rethinking introspection: a pluralist approach to the first-person perspective

New Directions in Philosophy and Cognitive Science

Series Editors: John Protevi , Louisiana State University and Michael Wheeler , University of Stirling

This series brings together work that takes cognitive science in new directions. Hitherto, philosophical reflection on cognitive science – or perhaps better, philo-sophical contribution to the interdisciplinary field that is cognitive science – has for the most part come from philosophers with a commitment to a representa-tionalist model of the mind.

However, as cognitive science continues to make advances, especially in its neuroscience and robotics aspects, there is growing discontent with the represen-tationalism of traditional philosophical interpretations of cognition. Cognitive scientists and philosophers have turned to a variety of sources – phenome-nology and dynamic systems theory foremost among them to date – to rethink cognition as the direction of the action of an embodied and affectively attuned organism embedded in its social world, a stance that sees representation as only one tool of cognition, and a derived one at that.

To foster this growing interest in rethinking traditional philosophical notions of cognition – using phenomenology, dynamic systems theory, and perhaps other approaches yet to be identified – we dedicate this series to ‘New Directions in Philosophy and Cognitive Science’.

Titles include:

Robyn Bluhm, Anne Jaap Jacobson and Heidi Maibom ( editors ) NEUROFEMINISM Issues at the Intersection of Feminist Theory and Cognitive

Jesse Butler RETHINKING INTROSPECTION A Pluralist Approach to the First-Person Perspective

Julian Kiverstein & Michael Wheeler ( editors ) HEIDEGGER AND COGNITIVE SCIENCE

Michelle Maiese EMBODIMENT, EMOTION, AND COGNITION

Richard Menary COGNITIVE INTEGRATION Mind and Cognition Unbounded

Zdravko Radman ( editor ) KNOWING WITHOUT THINKING Mind, Action, Cognition and the Phenomenon of the Background

Matthew Ratcliffe RETHINKING COMMONSENSE PSYCHOLOGY A Critique of Folk Psychology, Theory of Mind and Stimulation

Jay Schulkin ( editor ) ACTION, PERCEPTION AND THE BRAIN

Forthcoming titles:

Anne Jaap Jacobson KEEPING THE WORLD IN MIND Biologically Embodied Representations and the New Sciences of the Mind

Hanne De Jaegher PARTICIPATION SENSE-MAKING An Enactive Approach to Intersubjectivity

Robert Welshon NIETZSCHE, PSYCHOLOGY, AND COGNITIVE SCIENCE

Charles T. Wolfe ( editor ) BRAIN THEORY Essays in Critical Neurophilosophy

New Directions in Philosophy and Cognitive Science Series Standing Order ISBN 978–0–230–54935–7 (Hardback) 978–0–230–54936–4 (Paperback) ( outside North America only )

You can receive future titles in this series as they are published by placing a standing order. Please contact your bookseller or, in case of difficulty, write to us at the address below with your name and address, the title of the series and one of the ISBNs quoted above.

Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire RG21 6XS, England

Rethinking Introspection A Pluralist Approach to the First-Person Perspective

Jesse Butler University of Central Arkansas, USA

© Jesse Butler 2013 Softcover reprint of the hardcover 1st edition 2013 978-1-137-28037-4

All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission.

No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6–10 Kirby Street, London EC1N 8TS.

Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages.

The author has asserted his right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988.

First published 2013 by PALGRAVE MACMILLAN

Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS.

Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010.

Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world.

Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries

ISBN 978-1-349-44767-1 ISBN 978-1-137-28038-1 (eBook)DOI 10.1057/9781137280381

This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin.

A catalogue record for this book is available from the British Library.

A catalog record for this book is available from the Library of Congress.

