions provides much of the pre-reflective sense of self which allow us to operate as coherent social beings. Yet the ability to project an overall coherent sense of self is nevertheless virtual. This is not to deny that dialogue between

The Reality of the Virtual Self as Interface to the Social World 

267

self-positions might sometimes be necessary, nor that there are not other internal and worldly processes that help us integrate a coherent sense of self  29. But in general we have a rich pre-reflective sense of self because of our abilities to fluidly inhabit the social world, not because we can always access the very same sense of bodily self anchored in the body. There are two central implications for the theory of self as thus conceived. First, the theory offered expresses continuity with traditional theorizing that made the explanation of self, in this case reconceived around the idea of virtual self-positions, central to the understanding of certain forms of psychiatric illness. There is nothing unique here as many other psychiatric theories also posit the self as a central explanatory unit. Insofar as Metzinger argues that there will be no explanatory loss if we jettison the self we need to understand how its replacement offers us any explanatory advantage. Metzinger, although he gives extended analysis of many psychiatric illnesses, is practically silent on the question of how the PSM would help explain a central aspect of schizophrenia, namely the phenomenology of diminishment. And yet if the account here is on the right lines we have a good way of dealing with this problem. If the projection of the PSM is virtual in the sense of us experiencing a plenum when what we really do is inhabit self-positions, a reconceived PSM can help us explain this aspect of schizophrenia. Schizophrenics feel diminished because their abilities to express, and inhabit self-positions are diminished, because their interface to the social world through the projection of multiple appropriate PSM is, for whatever reason, curtailed and crucially this must compromise both their ability to operate as social beings and their sense of who they are. In terms of psychiatric explanation, if on no other grounds, the elimination of self seems premature. The second implication is how congenial a revised self-position based theory is to non-eliminativist approaches to self. Let us cast our minds back to the discussion of Dennett’s virtual presence discussed at the beginning of this paper. On this approach perceptual richness can be considered to be based on a lack of epistemic hunger about certain internal informational absences. Many of these absence can be 29 NB – We should not overlook here how our sense of self is constrained by an array of folk-psychological practices which help to produce being that are more unified and have the greater appearance of being more unified (Zawidzki, 2008).

268 

Robert Clowes

made good by the ability to lean on the world when necessary in order to bring in information as required; information that is typically left in the ambient environment. In similar ways we experience our social presence in the ways it is disclosed to us through the self-positions we inhabit. For we human beings at least, the body may be the ground for projecting a self among others. But if the claims here are correct the expectations that synthesize a self are derived from a social realm. The self is centrally an instrument produced by our brains in order to minimally integrate our actions as an individual agent and more expansively in the multiple social relationships in which we are involved. Its unity may be more projected and inhabited as a result of our abilities in the social world rather than based on any particular unities in the brain. This conception has several predictions. The self as such can be dissolved or disturbed in several ways, but to roughly dichotomise, we could say this can happen in top-down or bottom-up ways. Bottom-up ways in which the self may be compromised include when bodily content is distorted, attenuated or otherwise no longer able to be integrated or synthesized into a coherent construct (e.g. Parnas, 2003). But it may also be compromised in a top-down manner where bodily content cannot be integrated into to the currently active set of social expectations. Our sense of self may become problematic for us in multiple ways and there may be multiple routes to the forms of diminishment of self we find in schizophrenia. But does the virtuality of self imply its non-existence, or the necessity of an elimination of its use in science? Only if one adopts an extremely traditional – namely Cartesian – view of the self as an internal substance or immaterial spirit. Even reducing the self to a PSM-like partition within the representational space and grounded in the homeostatic processes of the body is not obviously a way of rendering the self inexistent. Rather it is a way of reducing the self to a particular sort of representational content / process involved in maintenance of the agent30. As I have begun to argue, such a picture is radically incomplete 30 As this content is playing an ongoing role in controlling the action profile of the system it might be regarded as an actualising virtuality. A content produced by certain bio-systems to project presence in order to play a particular role in the unification of action; or as Metzinger puts it, “a real-time world-model that can be

The Reality of the Virtual Self as Interface to the Social World 

269

in that it does not supply an explanation of the apparent richness of pre-reflective self experience. The version of self-positions defended here is both able to make sense of how embodiment enabled self, but also why the idea of the minimal self never seems to have a enough substance to explain central problems of the self such as personal identity or even the question ipseity, that is, what really makes us ourselves. But as I have argued, the self is best conceived of as an intrinsically social, or worldly, virtual projection. The virtualist view of self presented here, far from implicating the need to eliminate the self, invites us to investigate further how it might be grounded in neurologically describable systems and the social processes we inhabit. So is this a reduction or elimination of the self as traditionally conceived? The self as conceived here is a socially targeted construct – a virtual projection – anchored in bodily representation but projected through a rich set of expectations about our social roles and synthesized by an underlying sense of coherence and structure. Our pre-reflective sense of self, it is in large part not constituted by our basic sense of presence, but in our social self-positions that largely give us our identity.

Actualizing Virtualities and the Conceptual Role of Self In this article I have claimed: that the self can be understood as a virtual construct whose core is bodily content but the principle function of which is to allow us to inhabit the social world. On this model the pre-reflective experience of self derives from the way the experienced body is projected into a social world. The character of this experienced self is not limited to natively endowed expectations and modes of integration. Rather it is always an interactive construct, drawing much of viewed as a permanently running online simulation allowing organisms to act and interact” (2004a, p. 104). This looks like a way, as Damasio argued, of giving the self some substance rather than suggesting its elimination, as Metzinger contends. There is little reason to endorse elimination if ‘virtual selves’ can confer the right sorts of unity and integrity that we would expect from the traditional use of the concept. The contention is that the virtual self can play the roles required.

270 

Robert Clowes

its character from the social world, especially the self-positions we are required to articulate in order to exist as social beings. A great advantage of this version of the virtualist view is that it does not imply any identity between the self and the body, or imply that minimal self is somehow the real or more authentic self  31. Rather we can – following Lysaker and Lysaker – highlight how the retreat into a more fixed or minimal sense of self can be a dramatic problem for the human subject. As Lysaker and Lysaker’s analysis of schizophrenia makes abundantly clear, the failure of an individual’s ability to project self-positions consistently and appropriately undermines or diminishes one’s sense of self. The virtualist scheme requires the projection of socially refined senses of self around an embodied core. Thus we can understand these projections as extensions of the core rather than narrative fictions (Dennett, 1992). A further potential advantage of this approach is that self is identified neither with the body, nor with groundless narratives, but that the two poles are unified32. This is not to claim the virtual self is not embodied. The self may be grounded in the body in multiple ways including the production of homeostatic content that is at the heart of one important embodied theory of self (Damasio, 2000)33. But it is given structure and definition through the process of virtuality projection. Does this virtualist conception of the self imply compelling reasons to eliminate the self-concept as we inherit it from the Western tradition? I think not. Instead the self, somewhat conceptually revised perhaps, still seems to be a necessary concept in understanding the constitution of agents like us. On the contrary, the revised account helps us see how the self can be virtual, while still being a proper part of the real and central to the operation of agents like us. Selves 31

It is worth noting in passing that there may be, indeed almost are, other ways in which we can be said to have or be selves. This article is really groundwork in a theory of self which would have to be incorporated into a wider network of theories and explanation (Gallagher, 2013). There may be other forms of self than discussed here. 32 The basic tension between these two ways of seeing things is concisely developed in (Gallagher, 2000). 33 In fact I think aspects of Damasio’s approach highly congenial to the idea of the virtual self presented here. Unfortunately making these links goes beyond the scope of the current discussion.

The Reality of the Virtual Self as Interface to the Social World 

271

are best not conceived of as Cartesian essences but this does not imply their inexistence. Once the content of selves is thought of as tied up with inhabiting the social realm their necessity becomes apparent. Selves are real, indeed they are actualizing virtualities and that is good enough – at least provisionally – for them to play the required role in our conceptual scheme.

Acknowledgements I want to express my heartfelt thanks to the Cognitive Foundations of Self Group at the IFL, New University of Lisbon especially the long-suffering editors of this volume Jorge Gonçalves and João Fonseca for their extraordinary forbearance in allowing me to revise this text long after it was due. Special thanks also go to Shaun Gallagher who first suggested I should look at Lysaker and Lysaker’s book as a way of developing my ideas about schizophrenia. Dina Mendonça and Michael Baumtrog also deserve special thanks for commenting on advanced drafts of this paper. Many thanks also to other members of the Cognitive Foundations of Self Group to whom I presented several versions of this paper, especially Vera Pereira, Inês Hipólito and Caio Novais, and to Sean Bell for helping me copy-edit several versions of it. All remaining problems are of course my own. Although this paper was begun while I was a tutor at the University of Sussex its development was supported by a grant from Portuguese Foundation for Science and Technology grant (SFRH/BPD/70440/2010).

References Barry, P. & Phillips, T. (2001). Mixed up in Space. Science@ NASA, National Aeronautics and Space Administration. Botvinick, M. & Cohen, J. (1998). Rubber hands ‘feel’ touch that eyes see. Nature, 391(6669), 756–756.

272 

Robert Clowes

Brooks, R. (1991). Intelligence without Representation. Artificial Intelligence, 47, 139–160. Brugger, P., Kollias, S. S., Müri, R. M., Crelier, G., Hepp-Reymond, M. C. & Regard, M. (2000). Beyond re-membering: Phantom sensations of congenitally absent limbs. Proceedings of the National Academy of Sciences, 97(11), 6167. Cardinali, L., Brozzoli, C., & Farne, A. (2009). Peripersonal space and body schema: two labels for the same concept? Brain topography, 21(3–4), 252–260. Cermolacce, M., Naudin, J. & Parnas, J. (2007). The “minimal self ” in psychopathology: re-examining the self-disorders in the schizophrenia spectrum. Consciousness and cognition, 16(3), 703–714. Clark, A. (2008). Supersizing the Mind, Oxford: Oxford University Press. Clark, A. (2012). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behav. Brain Sci, 1–86. Clowes, R. W. & Chrisley, R. (2012). Virtualist Representation. International Journal of Machine Consciousness, 04(02), 503–522. Dainton, B. (2006). Stream of consciousness: Unity and continuity in conscious experience: Taylor & Francis US. Damasio, A. R. (2000). The Feeling of What Happens: body, emotion and the making of consciousness,London: Vintage. de Vries, R., Heering, H. D., Postmes, L., Goedhart, S., Sno, H. N. & de Haan, L. (2013). Self-Disturbance in Schizophrenia: A Phenomenological Approach to Better Understand Our Patients. The Primary Care Companion for CNS Disorders, 15(1). Deleuze, G. (1991). Bergsonism (H. Tomlinson & B. Habberjam, Trans.). New York: Zone Books. Dennett, D. C. (1991). Consciousness Explained, Boston: Little Brown. Dennett, D. C. (1992). The Self as a centre of Narrative Gravity. In F. Kessel, P. Cole & D. Johnson (eds.), The future of folk psychology, Cambridge: Cambridge University Press. Dennett, D. C. (2001). Surprise, surprise. Behavioral and Brain Sciences, 24(05), 982–982. Frank, M. (1991). Selbstbewußtsein und Selbsterkenntnis: Essays zur analytischen Philosophie der Subjektivität. Stuttgart: P. Reclam.

The Reality of the Virtual Self as Interface to the Social World 

273

Gallagher, S. (2000). Philosophical conceptions of the self: Implications for cognitive science. Trends in Cognitive Sciences, 4(1), 14–21. Gallagher, S. (2005a). How the body shapes the mind. New York: Oxford University Press. Gallagher, S. (2005b). Metzinger’s Matrix: Living the Virtual Life with a Real Body. PSYCHE, 11(5), 5. Gallagher, S. (2013). A Pattern Theory of Self. Frontiers in Human Neuroscience, (7). Gallagher, S., & Zahavi, D. (2012). The phenomenological mind. New York: Routledge. Ghin, M. (2005). What a self could be? PSYCHE, 11(5). Goffman, E. (1959). The Presentation of Self in Everyday Life. Garden City, NY. Grush, R. (2004). The emulation theory of representation Motor control, imagery, and perception. Behavioral and Brain Sciences, 27(3). Henriksen, M. G. & Parnas, J. (2012). Clinical manifestations of self-disorders and the Gestalt of schizophrenia. Schizophrenia bulletin, 38(4), 657–660. Henry, M. (1963). L’Essence de la Manifestation. Paris: Presses Universitaires de France. Hobson, A. (2005). Finally Some One. PSYCHE, 11(5), 5. Johnson-Laird, P. (1983). Mental models: Towards a cognitive science of language, inference and consciousness. Cambridge MA: Harvard University Press. Lakoff, G. & Johnson, M. (1999). Philosophy in the Flesh. New York: Basic Books. Legrand, D. (2007). Pre-reflective self-as-subject from experiential and empirical perspectives. Consciousness and cognition, 16(3), 583–599. Lombard, M. & Ditton, T. (1997). At the heart of it all: The concept of presence. Journal of Computer Mediated Communication, 3(2). Lysaker, P. H. & Lysaker, J. T. (2008). Schizophrenia and the fate of the self: Oxford University Press, USA. Maravita, A. & Iriki, A. (2004). Tools for the body (schema). Trends in Cognitive Sciences, 8(2), 79–86. Medford, N., Sierra, M., Baker, D. & David, A. S. (2005). Understanding and treating depersonalisation disorder. Advances in Psychiatric Treatment, 11(2), 92–100.

274 

Robert Clowes

Metzinger, T. (2000). The subjectivity of subjective experience: A representationalist analysis of the first-person perspective. In T. Metzinger (ed.), Neural Correlates of Consciousness (285–307). Cambridge, MA: MIT Press. Metzinger, T. (2004a). Being No One: The Self-Model Theory of Subjectivity. Cambridge, MA: Bradford Book. Metzinger, T. (2004b). The Subjectivity of Subjective Experience: A Representationalist Analysis of the First-Person Perspective. Networks, 3(4), 33–64. Metzinger, T. (2006). Reply to Ghin. In Self-Sustainment on the Level of Global Availability. PSYCHE, 12(4). Metzinger, T. (2009). The Ego Tunnel: The Science of the Mind and the Myth of the Self. New York: Basic Books. Metzinger, T. (2011). The no-self alternative. In The Oxford Handbook of the Self, 279–296. Minsky, M. (1985). Society of mind. New York: Simon and Shuster. Noë, A. (2002). Is the Visual World a Grand Illusion? Journal of Consciousness Studies, 9(5–6), 1–12. Noë, A. (2004). Action in Perception. Cambridge, MA: Bradford Books, MIT Press. O’Regan, J. K. (2002). The Grand Illusion and Sensorimotor Approaches to Consciousness. Journal of Consciousness Studies, 9(5–6). O’Regan, J. K. & Noë, A. (2001). A sensorimotor account of vision and visual consciousness. Behavioral and Brain Sciences, 24, 939–973. Parnas, J. (2000). The self and intentionality in the pre-psychotic stages of schizophrenia. Exploring the self: Philosophical and psychopathological perspectives on self-experience, 115–147. Parnas, J. (2003). Self and schizophrenia: a phenomenological perspective. The self in neuroscience and psychiatry, 217–241. Parnas, J. (2012). The core Gestalt of schizophrenia. World Psychiatry, 11(2), 67–69. Parnas, J., Bovet, P. & Zahavi, D. (2002). Schizophrenic autism: clinical phenomenology and pathogenetic implications. World Psychiatry, 1(3), 131.

The Reality of the Virtual Self as Interface to the Social World 

275

Parnas, J. & Handest, P. (2003). Phenomenology of anomalous selfexperience in early schizophrenia. Comprehensive psychiatry, 44(2), 121–134. Parnas, J., Møller, P., Kircher, T., Thalbitzer, J., Jansson, L., Handest, P., et al. (2005). EASE: examination of anomalous self-experience. Psychopathology, 38(5), 236–258. Pezzulo, G. (2008). Coordinating with the future: the anticipatory nature of representation. Minds and Machines, 18(2), 179–225. Radovic, F. & Radovic, S. (2002). Feelings of unreality: A conceptual and phenomenological analysis of the language of depersonalization. Philosophy, Psychiatry, & Psychology, 9(3), 271–279. Ramachandran, V. S. & Blakeslee, S. (1998). Phantoms in the brain. New York: Harper Collins. Ratcliffe, M., Slaby, J., Manzotti, R., Stephan, A., Holzhey-Kunz, A., et al. (2008). Feelings of being. Oxford: Oxford University Press. Revonsuo, A. (1995). Consciousness, dreams and virtual realities. Philosophical Psychology, 8(1), 35–58. Revonsuo, A. (2006). Inner Presence: Consciousness as a Biological Phenomenon. Cambridge MA: MIT Press. Sass, L. A. (2003). Self-disturbance in schizophrenia: hyperreflexivity and diminished self-affection. In T. Kircher & A. David (eds.), The self in neuroscience and psychiatry (Vol. 870539117): Cambridge University Press Cambridge. Sass, L. A. (2013, In Press). Self-Disturbance and Schizophrenia: Structure, specificity and pathogenesis (Current Issues, New Directions). Schizophrenia Research. Sass, L. A. & Parnas, J. (2003). Schizophrenia, consciousness, and the self. Schizophrenia Bulletin, 29(3), 427–444. Sass, L. A., Pienkos, E., Nelson, B. & Medford, N. (2013). Anomalous self-experience in depersonalization and schizophrenia: A comparative investigation. Consciousness and cognition, 22(2), 430–441. Seigel, J. (2005). The idea of the self: Thought and experience in Western Europe since the seventeenth century: Cambridge University Press. Seth, A. K., Suzuki, K. & Critchley, H. D. (2011). An Interoceptive Predictive Coding Model of Conscious Presence. Frontiers in Psychology, 2.

276 

Robert Clowes

Sierra, M., Baker, D., Medford, N. & David, A. S. (2005). Unpacking the depersonalization syndrome: an exploratory factor analysis on the Cambridge Depersonalization Scale. Psychological Medicine, 35(10), 1523–1532. Sierra, M. & David, A. S. (2011). Depersonalization: a selective impairment of self-awareness. Consciousness and cognition, 20(1), 99–108. Strawson, G. (1999). The Self. In S. Gallagher & J. Shear (Eds.), Models of The Self ( 1–24). Exeter: Imprint Academic. Taft, M. (2012). An Interview with Thomas Metzinger: What Is the Self? Retrieved 04 October 2013, from . Zahavi, D. (2003). Phenomenology of self. The Self in Neuroscience and Psychiatry, 56–75. Zahavi, D. (2005a). Being someone. Psyche, 11(5), 1–20. Zahavi, D. (2005b). Subjectivity and Selfhood: Investigating the First-Person Perspective. Cambridge, MA: MIT Press. Zawidzki, T. W. (2008). The function of folk psychology: mind reading or mind shaping? Philosophical Explorations, 11(3), 193–210.

Conceptual Personae of the “attentional self ” Alexander Gerner1

1. Introduction “Everything has to be rethought from the beginning” Mr. Palomar (Calvino 1985, 115)

In this paper I take attention as a constitutive ground of the self. I critically survey the claim of Metzinger that a subjective self, defined as the centre of awareness, is the possibility of being able to manipulate the focus of attention, thereby stabilizing subjective experience. Thus I propose an attentional self in which I will put Metzinger’s thesis of the “control of the focus of attention” and the resulting notion of the “attentional self ” critically into perspective by approximating the concept of the self by means of conceptual personae of the “impossible” attentional self in Paul Valery’s dyadic conceptual personae “Monsieur Teste”/ “Émilie Teste” and the “heautoscopic” attentional self in Italo Calvino’s “Mister Palomar”. Philosophical Concepts, according to Gilles Deleuze & Felix Guattari in “What is philosophy?”(WP) can be personae – as explicitly as the appearing of “Socrates” for Plato or “Dionysus” and “Zarathustra” for Nietzsche- that act from a constitutive plane of philosophy and create concepts on this plane2 at the same time. What 1

2

Affiliation: University of Lisbon, Center for Philosophy of Science (CFCUL), FCT Post-Doc SFRH/BPD/90360/2012, , . Member of the Research Project: “Cognitive Foundation of Self ” IFL FCSH-UNL Project Fund: PTDC/FILFCI/110978/2009. “There are innumerable planes, each with a variable curve, and they group together or separate themselves according to the points of view constituted by personae. Each persona has several features that may give rise to other personae, on the same or a different plane: conceptual personae proliferate.” (WP, p. 76).

278 

Alexander Gerner

Deleuze & Guattari call “A conceptual personae […] thinks in us”3 (WP, p. 69) is thus more complex than a specific representational model of thinking or of a represented narrative self in symbolic representation. Therefore we have to underline the importance of conceptual personae for the creation of different points of view by these personae that direct thoughts concerning concepts such as “self ” and “attention” by giving these concepts not just certain names but also providing concrete concepts to consider, putting these conceptual personae into theoretical and practicable perspectives and into coherent conceptual constellations with the phenomenon of the self. Going towards the “self ” and back over conceptual personae, the “self ” becomes a diagrammatic feature on superimposed conceptual maps4 of concrete personae of the “self ”. Thus to state the first point: 1) Complex concepts such as the “self” have personae with which we think. With these personae of thinking we create plural conceptual maps with which we actively explore the complexity of the phenomenon. The personae of the self may be extended into a conceptual space, but are always present as concepts that show their perspectives, connecting lines, strengths and weaknesses as well as their lines of flights by being superimposed on to and by others. The concept of the 3 4

Hereby the expression “personae” is used in the sense of different instances of one being and “personas” as different beings. “Maps […] are superimposed in such a way that each map finds itself modified in the following map, rather than finding its origin in the preceding one: from one map to the next, it is not a matter of searching for an origin, but of evaluating displacements. Every map is a redistribution of impasses and breakthroughs, of thresholds and enclosures, which necessarily go from bottom to top. […] Maps should not be understood only in extension, in relation to a space constituted by trajectories. There are also maps of intensity, of density, that are concerned with what fills space, what subtends the trajectory. […] A list or constellation of affects, an intensive map, is a becoming. […] We see clearly why real and imaginary were led to exceed themselves, or even to interchange with each other: a becoming is not imaginary, any more than a voyage is real. It is becoming that turns the most negligible of all trajectories, or even the fixed immobility, into a voyage; and it is the trajectory that turns the imaginary into a becoming. Each of the two types of maps, those of trajectories and those of affects, refer to each other” (Deleuze, 1998, p. 63.)

Conceptual Personae of the “attentional self ”

279

“self ” is seen here as a relational entity and as such can be oriented within a diagrammatic map5 of concepts. The self might be superimposed on to the concept of the “other”, or in distinction to the “Non-self ” (as in Buddhist philosophy and psychology; see: Olendzki 2010, Engler/Fulton 2012) or even be integrated or opposed towards a representational “self-model” (Metzinger) or a virtual, “disembodied” entity of a “mind”. Another way to think the self is to think of it as basically embodied in between the body-image and the body schema (Gallagher, 1995) or described by a double aspect of an basic bodily embodied self and a personal, reflexive or extended self6 (see: for example Fuchs 2012b). In the terms of Thomas Fuchs7 there is a fundamental second-person perspective in the development of the self and thus in the development of the attentional self (from simple8 to dyadic attention) as well, suggesting a central 5 On diagrammatic thinking and the map see: Gerner 2010a. 6 The extended, reflexive or personal self, according to Fuchs (2012b) is characterized by the aspects of (a) capacity of taking on the perspective of the other (Perspektivenübernahme), (b) the capacity of introspective or reflexive self-consciousness, (c) a capacity of verbalization of its own experience, (d) the creation of coherent narrative Identities and (f) a self-concept by obtaining conceptual and autobiographic self- knowledge. 7 The primary experience of self -in Fuchs account- is not a pure self-experience, but also includes a sense-motor relationship between the experiencing subject and the environment, – often expressed in spontaneous agency and in action-perception cycles – which is mediated by the body and its habitual capacity and later on leads to social inter-corporeity. Available through the senses, members and capacities the body is embedded in the surrounding space that presents itself to it as a field of possibilities and valences. By this structural pairing of the lived body and its complementary surrounding, the basic self becomes an embodied spatial “ecological self” (Neisser, 1988). This is expressed by the situatedness and attunement of the lived body as well as by the feeling of body related to potentialities of a situation, and specifically by bodily sensations implicit in world-directed attention. Thus a basic feature of the attentional self is a) the body as zero-point of spatial orientation and attentional coordination, b) sensory-motor cycles, the sensory-motor coupling with the environment and a basic bodily being-towards-the world and c) the experience of flexible enacted and transgressed boundaries of the lived body. 8 “From birth on, infants are attentive to external entities (call this ‘simple attention’) and are engaged in dyadic self-other interactions which involve dyadic attention where subjects are mutually attending to each other. Later, when the

280 

Alexander Gerner

ecological conceptual map (I-me [1st PP] & you-me [2nd PP]) of the self and its conceptual personae that then is able to socially share attention and intentionality (Tomasello & Carpenter, 2007). This developmental account of an attentional self, based on early mutual attentional engagements, are crucial for later shared triadic attentional relations (Me-You-object), which then become important for developing complex joint actions (Frith & Gallagher, 2012). These views will later on be explored in slightly different way in the responsive account of the dyadic self in between the conceptual personae of “Monsieur Teste” and his wife Emilie. Each superimposed conceptual map of the self has to show how productive it acts and engages in the clarification of the phenomenon of the self. I wish to propose to superimpose the concept of the “self ” on to the concrete working of “attention” (see Gerner PhD thesis 2012 upcoming) and propose to show how conceptual personae of attention, although being realistically “impossible” as in the case of a purely and isolated episodic attention function or a merely disembodied heautoscopic attention, nevertheless can help to elucidate how to conceive the “self ” by its relations to the phenomenon of attention. The officially “non-philosophical” fictional cases of “Monsieur Teste” (Paul Valéry) and “Mister Palomar” (Italo Calvino) can produce insights into the complex philosophical concept of the “self ” – in our case of the self as an “attentional self ” – before reviewing the philosophical position in Thomas Metzinger’s account of the self as “attentional self ” that in his account has the main function of stabilizing subjective experience by controlling the attentional focus. The possible/impossible and aesthetic/ heautoscopic notions of the self relate to the constitution of modes of appearance and disappearance of a bodily episodic self, that seems more fundamental than a narrative or reflexive conceptual self. Thus an “attentional self ” seems impossible as a permanently and constantly alert “hunter” that is permanently “on“ as in the ideal of a perfectly and permanently alert man. Attention as I see it – pertains to a double movement in between “waking up” infant begins to follow the gaze of the other person, it may occasion a new form of attention (call this ‘shared attention’): the infant is aware of the adult being attentive towards the object and of herself being attentive towards the object” Fiebrich & Gallagher (2012, p. 7).