10 9 8 7 6 5 4 3 2 1

22 21 20 19 18 17 16 15 14 13

For Blue

Contents

Series Editors’ Foreword ix Acknowledgements xi

Introduction 1

1 Introspection as Inner Perception 6 Historical overview 7 Perceptual models of introspection in psychology 10 Inner perception in contemporary philosophy of mind 12

2 Poking Out the Inner Eye 16 Is there a mind’s eye in the brain? 17 What are the objects of mental perception? 21 Do mental states exhibit perceptual qualia? 28 Perception, evolution, and knowledge of minds 31 Perception and recursion 36 Introspection and attention 37

3 Introspection as a Metaphor 41 The prominence of perceptual metaphor 41 What is introspection? 45 The diversity of introspection 48

4 Knowing Our Own Consciousness 51 Knowing what it is like 52 The inadequacy of standard models

of phenomenal knowledge 55 The existential constitution model

of phenomenal knowledge 62 Objections and clarifications 65 Putting skepticism of self-knowledge in its place 69 Conclusion 73

5 Introspection through Cognition 75 Representation, metarepresentation, and misrepresentation 76 Conceptualizing our own mental states 84 Attention and introspection 90 Introspection and the extended mind 94 Conclusion 97

vii

viii Contents

6 Understanding Our Own Beliefs and Desires 99 Folk psychology and first-person authority 102 The theory theory challenge to first-person authority 103 Empirical evidence of theory-mediated introspection 106 Mind-monitoring mechanisms and the phenomenology

of propositional attitudes 112 Conclusion 116

7 The Internal Monologue 119 Knowing our own thoughts through language 120 Objections and clarifications 130 Kinds of self-knowledge enabled by inner speech 137 From self-determined truth to self-deception and back 140 Conclusion 147

8 On the Social Side of Self-Knowledge 148 Social self-knowledge 149 Navigating one’s position in social space 151 Direct and indirect testimony about oneself 153 Bias and the rationality of testimonial deference 156 Conclusion 160

Conclusion: Is That All There Is? 161

Notes 165

References 169

Index 177

Series Editors’ Foreword

One of the oldest and yet most difficult of the ancient Greek philo-sophical imperatives is ‘Know thyself’. A perhaps equally ancient asso-ciation couples vision – theory – and truth. Combining the two leads us to introspection: the route to knowing yourself is to look within yourself. As venerable as the introspective tradition is, it came under severe attack when philosophy and scientific psychology became allies. Cast aside for its impressionism and relativism, introspection spent the twentieth century banished from scientifically oriented philosophy, lost in the wilderness of poets, novelists, and phenomenologists.

If introspection is to be readmitted to polite company then, Jesse Butler argues, it will have to be rethought. And that rethinking shows it not to be a literal inward vision – despite a long series of philosophers who have used language suggesting that it is – but a metaphor grouping together multiple and heterogeneous processes.

The most basic of those processes, Butler shows, is revealed by the phenomenological method, which is not, pace its enemies , navel-gazing yielding subjective impressions, but rather the careful demonstration that our basic cognitive self-relation is non-reflective subjective aware-ness. As Butler puts it, ‘We know ourselves by being ourselves, not as objects but rather through the qualitative character of our experiences as living subjects’. This is a unique sort of knowledge, Butler claims, irreducible to the types of knowledge currently discussed in analytic philosophy of mind. Butler calls it the ‘existential constitution’ model and discusses it in relation to the classical phenomenologists and to contemporary thinkers.

Butler does not stop there, however. He goes on to show how non-reflective self-awareness is only one of our ‘introspective’ capacities; we can and do objectify ourselves for genuine and valid cognitive purposes. Such higher-level, reflective introspection occurs through the use of normal outwardly directed cognitive processes such as representation, conceptualization, and attention that we turn back on ourselves. To complete the picture, Butler also discusses the self-application of Theory of Mind, inner speech, and sociality and self-knowledge.

ix

x Series Editors’ Foreword

All in all, then, Butler’s Rethinking Introspection shows how the ancient theme of introspection, when brought back from exile and appropri-ately rethought, offers widespread and important current application.