Conceptual Personae of the “attentional self ”

281

and “falling asleep” and therefore is exposed to modes of fatigue and change, and bound with it seems the notion of the self in the sense of an attentional self, that neither might be permanently “on”, although it might as well not be entirely “episodic” as conveyed by Strawson (2004) also make part of a diachronic temporal instances. Alertness has been studied in empirical psychology and cognitive (neuro-) science, being termed “vigilance9“ or “sustained attention”, to hold on to something for a longer amount of time, as distinct from the shorter rhythmic structure of spatial orientation in turning towards and turning away in attention. This vigilance function of attention has to be distinguished from the conceptual personae of the “hunter” as the attentively alert self, that – as Ortega y Gasset (2007) underlines – is not necessarily programmed for successful hunting, by catching the prey10, 9

Vigilance was first studied on the empirical level by the neurologist Henry Head (1923) in patients with brain injuries in the 1920s. Norman Mackworth (1948, 1950/61) started the “systematic study of vigilance during World War II in human factors”(Warm et al., 2008, 433). Later Parasuraman (Parasuraman, 1976; Parasuraman & Davis, 1977; Parasuraman, 1986; Parasuraman et al., 1986) dedicated his studies to the taxonomy and utility of vigilance (Parasuraman et al., 1987) and sustained attention (Parasuraman, 1984a) in detection and discrimination (Parasuraman, 1984b), in human factors and automation (Parasuraman & Reiley, 1997) as well as monitoring and search (Parasuraman, 1987) and the topics of the brain systems of vigilance (Parasuraman et al., 1989) of how the brain is put to work in the sense of “neuroergonomics” (Parasuraman & Rizzo (eds.), 2007; Parasuraman & Wilson, 2008). Vigilance may not be automatic and “effortless”, but may involve stress and “hard mental work” (see: Warm et al., 2008). 10 It should seem almost “natural” to start any kind of exposition about human attention with the traditional figure of the attentive “hunter” that besides the monitoring of other animals seems a classical conceptual persona of creature attention as different from machine attention and thus contains consequences for the self in its development. New et al. (2007) recently proposed an ancestral animate monitoring hypothesis that is confirmed by imaging studies (see Böttger et. al., 2010) by comparison with stimuli of moving animals in comparison to optokinetic stimuli. Attention is captured more easily by ancestral animals than for instance by equivalent contemporary moving computer stimuli: “For ancestral hunter-gatherers immersed in a rich biotic environment, non-human and human animals would have been the two most consequential time-sensitive categories to monitor on an on-going basis. As family, friends, potential mates, and adversaries, humans afforded social opportunities and dangers. Information about non-human animals was also of critical importance to our foraging ancestors. Non-human animals were predators on humans; food when they strayed close enough to be worth pursuing; dangers when surprised or threatened by

282 

Alexander Gerner

but might give us hints to understand how attention and the self could have co-evolved and differentiated over time. 2) Moreover, we can state that conceptual personae of the self give us the possibility of accessing something subjectively phenomenal even if the phenomenon in question shows itself to be difficult to be measured empirically and thus has to be combined and superimposed virtue of their venom, horns, claws, mass, strength, or propensity to charge; or sources of information about other animals or plants that were hidden or occluded; etc. Not only were animals (human and non-human) vital features of the visual environment, but they changed their status far more frequently than plants, artefacts, or features of the terrain. Animals can change their minds, behaviour, trajectory, or location in a fraction of a second, making their frequent reinspection as indispensable as their initial detection. For these reasons, we hypothesized that the human attention system evolved to reliably develop certain category-specific selection criteria, including a set designed to differentially monitor animals and humans. These should cause stronger spontaneous recruitment of attention to humans and to non- human animals than to objects drawn from less time-sensitive or vital categories (e.g., plants, mountains, artefacts). We call this the animate monitoring hypothesis.” New (2007, 16598). In animals the application of attention is crucial for survival. “Although attention is an efficient filtering mechanism, limited attention may be a major cause of mortality in nature.” (Dukas & Kamil, 2000, 502). As hunting or predation as well as predator avoidance (for example through the fear of snakes that catches attention in mammals) are principles of the animal kingdom, attention can be seen as a strategic biological mechanism that has evolved also in relation to survival in between predation and prey. On predation see Langerhans, R. (2006) ”Evolutionary consequences of predation: avoidance, escape, reproduction, and diversification.” In A.M.T Elewa (ed.). Predation in organisms: a distinct phenomenon. Heidelberg: Springer, 177–220. See Isabell (2006, 2009) on the category of the sensitive evolution of the mammal brain (including vigilance or strategic attention) towards those animals that have provided a recurrent survival threat from an evolutionary perspective (such as snakes) and those which have not: “Snakes have a long, shared evolutionary existence with crown-group placental mammals and were likely to have been their first predators. Mammals are conservative in the structures of the brain that are involved in vigilance, fear, and learning and memory associated with fearful stimuli, e.g., predators. Some of these areas have expanded in primates and are more strongly connected to visual systems. However, primates vary in the extent of brain expansion. This variation is coincident with variation in evolutionary co-existence with the more recently evolved venomous snakes.”(Isbell, 2006, p. 1). On the fear of snakes as attention “commander” in an evolutionary perspective see Soares (2010). In “Meditations on Hunting” Ortega y Gasset meditated, however, not from a biological evolutionary standpoint, but from a philosophical point of view on “hunting” as a fundamental human attentional diversion phenomenon, that we should have in mind as well besides its biological function.

Conceptual Personae of the “attentional self ”

283

on other concepts or only appears in the moment of its failure, as is documented by the large literature on pathological situations of the self as in schizophrenia or the missing body image (see: Gallagher, 1995), and is parallel to the case of “attention” that can be called a philosophical embarrassment (Waldenfels, 2004) that accompanies the literature of attention and its many non-unified theories and incoherent theoretical metaphors (see: Gerner 2012; Fernandez- Duques & Johnson 1999, 2002) through which it is described. The idea of proposing a necessary relation between attention and the self by means of an “attentional self ” shares difficulties in the description of “attention” and “self ” as independent concepts. And one might say that through the introduction of attention into the concept of the self increases the complexity in the understanding of what the self could be. Nevertheless, it is our aim to show that looking tangentially at the self from the perspective from attention we obtain concrete insights that can be made fruitful for the concept of the “Self ”. 1.1 Conceptual personae of the impossible attentional self: Monsieur Teste as the impossible “man” of attention Let us assume here for a few short paragraphs that the actual philosophical debate of the self would be fundamentally enriched if we took the philosophical personae of writers into consideration – as we will do in the first place with the conceptual personae Monsieur Teste of Paul Valéry as an attentional self in its conceptual constellations. The “impossible man of attention” Teste is considered by Valéry as being “no one” (Valéry, 1989, p. 131), and as such has “no personality that is unique and his own, no inseparable attributes.”(Ibid.) He is also characterized as well as the “Tête (Teste, Head)” (Valéry, 1989, 90) or “The game played with oneself ” (Valéry, 1989, p. 106) and its living self observation and will still always be “some stranger to me” (Valéry, 1989, p. 102), thus is fundamentally related to the “other” in its proper doubled definition as other-self. The Self is thus seen as a strange habit of interchange of two different perspectives in a kind of “self-game” between “me” (Teste himself) and “he” (Teste narrated or the narrative self and put into perspective by the narrator/author of Teste) – a higher-order

284 

Alexander Gerner

“narrative self ”: “Monsieur Teste had taken the strange habit of thinking of himself as a chessman in his own game. He could see himself. He could push himself across the table. At times he lost interest in the game. /The systematic use of Me and He.” (Valéry, 1989, p. 100). What, however, happens if the interest in the distinction of he/me is dropped and the self-game stops, seems a fundamental question that we should answer with the concept of a “self ”. “He” is also characterized as a plural episodic multitude of Selves “Teste. / The home of Selves/ The islands of Selves” (Valéry, 1989, p. 148) or man’s potential of a “pure” or virtual attentional self “beside his body” (ibid.) and as such will be helpful to understand better what – as we will see later- Metzinger in “Being no one” understands as an “attentional self ” as a precondition of stabilized subjective experience. But let us stick to Monsieur Teste, who Valéry paradoxically describes as between The “Everything” in the sense of all particular sensations (bodily needs, pain or caution) before any reflection and before a “pure reflexive self ” in which “Self ” and the sensational “Everything” would be merely polar properties of an attitude or disposition that can be turned into an automatic routine (body schema or body image). In the end for Valéry the maximum consciousness of the self is turned into a routine behavior and as such in Valery’s perspective is seen as animalized by the biological fact of being sensitive to typical biomechanical processes of animal life. Firstly, we will look at the case of the Paul Valery’s conceptual personae in “Monsieur Teste” in which not only Monsieur Teste shows itself as an a) impossible pure attentional self, but as well as part of an extended concept of the self by b) “Monsieur Teste’s” wife Émilie Teste and who is accompanied by a c) partial observer of Monsieur Teste, a friend of Monsieur Teste who observes him and describes him in an attentional scene at the “theatre” of consciousness as well as by an d) author-narrator-self of Paul Valéry that gains self-knowledge by reflecting on his own conceptual personae Monsieur Teste that he properly created in odd moments of his life11. 11 “Teste was conceived – in a room where August Compte spent his early years – at a period when I was drunk on my own will and subject to strange excess of consciousness of my self.” (Valéry, 1989, p. 3).

Conceptual Personae of the “attentional self ”

285

Teste is defined from the outset as a concept of the impossible12, a “man of attention” that shows different notions about the conceptual personae of the attentional self related by its author: [A] Monsieur Teste – alias the “attentional self ” – is regulated by an equilibrium of possibility and impossibility [B]   Monsieur Teste – alias the “attentional self ” – is a mechanism that sets up and regulates the relation between the known and the unknown [C]   Monsieur Teste – alias the “attentional self ” – deals with the experience of the finite. All three notions of Monsieur Teste as a thinking image of the attentional self are concepts of a multitude of attentional selves that are interrelated and connected and none of them alone can provide a realistic account of a unified “attentional self ” or the one and only persona of Monsieur Teste, which also means each conceptual personae being taken on its own and as a materialization of the “Self ” is impossible, as they would be neglecting necessary layers of the self (from the minimal to the narrative self). Even so, considering the minimal self we have to listen to what Bernhard Waldenfels states about the complex working of attention. According to Waldenfels (2004) there is not one but a double-event of attention, and thus – I would propose – also a double concept of – for instance – the minimal “self ”: as attention is not just “Auffallen” the “what happens” side that could be conceptually mapped by physiological processes alone, but is as well an “Aufmerken” – that is an event that shows “what happens to me”, and thus implies the perspective of an attentional self to which through my body/ a “me” is 12 “Why is Monsieur Teste impossible? That question is the soul of him. It changes you into Monsieur Teste. For he is no one other than the demon of possibility. His concern for the whole range of what he can do rules him. He observes himself, he manoeuvres, he will not allow himself to be manoeuvred. He knows only two values, two categories- those of consciousness reduced to its own act: the possible and the impossible. […]. Its brief and intense life is spent in watching over the mechanism, which sets up and regulates the relation between the known and the unknown. It even applies its obscure and transcendent powers obstinately to simulating the properties of an isolated system in which the infinite plays no part.”(Valéry, 1989, pp. 6–7).

286 

Alexander Gerner

affected and may respond before any conceptual epistemic noting of x, and any kind of narrative or narrative self is implied. According to Waldenfels, attention should first be approached through a double-view or a double-grasp of the two single movements (Auffallen- Aufmerken), in addition to the third part, the taking notice side of epistemic attention, which is when I attend to x and note or recognize that it is x (for example, a book on my desk). Thus, the “self ” for Valery’s thinking personae Monsieur Teste deals correspondingly to our view on attention with the incoherence of the mind that for example a new stimulus (It happens) creates that happens first as an event that is followed by the event of the self (something happens to me) – that what happens, is actually happening to me, before I note what it is that happens and what this me as such involves. The “self ” thus first appears as a mental pre-conceptual response to the instability of what happens and what can thus possibly happen to me. In this line of thought Paul Valéry has already stated: – I am the unstable. – The mind is maximum possibility- and maximum capacity for incoherence – The SELF is the immediate response to each partial incoherence which is a stimulus (Valéry, 1989, p. 73).

Thereby we can note three of Monsieur Teste’s notions of the attentional self: [1] The self is part of responsive attention, an direct or “immediate response” to a situation involving the incoherence event [2] The self is stimulus driven, and as such we can easily call it an empirical attentional bottom-up self [3] The self is both an actuality attentional mode, or a capacity/bodily mechanism and a possibility attentional mode, or a mere possibility of the mind. The self is seen here as a consequential response to the maximum of incoherence of the mind. The incoherence of the mind makes it alive, as does the coming to the point in its most exactness. This also means that Monsieur Teste would be not “mortal” but dead, if he would be fixed as a “pure absolute” or “pure transcendental” self. This impossible

Conceptual Personae of the “attentional self ”

287

purely transcendental notion of the concept of the “self ” leads us to several key questions of Monsieur Teste as the dynamic centre of awareness: […] how does one choose a character to be oneself? How does that centre take shape? Why, in the theatre of the mind are you You? You and not me? But is that not precisely Monsieur Teste’s research: to withdraw from the self, the ordinary self, by trying constantly to diminish, combat, compensate the irregularity, the anisotropy of consciousness. (Valéry, 1989, p. 109)

Hence Valéry that turns the self into a paradoxical concept makes an essential double notion of the self clear: a) The ordinary functional self-mechanism creating coherence in a sea of chaotic episodic impressions and stimuli is located in a permanent struggle with an implicit b) “extraordinary” self that in extremis properly not only compensates irregularity but even creates the incoherence of the mind by taking each episodic extraordinary attentional event of the self at face value. Let’s stick just for a moment to the second extraordinary notion of the self that breaks with diachronic self-habits by considering the relation of the conceptual personae of Monsieur Teste together with the conceptual personae of his “wife” Émily Teste: Émilie Teste achieves the possibility of self-formulation while reflecting on her husband, self- designating herself as being more than a mere witness, but already an “organ” (Valéry, 1989, p. 24)- though externalof Monsieur Teste. By reducing herself to an organic and organizational inter-dependence with Teste, however, being a “non- essential” (ibid.) or external organ in his life-world, she describes the essence of her character as longing “to be surprised” (ibid.) by Monsieur Teste’s episodic self and the relation she has to Teste as being strongly bound by “the uncertainty of his moods”(ibid.). Their conceptual relation is double a) of “husband” and “wife” and their distinct activities and necessities are said to be in tune b) in the sense of interdependence of lived organization. Émilie Teste and her “husband” are two conceptual personae of one attentional self and as such are interrelated13. The practical organ of sensible stimulus-driven 13 “But I am somewhat more than the witness of his life. I am a part, almost an organ of it, though non-essential. Husband and wife as we are, our actions are harmonized in marriage and our temporal necessities are well enough adjusted,

288 

Alexander Gerner

responsivity to the world and the ideal episodic or “epochal” attentional self of Monsieur Teste are two inseparable aspects of the attentional self. The way of investigating in the impossible autochthonous pure attentional self of Monsieur Teste finds its necessary responsive structure in the pragmatic function of a stabilizing attentional self – that here is called Madame Émilie Teste- and that in Thomas Metzinger could be described as the function of focusing the centre of attention in order to stabilize subjectivity. This, however coming from the double constellation of Paul Valéry’s Monsieur Teste/Émilie Teste is however just a simplistic picture of the attentional self that in its foundation is dyadic and contains a double movement of focus and distraction. Thus coming back to the key question of how the centre of the self gains form, and how Monsieur Teste becomes himself, we have to look at the description of Émilie Teste. In the metaphoric “conceptual marriage” between Edmond and Émilie Teste, her preoccupation about his episodic self is defined by the fear of his disappearance through absorption, or a purely “episodic” self that would be equaled with change itself: …A bit more of such absorption, and I am sure that he would be invisible!… (Valéry, 1989, p. 26).

Émilie shows herself as “almost” a bodily organ regarding the possibility of an episodic attentional self of Monsieur Teste in her function of self-preservation, for instance in regulating the “social life”14 of the attentional self of Monsieur Teste. This is exemplified by her actions of responding to the environment as in the case of her responding to a despite the immense and indefinable difference of our minds. So I am obliged to tell you incidentally about her, who is now telling you about him” (Valéry, 1989, p. 24). 14 In relation of a merely “spiritual” part of the attentional self alias Monsieur Teste and its social organ Émilie Teste one should consider the following quote by Valéry in which the vital function of self-preservation is attributed to the social “organ” of the ego while the pure spiritual Teste kills in its epochal way: “Ego./ There are things I could “clarify” and don’t wish to…/ I keep them in condition. The social keeps the spiritual in condition. /Social life does not want Monsieur Teste to discharge his duties. We must admit that the spirit kills and the letter gives life… or at least, preserves it.” (Valéry, 1989, p. 125).

Conceptual Personae of the “attentional self ”

289

letter to a friend of Monsieur Teste, that is, questioning and observing the “life” of Monsieur Teste from a distance. By being a conceptual personae or a figure of thought of sensibility that understands little but responds to the necessity of the attentional self to work, she would sometimes act instead of or as part of Monsieur Teste, the “man of attention”. The relation of Monsieur Teste and Émilie Teste shows itself through in-between events15 (Waldenfels 2002) of attention because “in the phenomenon of attention the mundane consciousness fights with transcendental consciousness.” (Blumenberg, 2002, p. 200), and in this perspective one has to see what could be formulated as a complex attentional self between constitutional stability and instability at the same time: The self is a marriage! Valéry gives us the most rich and thought-provoking conceptual personae of an attentional self: Monsieur Teste and his wife Émilie with whom we are able to explore the working of related but distinct complementary functions of the self that are in constant (dis-) equilibrium. Equipped with the pragmatic function of her “reading” Monsieur Teste, Émilie Teste describes herself as acting complimentary to the extraordinary, transcendental attention based on interior observation of Monsieur Teste: Kind sir (and friend) […] So I read your letter to Monsieur Teste. He listened without showing what he thought of it, nor even that he was thinking about it. You know that he reads almost nothing with his eyes, but uses them in a strange and somehow inner way. I am mistaken – I mean a particular way. But this is not it at all. I don’t know how to put it; let’s say inner, particular…, and universal!!! His eyes are beautiful; I admire them for being somewhat larger than all that is visible. One never knows if anything at all escapes them, or, on the other hand, whether the world itself is not simply a detail in all that they see, a floating speck that can 15 “Zwischenereignisse” or “In-between-events” for Waldenfels are events in which something appears [auftritt], while they connect to something else, without having being connected or related to it before. (Waldenfels, 2002, p. 174; see also Waldenfels, 1987, pp. 46–48). According to these in-between-events, Waldenfels inscribes in-between-events of attention in his philosophy of the “strange” or the “other”. In-between-events of attention happen either a) between me and the others or b) between a group (us) and another group or c) in general between the “proper” [dem Eigenem] and the strange [dem Fremdem] that unfolds in the phenomenon of attention consisting of a double event in auffallen-aufmerken and is irreducible to one of its double events.

290 

Alexander Gerner

besiege you but does not exist. Sir, in all the time I was married to your friend I was never sure of what he sees. (Valéry, 1989, p. 22, translation altered, A.G.)

Monsieur Testes attention is not only particular, but also universal, and above all an inner principle. The eyes of Teste are not just organs to perceptually pick up visual information. They are more actors of transformations in which a world densifies or vanishes, in which by attentive observation a detail of the world can become a whole world or in which the world expands or shrinks in an instant: zim-zum. What stays are the riddles of not understanding exactly what it is that the eyes of Monsieur Teste actually see: What is this inner attentive self? As Valéry says, the “proper soul” of Monsieur Teste is his impossibility of existing as a “real person”. On the other hand M. Teste is thinkable as a “proper demon of possibility”, a “personnage de fantaisie” (Valéry, 1946, p. 7) that extends in an episodic self that is measurable in exceptionally few “quarters of an hour”16. Paul Valery called “Monsieur Teste” in his introductory preface the “impossible man of attention”. Despite the rhetoric of failure and the impossibility of the model character Monsieur Teste postulated by Valéry as the constitutive impossibility of M. Teste as a real person, paradoxically he is taken as someone that could transform the one thinking Monsieur Teste into the proper Monsieur Teste – “[Elles change vous en Monsieur Teste] It changes you into Monsieur Teste” (Valéry, 1989, p. 6 [1945, p. 11]). This change into Monsieur Teste from the one thinking Monsieur Teste is central. At the same time it is even more central to note that there is another change that is excluded in the moment in which the reader or thinker of Monsieur Teste changes herself into Monsieur Teste. By becoming a “man of attention”: the change into change itself is excluded. This means per definition of an attentional self there can never be an absolute identity between the episodic change and the one experiencing the change. Thus this difference constitutes the self-variance or 16 “Coming back to Monsieur Teste, and observing that a character of his kind could not survive in reality more than a few quarters of an hour, I say that the problem of that existence and its duration is enough to give it a sort of life. The problem itself is a seed. A seed is a living thing; but there are some that could never develop. These make an effort to live, become monsters, and monsters die. In fact we know them only by this remarkable property of being unable to endure. (Valéry, 1989, p. 5).

Conceptual Personae of the “attentional self ”

291

instability of the attentive self. This also shows the openness of the self as unequal of the changing episode: The attentional self acts as the difference of “what happens” and “what happens to me”. Monsieur Teste, as described by his author Paul Valéry, shares with his author one basic intention regarding an episodic self: “to extend the duration of certain thoughts” (Valéry, 1989, p. 3)17. Parallel to this temporal limitation we find the neurobiological research of the Munich Neuroscientist Ernst Pöppel and the temporal empirical window of three seconds18 of attention. This limited capacity in time or load of an episodic attentional self that Valéry in his “Extracts from Monsieur Teste’s logbook” calls the centre of elasticity, “Centre de ressort” (Valéry, 1946, p. 65) can ideally prolong the attentive observation until a temporal maximum of a quarter of an hour. Recently research on the praxis of meditation has shown that this temporal plasticity differs and increases when carrying out a continuous praxis of attentive meditation19. The “centrality” into which these different episodic attentional periods come together and gain coherence while not dissolving into change and the episodic itself is exactly one of the most difficult philosophical problems of the self, from perceptual binding to the coherent identity of selves over time: What he calls: The Central Problem. To be a center. This problem is one of agreements – coincidences, stabilities, time intervals of various kinds in combination 17 “Monsieur Teste? / If reflection on one point could be prolonged, it is likely that the content of thought would allow its substance to appear, and that whatever it might represent, one would see the canvas, the panel, the texture- the true element of time. There would be an expansion of time during which a sort of pure sensation would arise in its place, as the substance of the meaningful illusion. But such an extension is impossible. Attention ends in the perception of attention itself, which is fatigue […]” (Valéry, 1989, p. 121). 18 Pöppel, E. (1985). cit. In Assmann, A. (2001, p. 17). 19 See for example: Lazar, S. et al. (2005). Meditation experience is associated with increased cortical thickness. Neuroreport 28; 16(17), 1893–7; Lutz, A. et al. (2008). Attention Regulation and monitoring in meditation. Trends Cogn Sci 2008, 12(4), 163–9; Manna, A. et al. (2010). Neural correlates of focussed attention and cognitive monitoring in meditation. Brain Res Bull. 82 (1,2), 46–58; MacLean, K. et al. (2010). Intensive Meditation Training Improves Perceptual Discrimination and Sustained Attention. Psychological Science, 21(6), 829–838.

292 

Alexander Gerner

or contrasts […] The Contrasts between specific times, various functions- that is to say, what keeps us from merging with change itself, and makes us aware of divergence, […] and gives us the Capacity of the moment […] its instability (self-variance). (Valéry, 1989, p. 120)

For Bernhard Waldenfels, Paul Valéry marks a hybrid between writing and thinking in a singular way that is preoccupied with attention in the Cahiers- but also in Monsieur Teste- that opens up a “new form of poetics” (Waldenfels, 2004, p. 28) as well as formulating a concept of the self by following the self’s traces of strangeness [Fremdheit], absences [Absenzen], deviations [Abweichungen] and “delays” [Retardandis], and as such considers this worrisome phenomena of an attentional self, that could here be called unbalanced attentional dives. As the diver never can maintain himself completely in his or her “element”, and as such needs to come back to the surface to take a breath, the man of attentional self – Monsieur Teste – feels like he is a fish out of water, that needs to prolong his being in a longer duration and thus stretch the episodic attentional self and the episodically appearing objects of attention. For Valéry the condition of something as an object of attention, of a thing, is only able to be understood when one is shown a rule of a medium value in the moment of its observation: “The things are only what they are as average duration and depth of observation.” (Valéry, Cahiers II, p. 262, cit. in Waldenfels, 2004, p. 34). Things are gazes directed towards them for a certain amount of time in comparison and relation to other possible things. The functioning of attention, according to Valéry, is explained in relation to a regular value that only shows up at the moment of a dislocation and thus the imbalance in relation to this medium value is constituted through a process of finding a re-equilibrium, or we could say: attention is what deviates from what should be expected as a medium value and thus the attentional self is the gesture between the balance and imbalance of this medium value of experience. This medium value seems already implicit in expectation regarding things and their givenness and by a possibly notable shift away from a pre-given expectation towards the actually attended thing in observation. To be attentive means – in this sense of Valéry – “to find or go looking to find a value of X that differs from its medium value X_” (Valéry, ibid.) and that both values show themselves only in their dyadic relation.

Conceptual Personae of the “attentional self ”

293

That is how the attentional self shows itself as an imbalance of the mind that calls for (re-) equilibrium. The moment at which attention happens – Valéry notes – can be minimal: “a tooth aches – a disquietness [Sorge] – a necessity” etc. are all seeds, luminous signs- of attention” (Waldenfels, 2004, p. 34). This is why the conceptual personae of a pure “unstable” episodic self that is devoid of any bodily multisensory experience – such as Monsieur Teste – is still a conceptual personae of an attentional self.

2.  Heautoscopic “attentional self ” 2.1 Italo Calvino’s Mister Palomar’s development from an aesthetic to an heautoscopic “attentional self” Calvino’s text “Mr. Palomar” (1985 [1983]) shows three steps of a phenomenology of attention – firstly, phenomenological description, secondly, narration, and thirdly, meditation. Moreover, Mr. Palomar deals with the self in three types of experience20 as Calvino explicitly remarks at the end of his book: a) the self of Mister Palomar in visual experience b) the self in experience involving language/meaning and symbols and c) the self in speculative experience (cosmos, time, infinity, the relation of self and world and the dimensions of mind). The name of the conceptual personae of the self, Mr. Palomar, is derived from a famous astronomical observatory21 (Calvino, 1985, p. 37), which is located in the north of San Diego County operated 20 Those marked “1” generally correspond to a visual experience, whose object is almost always some natural form; the text tends to the descriptive/ Those marked “2” contain elements that are anthropological, or cultural in the broad sense; and the experience involves, besides visual data, also language, meaning, symbols. The texts tend to the form of the story./Those marked “3” involve more speculative experience, concerning the cosmos, time, infinity, the relationship between the self and the world, the dimensions of the mind. From description and narrative we move into meditation” (Calvino, 1985, p. 128). 21 See .

294 

Alexander Gerner

by the California Institute of Technology, Palomar. Palomar in Calvino is the second conceptual personae of the attentional self we want to explore and is seen here as an exemplary conceptual personae of the attentive gaze of the self in observation of the self and the world. At the end of Calvino’s book, Mr. Palomar comes to the phenomenological conclusion of his own being, and that his task is “to look at things from the outside” (Calvino, 1985, p. 113) while all the observables (from the wave to the stars, animals or daily objects) “ask” him to prolong his attention towards them. His attentive gaze is caught by a multitude of details ready for his observation of his attentive gaze by “running all over the details” and is “then unable to detach itself ”22 (Calvino, 1985, p. 113) from these observable details. That is why in the end Palomar sees just one path that he could go and that is to extend his attention: Mr. Palomar has decided that from now on he will redouble his attention: first, by not allowing these summons to escape him as they arrive from things; second, by attributing to the observer’s operation the importance it deserves. (Calvino, 1985, p. 113)

But the following questions remain beyond the magnification of attention and thus the extension of the self: can Palomar actually observe everything that he wants or ought to observe in a universal and objective manner, and how long can he keep the empirically perceptible, observable details or run over all of them in an attentive gaze? Is it actually possible to run over all the details if they are for example all the stars in the universe, or all the crumbs of bread on a table, or all the atoms and neuronal relations of single neurons in a single brain? How is the complexity of the world and reality manageable with Palomar’s attitude? 22 “After a series of intellectual misadventures not worth recalling, Mr. Palomar has decided that his chief activity will be looking at things from the outside. A bit near-sighted, absent-minded, introverted, he does not belong temperamentally to that human type generally called an observer. And yet it has always happened that certain things- a stone wall, a seashell, a leaf, a teapot- present themselves to him as if asking him for minute and prolonged attention: he starts observing them almost unawares, and his gaze begins to run over all the details and then is unable to detach itself ” (Calvino, 1985, p. 113).

Conceptual Personae of the “attentional self ”

295

The perfect universal observer shows himself in Mr. Palomar as being impossible and Palomar becomes more and more the expression of this failure of universal observation, as his self is located in between the world and its pure observation “as it is”: At this point he faces a critical moment; sure that from now on the world will reveal to him the wealth of things, Mr. Palomar tries staring at everything that comes within eyeshot; he feels no pleasure, and he stops. A second phase follows, in which he is convinced that only some things are to be looked at, others not, and he must go and seek the right ones. To do this, he has to face each time problems of selection, exclusion, hierarchies of preference; […] (Calvino, 1985, 113–114).

After the failure of looking at all the things with the same importance or giving all things the same relevance (as he already tried while observing the naked bosoms of a woman at the beach along with everything else) his pleasure of observation dissolves and he finally stops. His difficulties are due to the involvement of the self in observation. However, in the second attempt Palomar gets involved in one of the basic problems extremely relevant to the theme of attention and thus the attentional self: the problem of selection, exclusion, creation of hierarchies of attention and preference – why Palomar is attentive to this and not that, as not everything can be attended to at the same time and there is a basic non-simultaneity in attention. All that we are attentive to has already happened in a double way: it has happened and it has happened to me and while I start noting my experience of what has happened it is already “too late” for the attentional double event. This leads for Palomar into the next question of how we can exclude the self, here called the “ego”, of the observer from the observed and impossibly, but desirably, let the “world look[s] at the world” (Calvino, 1985, p. 113) itself. Mr. Palomar therefore questions the “who” in observation, or as he says: “Who’s eyes are doing the looking?”: […] he soon realizes that he is spoiling everything, as always when he involves his ego and all the problems he has with his own ego./ But how can you look at something and set your own ego aside? Whose eyes are doing the looking? As a rule, you think of the ego as one who is peering out of your eyes as if leaning on a window sill, looking at the world stretching out before him in all its immensity (Calvino, 1985, p. 114).