John Protevi, Louisiana State University Michael Wheeler, University of Stirling

Acknowledgements

This book has developed over a considerable span of time, warranting the acknowledgement of many more than I can list here, starting with all the friends and family who have supported me throughout the years. In the end, it all comes back to you. I give my love and gratitude to you all, especially Julie, Lily, and Miles, who have endured more than their fair share of the escapades behind the production of this work. You have done much more for this book than the pages reveal.

Although much has changed since, this book began as my Ph.D. dissertation at the University of Oklahoma, where I benefited from the fine guidance of Jim Hawthorne, Chris Swoyer, Ray Elugardo, Wayne Riggs, Lynn Devenport, and many others while carving out the basic framework that has become this book. Many more people have contrib-uted to the discussion and refinement of the content since, especially my friends and colleagues at the University of Central Arkansas and all the folks who offered their thoughts on my work at conferences. I thank you all for your helpful feedback and kind support.

Some of the work presented in Chapter 4 appeared previously in my article ‘Introspective Knowledge of Experience and Its Role in Consciousness Studies’ in the Journal of Consciousness Studies , which I thank for permission to republish portions of that article here. The content of Chapter 8 (and aspects of Chapter 7 ) was initially mapped out in 2009 through a summer research stipend from the University Research Council at the University of Central Arkansas. I am grateful for the support.

xi

1

When considering our first-person knowledge of our own minds, whether with regard to immediate awareness of a sensory experience, reflection upon one’s own beliefs or desires, deliberation concerning one’s character, or metaphysical soul-searching about one’s ultimate nature, we commonly appeal to the concept of introspection. But what exactly are we talking about when we speak of introspection? Do we liter-ally perceive our own mental states, as the term seems to suggest, or are there other processes at work in our ability to know our own minds? Can introspection be trusted as a viable or even privileged source of knowl-edge, as many have claimed and many others have simply assumed, or are there significant barriers, problems, and limits regarding what we can know about our own minds from our own first-person perspective? This book offers a pluralistic framework for understanding these issues regarding the nature and epistemic properties of introspection. At the core of this framework is the idea that introspection is a multi-faceted phenomenon that cannot be limited to a single cognitive mechanism or epistemic characterization. There are many different ways in which we engage in introspection and, correspondingly, a variety of different epistemic dimensions involved in our first-person understanding of our own minds. Through presentation and analysis of these various aspects of introspection, I illustrate how we know, and sometimes fail to know, our own minds.

The first step in this reevaluation of introspection is to recognize that the standard and somewhat intuitive understanding of introspection as an inner perceptual faculty is mistaken. People often think of intro-spection literally, as inner perception, but we do not really perceive the contents of our own minds, inside a so-called mind’s eye. As I will argue, this perceptual account of introspection is a result of pervasive

Introduction

2 Rethinking Introspection

and unintentional use of perceptual metaphors in our general under-standing of knowledge. This is not to say that we should stop using the concept of introspection to refer to our first-person knowledge of our own minds, however. Just as we continue to talk of the sun rising and setting while knowing that the rotation of the earth is the actual cause of what we are observing, we can continue to employ the term ‘intro-spection’ to refer to our first-person knowledge of our own minds while recognizing that what we are referring to is actually a heterogeneous collection of different ways of knowing our own minds.

With this clarification regarding the concept of introspection estab-lished, we can ‘see’ the diverse phenomena at work behind our ability to know our own minds from our own first-person perspective. First, there is the experiential knowledge that we have of our own conscious mental states simply in virtue of being ourselves, as conscious beings embodied in the world. This very basic component of self-knowledge, which I characterize as a kind of phenomenal knowledge that is intrinsic to one’s own conscious experience, is the topic of Chapter 4 . Through the development of what I call the existential constitution model of phenomenal knowledge, I argue that this phenomenal knowledge is a unique kind of knowledge that plays a constitutive role in our knowl-edge of our own minds. Next is the higher-level introspective knowl-edge we acquire through our mind’s representational, conceptual, and attention-facilitated abilities, which we routinely use in our engagement with the external world but can also recursively apply to our own mental states. I introduce these cognitive processes as aspects of our introspec-tive capacities in Chapter 5 , defending a somewhat reductionist account of the cognitive processes involved in higher-order introspection. It is a reductionist account in the sense that it explains our introspective abilities through faculties that are already present in human cognitive systems, irrespective of any particular introspective functions they may perform, thereby explaining our high-level capacities to reflect upon our own minds through the more general cognitive capacities of the human mind. In Chapter 6 , I develop this account further by analyzing a somewhat more specific element of this domain: the use of folk psychology, the common-sense understanding of mental states that enables us to conceptualize others as intentional beings with their own beliefs and desires, to understand prominent aspects of ourselves. Drawing upon some important and revealing empirical discoveries, I argue that we often understand ourselves through the application of folk psychological concepts to our own mental states, thereby under-standing ourselves through the same interpretive structures that we use