296 

Alexander Gerner

Calvino’s Mr. Palomar does not just want to suspend the participation of the human “ego” in observation, but also wants to shift radically the perspective of attention constituting an observer or a self as if “leaning on a window sill” observing the world, in which the world observes the world without the blur or noise that the egoic bodily centered observer always seems to introduce. Palomar’s intent seems the opposite of a classical Husserlian phenomenological approach that by its program wants to bracket all statements on the existence or non-existence of the world, in order to reach phenomenological consciousness of things by epoché- the suspension of the world (bracketing) by eidetic variation. Palomar uses the image of the “window” through which the “ego” observes the world and that changes position as a thinking image of the attentional self (that has another image for instance in the “frame” ([Der Rahmen23] Pöppel, 2010[2006]) or the tunnel in the “Ego-Tunnel” (Metzinger, 2009) metaphor. However, Mr. Palomar is actually seen as the conceptual personae of an “attentive observing Ego” that observes and besides being linked to a certain egoic perspective, tries to change its perspective from looking from “inside out” to observing from “outside out” -as happens for example in Out-of Body experiences, as we will see later. In terms of Italo Calvino we can state a shift of perspective. From the perspective of subject-object relation in “the ego looks at the world” he shifts the gaze from one object to another. This means we get a radical new perspective in “The world looks at the world” (Calvino ibid.), in which a non- egoic self represents the looking world and the objects are the looked-atworld: The gaze becomes autonomous and seemingly independent of the ego. Palomar doubts that this radical shift brings about an expected “general transfiguration” (Calvino, 1985, p. 114) of the observed objects of the world, as he “casts his eyes around” and states that no such transformation of worldly objects occur:

23 The subtitle of the book of the German neuroscientist Ernst Pöppel on the brain and the self shows a similar inverted shift in which the observational ego or self is now looked at from the perspective of the brain as the title shows: “The Frame. A view of the brain onto our ‘I’” [Der Rahmen. Ein Blick des Gehirns auf unser Ich].

Conceptual Personae of the “attentional self ”

297

No such thing. The usual quotidian grayness surrounds him. Everything has to be thought of from the beginning. Having the outside look outside is not enough: The trajectory must start from the looked-at-thing, linking it with the thing that looks (Calvino, 1985, p. 115).

This higher attention shift, not just from one observed object A to another object B, but an radical change of perspective in observation – in which the proper frame or window of the self is pushed outside the proper sensual limitations of the body – seems to fail and always bounces back on the ego of Palomar, his eyes and glasses that the world “needs” to be perceived: But how can you look at something and set your own ego aside? Whose eyes are doing the looking? As a rule, you think of the ego as one who is peering out of your own eyes as if leaning on a window sill […] So then: a window looks out on the world. The world is out there; and in here, what do we have? The world still- what else could there be? With a little effort of concentration, Mr. Palomar manages to shift the world from in front of him and set it on the sill, looking out. Now, beyond the window, what do we have? The world is also there, and for that occasion the world has been split into a looking world and a world looked at. And what about him, also known as “I”, namely Mr. Palomar? Is he not a piece of the world that is looking at another piece of the world? Or else, given that there is world that side of the window and world this side perhaps the “I”, the ego, is simply the window through which the world looks at the world. To look at itself the world needs the eyes (and the eyeglasses) of Mr. Palomar. (Calvino, 1985, p. 114)

The understanding of the “I” or ego here as the window through which the world looks at the world could be called the attentional self that can be shift its perspective from an aesthetic bodily-centered perceptive point of view towards an heautoscopic point of view among others. The basic metaphor of the self therefore seems the window through which the world is observed. Palomar’s undertaking seems derived from the transformation of a window scene by Caspar David Friedrich famous drawing “Window looking over the park” to the surreal visual experience of the window of René Magritte and his painting “The human condition”. Can we imagine the window through which we look as being put outside – beyond the frame of the window of subjectivity – and even turned around towards the window?

298 

Alexander Gerner

Palomar is confronted with the dilemma that he wants to look at things “from the outside” (Calvino, 1985, p. 113) and not from the inside. This means he wants to bracket and thus suspend his ego in the moment of observation. Palomar is trying to change the anchor of his attention from a bodilycentered aesthetic point of view (inside-out) towards a heautoscopic attention point of view (outside-out) in which even his body is disembodied and his gaze and regard can start not from himself but from the observed object. This absolute bracketing of the aesthetic attention seems, however, impossible. The attentional self can be seen as the play of perspectives and gazes in between aesthetic and heautoscopic poles. Similarly we will see that attention purely without a body in out-ofbody experience and heautoscopy (Metzinger, 2009) seems impossible. Breyer24 (2011) also notes that the theoretical transformation of pure aesthetic towards pure heautoscopic attention in which the own lived body [Leib] can be seen as an objectified body [Körper] among others in the thing world and scrutinized from the point of view of a neutral objective observer is a “liminal phenomenon” that shows the two poles of attention between the pure immanence of inwardness of the sensing living body and the pure outward directedness of the neutral objective observer as impossible, as in reality it can never be reached by phenomenological experience permanently suspending one of these poles. Through this meditation Palomar reaches the problem of identity and differentiation in attentional observation, as for him observation is linked to the observation of a self-signifying activity (a sign, a summons, a wink), and the signification of a self, an “itself ” that signifies something and thus catches and attaches attention to both poles- of the observer and to the observed- by separation or distancing one thing from the other: From the mute distance of things a sign must come, a summons, a wink: one thing detaches itself from the other things with the intention of signifying something… 24

“Die Kraft der Aufmerksamkeit reicht […] nicht zu ihrer eigenen Verabsolutierung in die eine oder andere Richtung hin. Zwar kann man sich den Polen, zwischen denen sich das attentionale Spektrum erstreckt, je nach den Umständen und den eigenen Vermögen, die durch Übung modifiziert werden können, relative stark annähern, doch bleibt stets die anziehende bzw. rückbindende Kraft des jeweils anderen Pols bestehen” (Breyer, 2011, p. 161).

Conceptual Personae of the “attentional self ”

299

what? Itself, a thing is happy to be looked at by other things only when it is convinced that it signifies itself and nothing else, amid other things that signify themselves and nothing else (Calvino, 1985, p. 115).

Mr. Palomar is, however, also a conceptual personae of failure in the attentive gaze in observation, as he- from the outset of an naïve phenomenological approach- fails to observe all that there is to see in a single observation with his bare naked eye: His plan is to maintain the duration of observation as long as possible to simply “see a wave” (Calvino, 1989, p. 4) until he would notice “that images are being repeated, he will know he has seen everything he wanted to see and he will be able to stop” (Calvino, 1989, p. 4) his phenomenological observation of the single wave. The succession of singularities in which he wants to notice a pattern of repetition seems a difficult task and time-consuming as well as tiring for his attention, but also his disposition as being “impatient to reach a complete, definitive conclusion of his visual operation, looking at waves […]” (Calvino 1989, p. 6). Palomar wants to simplify his observation through the method of “mastering the world’s complexity by reducing it to its simplest mechanism.” (Calvino, 1989, p. 6) in this case a single wave. However, the observation of everything there is to see with “scientific attention” – a single wave that then may become a universal of seeing waves – does not seem at all easy and brings us to the paradox of seeing the regularity of the singular that in the case of the wave is a very complex dynamic phenomenon, thus becoming another conceptual personae of impossible attention. In another episode regarding the conceptual personae of impossible attention Mr. Palomar rehearses four modes of attentional gaze (Bernhard Waldenfels, 2006, pp. 96–98) in being attentive to the naked breasts of a woman lying on the beach (the empty gaze, the fixed gaze, the overflying gaze and the universal all-embracing gaze) that show the important relation between attention and the ethos of the senses. Attention and responsibility, amenity, or courtesy of how to be attentive is closely related. This means: not paying heed and thus overhearing or overseeing someone can create a socially conflicting situation as well as being too attentive or (over-) attentive. This also means that attention is not just programmable but has to be offered and can also be withhold, can be voluntary given or involuntary

300 

Alexander Gerner

imposed, attention can be withdrawn or denied and as such is culturally and historically codified by values concerning how to be attentive to certain situations or not. Attention is valued and socially shared and thus is joined by social claims of validity. This also means that the attribution of attention can in a social situation become too much for the “normal” claim of validity of a certain situation. Therefore paying close attention – for example to the breasts of a naked woman on the beach- can be perceived as penetrating over-attention, intrusiveness or pushiness. At the same time a lack in giving attention for instance to the beauty of a new haircut or a new dress someone wears may be being seen as rudeness.

From aesthetic to heautoscopic attention In the end Mr. Palomar comes to the phenomenological conclusion of his own being, and that his task is “to look at things from the outside” (Calvino, 1985, p. 113) while all the observables (from the wave to the stars, animals or daily objects) “ask” him to prolong his attention towards them. His attentive gaze is caught by a multitude of details ready for the observation of his attentive gaze by “running all over the details” and is “then unable to detach itself ” (Calvino, 1985, p. 113) from these observable details. That is why in the end Palomar sees just one path he could go and that is to increase his attention, even if the following questions remain:

Conceptual Personae of the “attentional self ”

301

1. Can Palomar as conceptual personae of the attentional self actually observe everything that he wants or ought to observe in a universal and objective manner, and how long can he stay with the details or run over all of them in an attentive gaze? 2. Can Palomar anchor his attention not from an bodily centered perceptive (aisthetic25) point of view, but instead from a point of view “out of his body” and from “inside out of the body of the thing” that he wants to observe? A heautoscopic attentional self as a scientific observer is proposed in which even his body is disembodied and his gaze and regard can start not from himself but from the observed object. This means that the theoretical transformation of a pure aesthetic towards a pure heautoscopic attentional self in which the own lived body [Leib] can be seen as an objectified body [Körper] among others in the thing-world and scrutinized from the point of view of a neutral objective observer is a “liminal phenomenon” that shows the two poles of attention between the pure immanence of inwardness of the sensing living body and the pure outward directedness of the neutral objective observer in an impossible split. This means in reality it can never be reached by phenomenological experience. This absolute bracketing of the real aesthetic attentional self seems impossible; however, it could be imagined and thought of, as a metaphysical form of virtual attention, that would, however, never be independent from a self in a body. Now52I will concentrate on the implicit “attentional ghost” of Thomas Metzinger’s virtual and attentional self that in out-of-(real)-body experiences can shift the self-localization and viewpoint of attention away from an intra-or extramissionist attention, and thus becomes autoscopic from a viewpoint beyond embodiment. All conceptual personae are fundamental diagrammatic features of the concept of attention which, if taken on its own as the only necessary feature or function, show themselves to be impossible, and as such would provoke a strange and impossible phenomenon in which an episodic attentional self would gain no dynamic of experience in between stability and imbalance. 25 Aisthesis (perception) refers to a form of aesthetics that is based on perception. But this does not mean that one could reduce aesthetics to attentive perceptual routines. (See: Gerner, 2010b).

302 

Alexander Gerner

2.2 Metzinger’s “attentional ghost”: the virtual self floating out of the body and the impossible full body illusion Let’s look at a non-orthodox experience of attention that lies in between sleep and wakefulness. Here attention seems to do the impossible and follow a virtual self, leaving the material body or the attentional focus, going up in the air and floating ghostlike over one’s body. This figure of impossible attention is referred to as Out of Body experiences (OBE) and has been recently investigated as an empirical phenomenon in the Cognitive Neurosciences as we will see. Metzinger makes the point that non-standard cases of self experience are an epistemological important entrance door for verifying and researching into his theory of the minimal phenomenal self and its model (even though non-standard, deviant or pathological, extraordinary cases of selfhood might not automatically lead to a correct description of having a self or “selfhood”. For Metzinger what is important in these non-standard, altered and/or pathological experiences is what fails and shows up and this gives us access to in comparing 1st PP reports (of patients or research subjects) with neuro-scientific 3rd PP studies (trying as well to understand the “neural correlate” to the phenomenological description of what appears in experience, which leaves the difficult questions of causality still open for debate and the philosophical foundation of correlates in general) into how a permanent body-“model” is transformed. The phenomena he deals with in the third chapter of The Ego Tunnel (2009) called “Out of the Body and into the Mind: Body-image, Out-of-Body Experiences, and the Virtual Self ” are non-orthodox phenomena such as the Out-of-body experience which for a long time where treated as and in para-/pseudoscience, but which gained momentum in psychology in the 1980s (see Blackmore’s research) although they have long been “career obstacles” for researchers choosing these topics. However, since the end of the late 1980s and 1990s they have been picked up by neuroscientists, (neuro-) psychologists and psychiatrists such as Devinsky (1989) or have been treated in relation to seizures and autoscopic phenomena as well as concerning felt presences, multiple perspectives and duplications of one’s body (phantom body); Botvinick & Cohen (1998) regarding the rubber hand illusion, and which recently entered a new phase of research in which Metzinger became involved with as did the Swiss neuroscientist

Conceptual Personae of the “attentional self ”

303

Olaf Blanke (2002; 2004; 2008) and the Swedish researcher Henrik Ehrsson (2007; 2012). The latter actively induced these phenomena in test persons through brain-area stimulations, that of course also raises lots of ethical questions that Metzinger is also aware of (in the last chapter of his book (2009) he asks what a good state of consciousness is (and what are non-desirable or eliminable mental states or experiences; also Metzinger tries to induce certain abnormal body experiences in test persons by stimulating certain brain areas, swapping bodies or to evoke a full body illusion as in Blanke and Metzinger’s test case (2008). Other phenomena in which our attention is working in between sleep, dream and wakefulness are for example trance-like states in between dreaminess and wakefulness and out-of-body-experiences which are an example, that through attention guide our body image into an unusual or non-orthodox experience of our body-selves. It seems in this case that our attention is directed in a way that a regular perceptual state would not enable us to. We can say that our perceptual map of ourselves is dislocated, our attention is placed on a mapping of a virtual body self in an out-of-body experience, which means that OBE form part of being attentive to ourselves in a non-orthodox way, and are less grounded by our surroundings than in perceptual attending. The in-between event between wakefulness and dream attending of ourselves from an out-of-body point of view had happened to myself, when I personally experienced something that I had never felt before and that was not a mere dream. At the same moment that I had broken my leg and had been prepared to be operated on after being anaesthetized in Germany over 20 years ago, I had a proper out-of-body experience in which my body left my fixed body, which was not able to move on the operation table and the virtual body floated as an “out-of-body” ethereal body trying to start to play football but my body was floating across the operation room looking down on me and I couldn’t control it myself, and then the OBE stopped as the anesthesia had put me in a sleep state. 2.2.1 The concept of disembodied attentional self in an out-of-body experience (OBE) It seems evident: attention and an attentional self both are embodied. But having said this, is it actually always necessarily put into perspective

304 

Alexander Gerner

from one unique and unified fixed body position? Can attention actually wander its centeredness or attentional self-location, and can it be dislocated, willingly or not, can the perspective with what we feel, perceive, reason or image be experienced subjectively out of a, my or the body? The heautoscopic and autoscopic, as well out of body experience will now be considered and what this can tell us concerning the “impossible” attention out of a living body and how this relates to the attentional self in general: Let’s consider this phenomenal relation of attention in the instance of a body by taking a closer look at Thomas Metzinger’s research into the seemingly impossible but actually experienced out-of-body experiences (OBE) (see Metzinger, 2009, chapter 3, Metzinger, 2005). Even though I will not come to the same representionalist conclusions about a permanent “virtual body model” as Metzinger does, his observations and research is valuable in respect of this phenomenon in between wakefulness and dream in which attention plays a major role in non-orthodox self-experience such as Out-of Body experiences. Metzinger refers to his own OBE in his recent “The Ego Tunnel” book (2009) and refers to several other researchers, specifically the OBE and consciousness researcher Susan Blackmore (1982; 1984; 1987, see also: 1993) whom he met for the first time in 1995. Moreover, he draws on the work of the Swiss biochemist Ernst Waelti and also research into the pathological conditions of OBE, for example OBE induced by seizures. In the domain of empirical psychology Blackmore carried out several review studies and came up with a scientifically sound empirical theory for these strange body phenomena. What is interesting in her book on near death experiences (1993) is how she collects data from phenomenological 1st person reports and structures their “grammar”. One of the characteristics she found out was the jump-like movement of the “ethereal” body movement. Another point Metzinger picks up from Blackmore’s26 research is the 26 Blackmore (1993) shows how subjectively experienced OBE as “religious limit phenomena” of a “soul” flouting out of the body or having “tunnel” visions of light at the end of the tunnel merely are phenomenological evidence for an injured and altered self (under the influence of drugs or anaesthetics) or even a “dying” brain state of consciousness, such as triggered by the fact that the brain in certain moments (dying) undergoes lack of oxygen (and as we might all know

Conceptual Personae of the “attentional self ”

305

following idea: Out-of-Body experiences are “models of reality created by the brains, that are cut off from sensory inputs during stressful situations and have to fall back on internal sources of information” (Metzinger, 2009, p. 87). Here a conflict of two processing sources (perceptual or observational feedbacks) and internal models could create different non-unified body representations and thus two types of attentional selves. This could explain why during the OBE I was angry that my body would not move towards the ball but floated over the ball in the operation room. Metzinger also underlines in Blackmore the idea that memory reconstructing visual cognitive maps of past experiences – say for example movement maps of walking on the beach – are most often organized from a geometrical 1st PP that is not our actual real but motionless body perspective – that is, a bird’s eye perspective27 rather than in real life (online) human- height perceptive situation. But what exactly are Out-of-Body experiences (OBE) and what have they in common with other autoscopic phenomena? And especially why are they important for understanding the self? We know that the representation of the self as bodily self is complex, and if we take the idea of a basic bodily self serious we are confronted by a multitude of bodily systems that have to be integrated into one more or less coherent body schema and a body image that include motor signals as well as multisensory signals (somato-sensory, vestibular, visual, auditory, visceral). First of all OBE are autoscopic body phenomena. Autoscopic phenomena as Bolognini et al (2012) put it, “refer to complex experiences involving the illusory reduplication of one’s own body”. One could add that OBE not only refer to the dislocation of body parts, but that introduce a global shift in the body image. According to Mohr & Blanke (2005) autoscopic phenomena (AP) are rare, illusory visual experiences how the excess of oxygen and breathing rate is a hallucinogenic “natural drug” experience, we can imagine how – on the contrary – a lack of oxygen in the brain can trigger specific experiences while the brain and the “one” that experiences, dies (if – of course – the people coming close to death, survive and can still report their experiences). 27 “Close your eyes and remember the last time you were walking on the beach. Is your visual memory one of looking out at the scene itself? Or is it of observing yourself, perhaps from above, walking along the coastline? For most people the latter is the case” (Metzinger, 2009, p. 87).

306 

Alexander Gerner

during which the subject e.g. has the impression of seeing or feeling a second own body in extrapersonal space. AP – in their view – consist in “out-of-body experience, autoscopic hallucination, and heautoscopy”. For Brugger et al (2006), Heautoscopy is the encounter with one’s double (the reduplication of a single body and self and thus a breakdown of integrative processes that let me identify with my body or my self), in the sense of a multimodal “illusory” reduplication of one’s own bodily self. Aspell et al (2012) express it as following: An individual undergoing an OBE usually experiences a dissociation between his self-location and his first-person visuospatial perspective with respect to the seen location of his own body – in other words, he perceives his own body (and the world) from a spatial location and perspective that does not coincide with the seen position of his body.

The phenomenon of polyopic heautoscopy (a multiplication of body and self) according to Brugger et al (2006) “points to the multiple mappings of the body, whose disintegration may give rise to the illusory experience of multiple selves.” Autoscopic phenomena deal with a) viewpoint changes, b) “illusionary” self-identification, c) altered or abnormal self-location(s) and d) changes in the first- person perspective (see: Aspell et al., 2012) as studies of Ehrsson 2007 and Langenegger et al. 2007 show by the manipulation of multisensory cues that the brain is supposed to use in order to “create a representation of self-location and identity” (Aspell et al., 2012). Blanke (2005) defines out-of-body experiences in the following manner: In an out-of- body experience, subjects feel that their “self ”, or centre of awareness, is located outside the physical body and somewhat elevated. (Blanke, 2005, p. 173)

OBE can be idiopathic, self induced or induced by non-invasive technological aid (Blanke & Metzinger, 2009) using for instance video (Lenggenhager et al., 2007), virtual reality (Ehrsson, 2007) or robotic devices (Ionta, 2011), inducing changes in the self-location, self-identification or first person perspective in healthy subjects. Moreover, recent research has not only described phenomenologically these strange doubling, mirroring or shadowing phenomena of a “disrupted” self (Mishara, 2010) but has shown as well that invasive manipulation of the brain can even induce a “illusory shadow person” (Arzy

Conceptual Personae of the “attentional self ”

307

et al., 2006) by artificial brain stimulation for instance in the paradigm of asynchrony stimulation of different modes of perception (e.g. tactile and visual). * 28 Metzinger, following Blanke calls 1) OBE, as well as 2) autoscopic hallucination and also 3) Heautoscopy “illusionary global own body perceptions” (Blanke/Metzinger, 2008, pp. 9–10) or autoscopic phenomena “leading to alterations of the minimal phenomenal self ” (ibid, p. 10).

Diagram 1, © Gerner. This diagram (adopted from: Blanke & Metzinger, 2008, p. 10) shows the dynamics of the attentional self-location [SL] in autoscopic experience: Three cases of direction of the attentional point of view, either from the hallucinated body towards the somatic body in Out-of-body experience (virtual observer perspective) or from the somatic body towards the hallucinated virtual body (somatic 1P perspective) in Autoscopic hallucination or both ways in Heautoscopy (switching between somatic 1P perspective and virtual observer perspective). The somatic and the virtual body in these three cases of observer perspectives always face each other on the contrary to a “felt presence” from behind (existential feeling perspective of autoscopy), another form of autoscopic experience.

These altered self phenomena are defined by “seeing a second own body in extracorporeal space” (see: Metzinger 2005; Blanke 2008 my underlining) and are phenomenologically distinct in the sense of the 28 Blanke (2005) in a very interesting article on the relation of heautoscopic experience and self-portrait in artists resumes this phenomenon as follows: “To conclude, one might say that autoscopic hallucinations and AH [autoscopic hallucination] self-portraits reflect that the self perceives its body from the outside and as a mainly visual body. The reduplicative character of the double is only implicitly present and the mechanisms of ‘seeing’ predominate over spatial and body-perceptual mechanisms. The self ‘sees’ its body as the body of somebody else.

308 

Alexander Gerner

a) self-identification, b) the direction of the visual weak 1st PP and c) the self-location either in the physical body as in autoscopic hallucination, or in the floating above body looking down onto the physical body as in OBE, or in the either/or direction of the weak 1st PP in the physical or illusionary body in Heautoscopy. All these illusionary global body perceptions can be caused by different pathological conditions in specific brain areas. In general heautoscopy and OBE are abnormal minimal phenomenal self-extensions towards a strong 1st PP in which by self-identification with the incorrect content of a global body representation attention is misled by wrong global body perceptions. This seems to be the difference of heautoscopy to the state of dreaming in which attention cannot be specifically directed. In an OBE self-location is never at the position of the physical body as represented, therefore OBE are associated with the feeling of “disembodiment (the experience that the subject of conscious experience is localized outside the person’s bodily borders)” (Blanke/Metzinger, 2008, p. 9) There is another non-standard phenomena of a feeling of a (nonvisual!) presence of another body behind the actual one, not seen but felt as a presence behind, that can also be induced by stimulation of certain brain areas (in this case the left angular gyrus, while in the same study by Blanke (2002), the stimulation of the right angular gyrus resulted in an OBE, “as if the patient were floating from the ceiling and looking down and being directed from the floating body onto herself ” (Metzinger, 2009, p. 96).



Out-of-body experiences and OBE self-portraits also reflect the self, perceiving its body from the outside, but not only as a visual body but as a body in a complex spatial scene of which the body is only a part. The reduplicative character of the double is explicit and the painter’s self is distanced and elevated from its body. The self is ‘out of touch’ with its somato-sensory perceptions and ‘sees’ its body with the eyes of somebody else. Finally, heautoscopy and heautoscopy self-portraits reflect the perception of the body from the inside via somatosensory and motor mechanisms. The self is in touch not only with her or his bodily feelings but also with the bodily feelings of the double. The self does not ‘see’ but mainly ‘feels’ its body and the body of the double to be the same body.”

Conceptual Personae of the “attentional self ”

309

Diagram 2, A felt presence behind, © Gerner.

In recent studies Blanke et al have discovered other brain areas that are supposedly involved in the causation of OBE when leisured brain areas are stimulated (as in seizure patients that report OBE): the extrastriate body area (on the right half of the temporo-parietal junction) and the left half of the temporo-parietal junction (Blanke et al. 2004; 2005). The brain region of the temporo-parietal junction (TPJ) is also activated in “less than half a second” (Metzinger, 2009, p. 97) if a test person is asked to imagine himself as if in the position of an OBE watching his body from above. “If this brain region is inhibited by a procedure called transcranial magnetic stimulation, this transformation of the mental model of one’s own body is impaired.” (Metzinger, 2008, p. 97) Metzinger, as Blanke, expresses two pathological conditions for OBE: 1) disintegration on the level of the self (or self-model in the words of Metzinger) 2) the conflict between the visual space and the vestibular sense of balance: Two separate pathological conditions may have to come together to cause an OBE. The first is disintegration on the level of the self-model, brought about by a failure

310 

Alexander Gerner

to bind proprioceptive, tactile and visual information about one’s body. The second is conflict between external, visual space and the internal frame of reference created by vestibular information, i.e., our sense of balance. In vertigo or dizziness, for example, we have problems with vestibular information while experiencing the dominant external, visual space. If the spatial frame of reference created by our sense of balance and the one created by vision come apart, the result could well be the conscious experience of seeing one’s body in a position that does not coincide with its felt position. […] Finally when an epileptic patient whose OBEs were caused by damage to the temporo- parietal junction was asked to stimulate mentally an OBE self-model, this led to a partial activation of the seizure focus. Taken together these observations point to an anatomical link among three different but highly similar types of conscious experiences: real, seizure-based OBs; intended mental simulations of OBs in healthy subjects, and intended mental simulations of OBEs in epileptic subjects (Metzinger, 2009, p. 96).

Important in my view is what Metzinger puts at the end of his neuronal explanation of OBE, and that is that in deviant or pathological conditions, as in the case of seizure patients, the phenomenology in these patients in comparison to healthy subjects differs in the first seconds of an OBE, and again show a different phenomenology in those you learned to influence mental states by meditation. Metzinger does not mention examples such as those of attentional absorption into a task or a situation such as in post-voluntary flow-experiences (See Bruya, 2010; Dormashev, 2010) which makes his assumption of getting closer to what the minimal form of selfhood is by examining non-standard cases of self experience epistemologically at least a more complex issue: The initial seconds clearly seem to differ between spontaneous OBEs in healthy subjects and those experienced by the clinical population, such as epileptic patients. The onset may also be different in followers of certain spiritual practices. Moreover, there could be a considerable neuro-phenomenological overlap between lucid dreams and OBEs as well as body illusions in general (Metzinger, 2009, p. 98).