Introduction 3

to understand the minds of other people. In Chapter 7 , I present and discuss several different ways in which language, through the promi-nent phenomenon of inner speech, contributes to our ability to know our own minds, but also simultaneously allows for epistemic error and even self-deception. In the final chapter, I take a look at the social side of introspection, considering how our introspective knowledge extends into the social processes that surround our lives. Taken together, these various faculties and processes provide a viable framework for under-standing introspection.

Throughout my discussion of these rather diverse aspects of intro-spection, one key goal is to map out the epistemic dimensions of our capacity to know, and fail to know, our own minds from our first-person perspective. Corresponding to the fundamentally different kinds of introspective processes outlined above, there are distinctly different epistemic characterizations of the nature and extent of our knowledge of our own minds. As I will argue, some introspective faculties provide unique ways of knowing our own minds that do not fit standard models of knowledge and are not subject to typical epistemic error, such as the experiential self-knowledge intrinsic to being in a conscious state and the self-determining truth values of some types of inner speech utterances (such as thinking the thought ‘I am thinking’ to oneself). However, many other kinds of introspection are subject to the possi-bility of epistemic error, such as the capacity to misrepresent anything that is mediated through representational concepts and the capacity to deceive ourselves through the narratives we construct with inner speech commentary upon ourselves and our experiences. The consideration of these various traits illustrates how introspection includes a rather broad range of epistemic properties that cannot be neatly contained within a one-dimensional characterization of the epistemology of introspection.

By drawing attention to this broad epistemic range of introspection, I hope to clarify the needless and misguided debate concerning whether introspection is special and privileged or fallible and untrustworthy. Can we trust introspection? This question simply cannot be answered with a singular yes or no. Instead, introspection, as a diverse phenom-enon, is spread out across numerous characteristics, some of which are trustworthy and some of which are not. Realization of this fact is long overdue in the various disciplines investigating the human mind, where perspectives on introspection have been polarized by one-dimensional characterizations, from the categorically dismissive attitude that has been prominent in mainstream psychology since the decline and


rejection of the early ‘introspectionists’ to those in philosophy who regard introspection as a privileged, authoritative, and/or foundational source of knowledge. Introspection cannot be accurately squeezed into either of these characterizations, and is best regarded as a multi-faceted phenomenon with both privileged and fallible epistemic elements.

In addition to offering a pluralist account of introspection, this book is methodologically pluralistic. I will follow a naturalistic but meth-odologically diverse explanatory approach, drawing upon a variety of fields of study for both conceptual development and supporting evidence. This book is first and foremost a work in philosophy, but it also integrates ideas and observations from relevant sciences, including cognitive, social, and evolutionary psychology, neuroscience, linguis-tics, and sociology. Even within the domain of philosophy, it will be methodologically diverse and integrative. Much of what I will address stems from, and critically engages with, work in contemporary analytic philosophy of mind, but I will also utilize and engage in varying degrees with phenomenology, existentialism, pragmatism, and ordinary language philosophy, insofar as they present relevant considerations for understanding the phenomena involved in introspection. I do not draw any principled exclusionary lines in the sand between theoretical orientations, contrary to the common tendency to divide philosophy into warring camps pitted against one another. Instead, I prefer to draw upon multiple theoretical backgrounds, pragmatically determined by whatever considerations are needed to pursue truth, understanding, and clarification regarding the topic at hand.