Metzinger proposes – and indeed has to do so if OBE were to be of value to discovering the minimal phenomenal self – that OBE are general human experiences, and do not only appear in deviant or pathological conditions or are merely culturally based narrative inventions. As further examples he refers to situations such as meditation and I would add flow experiences (skilled experiences that had to be learnt with effort

Conceptual Personae of the “attentional self ”

311

and become habit in a way that they are experienced as post-voluntary effortless flowing, such as skiing down a ski slope after having learnt how to ski well), and the example he gives (but not designates as such) of a Marathon runner, where after some kilometers the woman goes into a trancelike state in which her body floats above her and her vision is of a bird-eye’s view watching the landscape from above. This unorthodox change in the perspective or the center of projection of the “I” I would designate as unorthodox attentional self maps, in which the dimensionality, the projection type and the material reference towards a real world scene is altered with attention occurring in an unusual way that we have no habitual experience of. It is a new map that occurs to us as an unusual or non-habitual orientation, but where we still can alter and move around the focus points we want to reach attentively. Seeing myself, however, floating out of my material body position is, however, an altered non-orthodox experience of my self. This, however, brings us deeper into the concept of a virtual “self-model” that attention supposedly controls. For Metzinger no such “things” as selves as substance exist, as he explained in the Introduction to his book “Being No one. The self-model theory of subjectivity”: […] no such things as selves exist in the world: Nobody ever was or had a self. The phenomenal self is not a thing but a process – and the subjective experience of being someone emerges if a conscious information processing system operates under a transparent self-model. (Metzinger, 2004, p. 1)

The fiction of the “self ” and its transparent model is said to have subjective selective advantages to form, for example, “higher forms of thoughts” (Ware, 2006, p. 1). In conscious experience Metzinger identifies three characteristics related to the self: a) Mineness, the notion that something belongs to the self (i.e. my thought, my feeling, my legs, my choice) b) selfhood, the idea that “I am a person”, the maintenance of an identity over time c) Perspectivalness a certain understanding of orientation in a representational space (giving a self a positioning and a relation to other representations) implying experience of a global attentional self and the relation of the self and some other object. The self, here defined as the “center of awareness”, is defined by Metzinger as

312 

Alexander Gerner

the possibility of being able to manipulate the focus of attention. This is the reason why with Metzinger we can call it an attentional self. In Metzinger’s representational view we have no direct access to the world, but the senses and their underlying neurobiological complex systems and functions create internal objects and contents that can be observed and manipulated by conscious “processing”. Metzinger proposes the elimination of an ontological self not for the sake of reductionism but for the sake of a more sound explanation of what conscious experience is said to be: It has now become clear that we will never solve the philosophical puzzle of consciousness- that is, how it can arise in the brain, which is a purely physical object- if we don’t come to terms with this simple proposition: that to the best of our current knowledge there is no such thing, no indivisible entity, that is us, neither in the brain nor in some metaphysical realm beyond this world. (Metzinger, 2009, p.1)

Thus Metzinger tries to address exactly the next question that arises from a “being no one” perspective of the self in conscious experience: So when we speak of conscious experience, what is the entity having these experiences? (Metzinger, 2009, p. 1)

Metzinger tries to answer this question of the first-person perspective through the intention to explain a) the theory of an apparent self within his naturalist “phenomenal Self-Model (PSM)” a multimodal plastic structure that will in principle for Metzinger be explained by neuro-scientific findings probably involving distributed processing and the b) the conceptual metaphor of the “Ego-Tunnel” as the “Consciousness tunnel that has evolved the additional property of creating a robust first person perspective, a subjective view of the world” (Metzinger, 2009, p. 11, my underlining). For him these two are the explanatory challenges in describing how “[…] a genuine sense of selfhood appears.” (Ibid, my underlining) He continues as follows: “We have to explain your experience of yourself as feeling the tactile sensation in the rubber hand, of yourself as understanding sentences you’re reading right now. This genuine conscious sense of selfhood is the deepest form of inwardness, much deeper than just being “in the brain” or in a “simulated world in the brain”. This

Conceptual Personae of the “attentional self ”

313

non-trivial form of inwardness is what the book is about.” (Metzinger, 2009, p.12, my underlining). He evokes the “transparency” of the underlying (neuroscientific) mechanisms of conscious experience as an argument for the process of a self- modeling its existence. In general Metzinger’s studies want to “contribute to a deeper understanding of non-conceptual, pre-reflective layers in conscious self-representation” (Metzinger, 2010, p. 25), in the embedded bodily self, that are distinct from for example the contents and the relations of visual perceptive experience in the “seeing self ” and what is perceived, in which he assumes the following theses: 1) Metzinger conceptually defends the claim that agency is not “part of the metaphysically necessary supervenience- basis for bodily self-consciousness.”(Metzinger ibid, see also: Blanke/Metzinger, 2008). But what are the constitutive conditions for selfhood after Metzinger, what is “truly necessary and not only sufficient to bring about an ego” (Metzinger, 2009, p. 101)? For Metzinger agency (see chapter 4 of his book and the phenomenon of the alien hand) is not a necessary condition, as it only selectively manipulates two dimensions: a) self-identification or “[…] the degree to which an organism identifies with the content of a global body representation” (Blanke/Metzinger, 2008, p. 7) or with the “content of a conscious body image” (Metzinger, 2009, p. 101) and b) self-localization as the embedding principle of the self in space. Self- localization is necessarily spatial-temporal including the feeling of “nowness”. For Metzinger the target phenomenon of consciousness/ conscious experience lies in basic minimal phenomenal embodied selfhood that can be “causally controlled by multisensory conflict alone” (ibid.) without any meta-control or author of action. For this minimal phenomenal self a “seeing” self for him is neither necessary as an “emotional” self nor a self-reflective or narrative self. The seeing self is not necessary because […] you can shut down the windows in front of the little man behind your eyes. The seeing self disappears. The Ego remains. You can be a robust, conscious self even if emotionally flat, if you do not engage in acts of will, and also in the absence of thought. Emotions, will and thoughts are not necessary to the fundamental sense of selfhood (ibid.)

314 

Alexander Gerner

2) On the level of the empirical and phenomenological foundation of his theory he wants to show how several phenomena of a) non-standard partial (body-) self experience such as the rubber hand illusion and b) “illusionary own global body perceptions” in which he includes “Out of Body experiences”, (virtually induced) “full body illusions”, “autoscopic hallucinations” and “Heautoscopy” that can be seen as an interesting “entry point” for researching into “the nature of the phenomenal self ” (Metzinger, 2010, p. 25) in general and the “Minimal Phenomenal Self ” (MPS) in particular. The MPS is related to embodiment and the simplest form of self-consciousness. Embodiment deals with the sense of “ownership” and identification with the body as a whole (the phenomenological sense of “mineness”) and is specifically important as it anchors the self, situating cognition through the subjective experience of having a body. One can, however, interpret the rubber hand phenomenon not merely as an “illusory” phenomenon, but as an extension or better an addition of an extra co-present object (the rubber hand) to the bodily self as Thomas Fuchs underlines: Functionally the rubber hand connects temporarily to the body, because by that the best possible coherence of different sense-modalities is installed […] this does not turn the sensed body into an illusion. (Fuchs, 2012, p. 33)

3) Metzinger tries to install a new research target in the investigation of the attentional self, which he calls the Self Model Theory of Subjectivity (SMT) in which the most important research entity – besides his questionable general constructivist “Phenomenal Self-Model” (PSM) – is what he calls the “Minimal Phenomenal Selfhood” Hence he differentiates epistemologically between different 1st Person perspectives (see Blanke/Metzinger, 2008, p. 7): He calls this first and basic state of a minimal phenomenal self, “selfhood-as-embodiment” (Metzinger, 2009, p. 102) or “weak 1 PP” (Blanke/Metzinger, 2008, p. 7), which he describes as a “calm, emotionally neutral state, deeply relaxed and widely alert, a state of pure observation, without any thought, while a certain elementary form of bodily self-consciousness remains” (Metzinger, 2009, p. 102). On his immanent bodily account of the self it is not a possessive self in the sense of “global ownership or “The feeling of this body being mine”, because global ownership already presupposes three entities: a) an

Conceptual Personae of the “attentional self ”

315

invisible self b) the body and c) a relation of body and self. On the contrary this basic selfhood for Metzinger is “the body that possesses itself ” (ibid.) The question of ownership is put into perspective in the following way: Owning means to be able to control it, and selfhood is intimately related to the very moment in which the body discovers that it can control itself- as a whole”(ibid.) The example he gives is the initial moment of “coming to yourself ” in the moment of waking up. The self is however not about control as such but about the enabling conditions for control. The minimal phenomenal self thus includes a bodily sense of space (spatial frame of reference) and time (the feeling of now), a body image and the transparency of this body image (the organism creating this image does not recognize it as an image). Moreover, the Here- and Nowness has to be accompanied by a “visual (or auditory) perspective originating within the body volume, a centre of projection, embedded in the volume of the body. (Ibid.)

*

Metzinger proposes a three levels of the self in which attention plays a significant role, however, only on the second and third level, as for him phenomenal “attentionality” would be only linked to a weak 1st PP in which in his view still there is no attention (in the sense of an active control mechanism of the focus of attention). Firstly, for Metzinger there is the weak 1st person perspective (1PP) a “purely geometrical feature of an egocentric model of reality” that is a spatio-multimodal model of reality (spatio-visual, spatio-auditory etc.). This includes a) a spatial frame of reference and b) a global body representation. Hence the body projects c) a perspective “originating within this body representation”. The weak 1st person perspective orientates the body through a centre of projection, “[…] which functions as the geometrical origin of seeing (or hearing etc.) Organism’s perspective.” I would call this the initial orienting, attentional bodily self. Secondly, and important in relation to attention in the sense of Metzinger is the “selfhood-as-subjectivity” (Metzinger, 2009, p. 102) or “strong 1 PP” (Blanke/Metzinger, 2008, p. 7), or stronger selfhood. The transition from the minimal phenomenal self to the self-as-subjectivity is marked -for him- by the supplementary selfrepresentation of the minimal selfhood as being directed towards an object that is by being able to change the location and duration of

316 

Alexander Gerner

attention in attentional agency or how he calls this by the “controlling the focus of attention”. Thus the stronger -not just minimal but also subjective phenomenal- self that first of induces a perspective after Metzinger’s hypothesis happens […] when we first discover that we can control the focus of attention. We understand that we can draw things from the fringe of consciousness into the centre of experience, holding them in the spotlight of attention or deliberately ignoring them – that we can actively control what information appears in our mind. Now we have a perspective, because we have an inner image of ourselves as actually representing, as subject directed at the world. Now we can, for the first time, also attend to our own body as a whole – we become self-directed. Inwardness appears. […] when we attend to the body itself […] Consciousness is the space of attentional agency. Selfhood as inwardness emerges when an organism for the first time actively attends to its body as a whole. If a global model of the body is integrated into the space of attentional agency, a richer phenomenal self emerges. It is not necessary to think. It is not necessary to move; the availability of the body as a whole for focal attention is enough to create the most fundamental sense of selfhood-as-inwardness – that is, the ability to become actively self-directed in attention. The body model now becomes a self- model […]. The organism is now potentially directed at the world and at itself at the same time. It is the body as subject (Metzinger, 2009, p. 102).

Thus the strong 1PP seems to be a global representation of the organism as a whole, given through a) Minimal Phenomenal Self and b) as being “directed” towards something, an “object component” (including the body itself). This direction of the 1PP towards a thing of experience for Metzinger necessarily includes the ability “to control the focus of attention”. For me this could be called the noting self. Thirdly, Selfhood as cognitive 1PP: The cognitive 1PP occurs “[…] when a system possesses a concept of the strong 1PP and is able to apply this concept to itself (i.e. it has an abstract and active mental representation of itself as a subject of experience which includes a special form of cognitive self-reference).” The cognitive 1PP presupposes the capacity of self-reference in the sense of using concepts such as “I” in “I (myself)-thoughts”. “I myself am a subject of experience”. This cognitive 1PP for Blanke/Metzinger (2008, p. 7) might be exclusively human “Many organisms might have phenomenal self-models, but perhaps only humans have self-concepts.”(ibid.). This could be called the self-attending (and self-observing) self.

Conceptual Personae of the “attentional self ”

317

2.2.2 Virtual Out- of-body experiences and impossible full body illusions Metzinger and Blanke (2008) amplify the notion of a partial body illusion as in the rubber hand illusion that I will sketch out here briefly. The rubber hand illusion takes up a strange phenomenon that Botvinick & Cohen (1998) called “Rubber Hand ‘Feels’ Touch That Eyes See” regarding an artificial body part lying on the table (the rubber hand) while the real left hand is hidden under the table (as the right arm and hand). Then a probe will repeatedly simultaneously stroke the visible rubber-hand, as will the real corresponding body part under the table. After a while (about a minute) the body image is redirected towards the visible rubber hand (body part), even creating a link (an illusory arm) that normally is accompanied with a sound or a crack felt when redirection of the virtual body part is active in this illusion. Metzinger and Blanke (2008) now distinguish between partial vs. global ownership of a body and carried out an experiment within a virtual reality setting to simulate a full body illusion and a swapping of the notion of one’s 1st PP towards an illusionary body in front of the test person. A head mounted display consisting of goggles that showed two separate images to each eye, creating the three-dimensional illusion of being in a virtual room. Subjects were able to see their own backs, which were filmed from a distance of two meters and projected into the three dimensional space in front of them with the help of a 3D decoder. When I acted as the subject of the experiment I felt as if I had been transposed into a 3D-version of René Magritte’s painting La Reproduction interdite. Suddenly I saw myself from the back standing in front of me. While I was looking at my back Bigna Lenggenhager was stroking my back, while the camera was recording this action. As I watched my own back being stroked, I immediately had an awkward feeling, I felt suddenly drawn towards my virtual body in front of me, and I tried to “slip into” it. This was as far as things went. (Metzinger, 2009, p. 99)

The most “interesting” outcome of these experiments (even though Metzinger’s conclusion of a reductive “geometrical view” of the self – and thus separated from biological matter and body structures – does not convince me) seems to be the conceptual conclusions Metzinger proposes after a series of tests with one’s own body being stroked (own body condition) or a fake body being stroked (fake body condition) and

318 

Alexander Gerner

an object being stroked synchronically and out of synch (control condition) the “embedding principle”: The bodily self is phenomenologically represented as inhabiting a volume in space, whereas the seeing self is a point without extension- namely, the centre of projection for our visio-spatial perspective, the geometrical origin of our perspectival visual model of reality. Normally this point of origin (behind the eyes, as if a little person were looking out of them as one looks out of a window) is within the volume defined by the felt bodily self. Yet as our experiments demonstrated, seeing and bodily self can be separated, and the fundamental self of selfhood is found at the location of the visual body representation. (Metzinger, 2009, p. 100)

3. Outlook We are confronted with lots of questions – that we cannot answer for now – but it is still important to clarify what the results of this research are, even if we do not have all the final answers but are rather confronted with posing future questions to be investigated. In this bold but strange account of the very interesting phenomenon of the so called “full body illusion” and the out of body experience the following questions and critical points, and the approximation towards the phenomenon of the “attentional self ” serve not as a conclusion but as an outlook within the impossible and the more possible fields of attention. It seems not so clear that epistemological questions concerning standard vs. non-standard cases of experience are easily compared. Does a dying brain always inform us about the living brain in all its functions or does a dying brain simply give us more insight into how a brain with impaired blood-flow or lack of oxygen alters our states of mind. Another point remains unclear: Why is it that the visual geometrical self should be the basic condition for the minimal phenomenal self (what about time consciousness, what about the basic feeling of being affected? What about touch and the skin as a compass and map of the body, that is not to be dislocated and touched elsewhere, as it is the proximate sense and not, as with the case of vision, the sense for approximating what is in the distance). Attention can arouse what is affected involuntarily through

Conceptual Personae of the “attentional self ”

319

sensory stimulation of one’s own body skin, though not from the stimulation of merely seeing another body stimulated. This question can be linked to Metzinger’s representative account of the self as a fiction of a transparent model in general: Metzinger himself asks […] The organism is now potentially directed at the world and at itself at the same time. It is the body as subject./ But again- who controls the focus of attention? In our Video Ergo sum study, who is the entity misidentifying itself? (Metzinger, 2009, p. 103)

There logically cannot be a deception happening in Metzinger’ s illusionary self-model if there is “no one” or “no entity” to be deceived in the first place by the “self ”. And still there is that feeling to know what it is like to be myself today this morning different from yesterday night when I finished this text in the year 2012, and still the one I am referring to and comparing the two notions of myself from yesterday and now, as well as Valéry’s question of what it is that makes up the center, what forms the center of attention? How are the episodic attentional self – Monsieur Teste – and the stabilizing and responding attentional self – Emilie Teste – interconnected, how are they “husband” and wife” inside a unified attentional self? These are deeper philosophical questions of ascribed identity that Metzinger leaves out in his “illusionary” account of the self. Again we can pose another question to be clarified in the future: Does the search for a minimal phenomenal self help in clarifying the question of conscious experience and the different modes of experience of an attentional self? To deeply link the different modes of attention such as in aesthetic and heautoscopic attention, attention in distraction (self-absorption, meditation etc.) and self-consciousness in relation to the non-self is beyond the scope of this paper but should be developed in the near future when describing a complete account of an attentional self. Thus, what the attentional self is and how it works in all its variety is an important future task of a philosophy of cognitive science. Is an attentional self an epistemological construct that joins separate research areas, but does not clarify what a non-dualist, or immanence account of the self should be about? This remains unanswered as well as the following question: How can OBE exactly help to clarify what the self is as a centre of projection and how does attentional control of focus actually stabilize the first person perspective? What role does distraction play in the

320 

Alexander Gerner

formation and the working of the self? Does attention therefore, besides attending with a specific sense modality, introduce a specific modus, in between virtuality (self) and the actuality of attending perception? If attention can follow the virtual body floating and look from above, then it should not necessarily be seen as a perceptive phenomenon, but as a virtual phenomenon that should also combine important conflicting inputs (as for instance visual and tactile) in a mereological account of a self. This also includes not only the notion of the visual body but the whole range of the multimodality of senses (somato-sensory, vestibular, visual, auditory, visceral) and especially the question how these sensible parts of the self are interconnected synchronically and even diachronically as well as synthesized in a global embodied self experience and linked to – what Metzinger calls – the orientation aspects of perspective (for him the weak 1st PP). The question remains however: Are these aspects merely representational and just “illusionary” or are they real and imminent in the bodily senses? These important questions that have arisen in our research show that it is fruitful to think more about attention and the self together in a mutual close relation and discover more about what can be called an “attentional self ” that we modestly started to explore with two of its conceptual persona.

References Aspel, J. et al. (2012). Multisensory Perception and Bodily Self-Consciousness. From Out-of-Body to Inside-Body Experience. In Murray, M., Wallace, M. (eds.). The Neural Basis of Multisensory Processes. CRC Press: Boca Raton online: . Assmann, A. & Assmann. J. (eds.) (2001). Aufmerksamkeiten. Archäo­ logie der literarischen Kommunikation VII. München: Fink. Blackmore, S. (1982). Beyond the body: An Investigation of Out-ofthe-body Experience. Granada: London. Blackmore, S. (1984). A psychological theory of the Out-of-Body Experience. Jour. Parapsychol. 48, 201– 218.

Conceptual Personae of the “attentional self ”

321

Blackmore, S. (1987). Where am I? Perspectives in Imagery and the Out-of-Body Experience. Jour. Mental Imagery, 11, 53– 66. Blackmore, S. (1993). Dying to live. Near Death Experiences. New York: Prometeus. Blanke, O., et al. (2002). Stimulation of Illusionary own body perceptions. Nature, 419, 269–270. Blanke, O. (2004). Out-of-body experiences and autoscopy of neurological origin. Brain, 127, 243–258. Blanke, O. (2005). Perception and experience of the self in autoscopic phenomena and self-portraiture. Schweizer Archiv für Neurologie und Psychiatrie, 156(4), 173–188. Blanke, O., Mohr, C. (2005). Out-of-body experience, heautoscopy, and autoscopic hallucination of neurological origin. Implications for neurocognitive mechanisms of corporeal awareness and self consciousness. Brain Research Reviews, 50(1), 184–199. Blanke, O., Metzinger, T. (2008). Full body illusions and minimal phenomenal selfhood. Trends in Cognitive Science, 13(1), 7–13. Blanke, O. et al. (2008). Illusionary Reduplication of the human body and self. Handbook of Clinical Neurology, 88, 429–458. Blumenberg, H. (2002). Zu den Sachen und Zurück. Frankfurt a. M.: Suhrkamp. Bolognini et al. (2011). Spatial perspective and Coordinate Systems in autoscopy. A Case Report of a “Fantôme de Profil” in Occipital Brain Damage. Journal of Cognitive Neuroscience, 23(7), 1741– 1751. Botvinick, M. & Cohen, J. (1998). Rubber hands “feel” touch that eyes see. Nature, 391, 756. Breyer, T. (2011). Attentionalität und Intentionalität. Grundzüge einer phänomenologisch-kognitionswissenschaftlichen Theorie der Aufmerksamkeit. München: Fink. Brugger et al. (2006). Polyopic heautoscopy: Case report and review of the literature.Cortex, 42, 666–674. Bruya, B. (ed.) (2010). Effortless Attention. A new perspective in the cognitive science of attention and action. Cambridge MA: MIT Press. Calvino, I. (1985 [1983]). Mister Palomar. Orlando: Harcourt Books. Deleuze, G. & Guattari, F. (1992 [1990]). What is philosophy? New York: Columbia University Press (WP).

322 

Alexander Gerner

Deleuze, G. (1998). What children say. In G. Deleuze, Essays critical and clinical. London: Verso, 61–67. Devinsky, O. (1989). Autoscopic Phenomena With Seizures. Arch. Neurol. 46(10), 1080–1088. Dormashev, Y. (2010). Flow experience explained on the grounds of an activity approach to attention. In Brian Bruya (ed.), Effortless Attention. A new perspective in the cognitive science of attention and action. Cambridge MA: MIT Press, 287–333. Ehrsson, H. (2007). The Experimental Induction of Out-of-Body Experiences. Science, 317, 1048. Ehrsson, H. (2012). The concept of body ownership and its relation to multisensory integration. In Barry E. Stein (ed.), The New Handbook of Multisensory Processes. Cambridge MA: MIT Press, 775–792. Engler, J. & Fulton, P. (2012). Self and No-Self in Psychotherapy. In Christopher Germer & Ronald Siegel (eds.), Wisdom and Compassion in Psychotherapy. New York: Guilford Press, 176–188. Fernandez-Duque, D. & Johnson, M. (1999). Attention metaphors: How metaphors guide the cognitive psychology of attention. Cognitive Science, 23, 83–116. Fernandez-Duque, D. & Johnson, M. (2002). Cause and effect theories of attention: The role of conceptual metaphors. Review of General Psychology, 6, 153–165. Fiebrich, A. & Gallagher, S. (2012). Joint Attention in Joint Action. Philosophical Psychology, 1–17, DOI:10.1080/09515089.2012.690176. Fuchs, T. (20124). Das Gehirn- ein Beziehungsorgan. Eine phänomenologisch-ökologische Konzeption. 4. Akt. u. erw. Aufl. Stuttgart: Kohlhammer. Fuchs, T. (2012b). Selbst und Schizophrenie. DZPhil, 60(6), 887–901. Gerner, A. (forthcoming). Philosophical Investigations of Attention. PhD Thesis in History and Philosophy of Science at the University of Lisbon (2012). Gerner, A. (2010a). Notes on Diagrams and Maps. In Olga Pombo, Alexander Gerner (eds.), Studies in Diagrammatology and Diagram Praxis. (=Logic and Cognitive Systems, Studies in Logic 24). London, College Publications.

Conceptual Personae of the “attentional self ”

323

Gerner, A. (2010b). Critical Notes on Two Forms of Reductionism: Some Shortcomings of Both Conceptual and Strong Neuroaethetic Accounts of Aesthetics”. In Olga Pombo, Silvia Di Marco, Marco Pina (Org.), Neuroaethetics. Can Science Explain Art? Lisboa: Fim de Século, 139–163. Head, H. (1923). Speech and cerebral localization. Brain, 46, 355–528. Lazar, S. et al. (2005). Meditation experience is associated with increased cortical thickness. Neuroreport 28, 16(17), 1893–7. Lutz, A. et al. (2008). Attention Regulation and monitoring in meditation. Trends Cogn Sci 2008, 12(4), 163–9. Manna, A. et al. (2010). Neural correlates of focussed attention and cognitive monitoring in meditation. Brain Res Bull, 82(1, 2), 46–58. MacLean, K. et al (2010). Intensive Meditation Training Improves Perceptual Discrimination and Sustained Attention. Psychological Science, 21(6), 829–838. Mackworth, N. H. (1948). The breakdown of vigilance during prolonged visual search. Quarterly Journal of Experimental Psychology 1, 6–21. Metzinger, T. (2004). Being No-one. The Self-Model Theory of Subjectivity. Cambridge MA: MIT Press. Metzinger, T. (2005). Out-of-Body Experiences as the Origin of the Concept of a “Soul”. Mind & Matter, 3(1), 57–84. Metzinger, T. (2009). The Ego Tunnel. The science of the mind and the myth of the self. New York: Basic Books. Metzinger, T. (2010). The Self-Model Theory of Subjectivity: A brief Summary with Examples. Humana.Mente Journal of Philosophical Studies, Vol. 14, 25–52. retrieved online 17.12.2010: . Neisser, U. (1988). Five kinds of Self-Knowledge. Philosophical Psychology, 1, 35–59. Olendzki, A. (2010). Unlimiting Mind: The radically experiential psychology of Buddhism. Boston: Wisdom. Ortega y Gasset, J. (2007[1943]). Meditation on Hunting. Belgrade, CA: Wilderness Adventures Press. Parasuraman, R. (1976). Consistency of individual differences in human vigilance performance: An abilities classification analysis. Journal of Applied Psychology, 61, 486–492.

324 

Alexander Gerner

Parasuraman, R. & Davies, D. R. (1977). A taxonomic analysis of vigilance. In R. R. Mackie (ed.), Vigilance: Theory, operational performance, and physiological correlates. New York: Plenum, 559–574. Parasuraman, R. (1987). Human computer monitoring. Human Factors, 29, 695–706. Parasuraman, R. (1984a). The psychobiology of sustained attention. In J. S. Warm (ed.), Sustained attention in human performance. Chichester, UK: Wiley, 61–101. Parasuraman, R. (1984b). Sustained attention in detection and discrimination. In R. Parasuraman, R. & Davies, D. (eds.), Varieties of attention. Orlando, FL: Academic Press, 243–271. Parasuraman, R., Warm, J. S. & See, J. E. (1998). Brain systems of vigilance. In R. Parasuraman (ed.), The attentive brain. Cambridge, MA: MIT Press, 221–256. Parasuraman, R. & Wilson, G. (2008). Putting the brain to work: Neuroergonomics past, present, and future. Human Factors, 50, 468–474. Peirce, C. S. (1994). “Application to the Carnegie Institution, July 15, 1902”. Manuscript L75, Analytical reconstruction and editorial work by Joseph Ransdell, Indiana University. Pöppel, E. (2006). Der Rahmen. Ein Blick des Gehirns auf unser Ich. Stuttgart: DTV. Strawson, G. (2004). Against Narrativity. Ratio (new series) XVII, 428–452. Tomasello, M. & Carpenter, M. (2007). Shared intentionality. Developmental Science, 10(1), 121–125. Valéry, P. (1989[1973]). Monsieur Teste. The Collected Works of Paul Valéry, ed. by Jackson Mathews (=Bollingen Series XLV, 6). Princeton: Princeton University Press Valéry, P. (1946). Monsieur Teste. Paris: Gallimard. Waldenfels, B. (2006). Grundmotive einer Phänomenologie des Fremden. Frankfurt a. M.: Edition Suhrkamp. Waldenfels, B. (2004). Phänomenologie der Aufmerksamkeit. Frankfurt a. M.: Suhrkamp. Waldenfels, B. (2002). Bruchlinien der Erfahrung. Phänomenologie, Psychoanalyse, Phenomenotechnik. Frankfurt a. M.: Suhrkamp. Warm, J. S. et al. (2008). Vigilance requires hard mental work and is stressfull. Human factors, 50, 3, 433–441.

Notes on Contributors

Eric T. Olson – Professor of Philosophy at University of Sheffield. Rui Vieira Da Cunha – Member of MLAG (Mind, Language and Action Group) and Research Fellow at the Institute for Philosophy, University of Porto. Klaus Gärtner – Doctoral Research Fellow of Institute for Philosophy of Language, New University of Lisbon. Clara Morando – Doctoral Research Fellow at the Institute for Philosophy, University of Porto. Dina Mendonça – Post-Doctoral Research Fellow of Institute for Philosophy of Language, New University of Lisbon. António Marques – Full Professor of Philosophy of Knowledge and Communication at the New University of Lisbon. Erich Rast – Post-Doctoral Research Fellow of Institute for Philosophy of Language, New University of Lisbon. Vasco Correia – Post-Doctoral Research Fellow of Institute for Philosophy of Language, New University of Lisbon. João Fonseca – Post-Doctoral Research Fellow of Institute for Philosophy of Language, New University of Lisbon. Jorge Gonçalves – Post-Doctoral Research Fellow of Institute for Philosophy of Language, New University of Lisbon. Robert Clowes – Post-Doctoral Research Fellow at the University of Sussex. Alexander Gerner – Post-Doctoral Research Fellow of the Center of Philosophy of Science, University of Lisbon.