It is worth noting that this work is not unique in following a pluralist methodology. For example, some philosophers concerned with under-standing the mind have advocated a triangulated approach, such that experiential phenomenology, explanatory theory, and empirical obser-vation are all each given their due in approaching a sufficient under-standing of the mind (e.g., Flanagan, 1992 ; Gallagher, 2008 ). I agree that all three of these factors must be taken together if we are to progress in our understanding of the mind, and likewise follow a similar pluralist approach in this book.

Not all accounts and perspectives are equally plausible, however. It is important to note that pluralism does not entail that all methods, accounts, and/or explanations are to be accepted. Indeed, as we will see in the first two chapters of this book, some accounts do not hold up to sustained pluralistic inquiry, even though they may pass muster from one particular perspective or other. In approaching truth and avoiding falsity, our claims about the mind ought to be held accountable to all

Introduction 5

applicable standards together, including empirical evidence, logical consistency, theoretical viability, phenomenological accuracy, and plain ole common-sense plausibility, none of which is sacrosanct alone, and some of which can trump others with enough support. Through the use of multiple explanatory methods and theories held to these stand-ards, the hope is that we can weed through false claims and misguided conceptions, approaching an accurate and well-rounded understanding of our introspective capacities.

Before we get started, however, I ought to emphasize one more thing about the approach I will take in this book. There is a growing consensus, across disciplinary and theoretical boundaries, towards a naturalistic understanding of the human mind and this book follows suit with this as well. Through multiple well-known developments (e.g., evolu-tionary biology, neuroscience, anthropology, human ecology), it has become quite clear that we are natural beings, biologically embedded in the physical world. We are not transcendent disembodied egos, what philosopher Gilbert Ryle aptly and derogatorily labeled Cartesian ‘ghosts in the machine’, that somehow mysteriously and only contingently inhabit our bodies ( 1949 ). Instead, we are our bodies, thoroughly phys-ical and inextricably situated within the natural world. I will assume this naturalized perspective in this book. The primary aim, however, is not to defend this naturalized perspective, at least not directly. Instead, what I offer here is a way of thinking about introspection that is both compatible with and developed from a broadly naturalistic and embodied approach to the mind. Readers who do not follow this natu-ralistic approach may still find something worthwhile in the book, however, insofar as they may be interested in understanding the nature of introspection and self-knowledge regardless of the broader nature of the mind. I will regularly and unapologetically write from a naturalistic perspective, but not everything I say is dependent upon that perspec-tive being true. If my overall account is right, however, some cherished anti-naturalistic ideas may be brought into serious doubt, including especially the notion that we are transcendent immaterial observers, perched in a privileged position to perceive the contents of our own minds. Despite much progress in overturning the picture of ourselves as disembodied observers that we inherited from Descartes and others, the idea of a perceiving ego within the head persists in much thought about the mind today, sometimes in quasi-naturalized form. As we will see over the next few chapters, this misguided notion must be pulled out by the roots, clearing the way for a more accurate understanding of how we know (and fail to know) our own minds.

6

Literally speaking, the term ‘introspection’ means to look within. The concept of introspection thereby suggests an ability to perceive within oneself. If we regard introspection in this literal manner, we are led to posit a perceptual faculty in the mind through which we observe our own mental states. Just as we perceive the external world through specific perceptual mechanisms, such as eyes, ears, and skin, we might perceive internal events through some similar sort of mechanism or process that is specially aimed toward the perception of mental states. A number of people, both past and present, have regarded introspection along these lines. Moreover, ordinary talk of a ‘mind’s eye’ or ‘looking within’ oneself likewise suggests some kind of perceptual process regarding introspection. However, as I will argue later, there are good reasons for thinking that this view of introspection is mistaken as a literal account of what goes on in our minds when we come to know our own mental states. Dismissing this common view of introspection will open the concept of introspection to a broader range of under-standing and analysis. It will allow us to conceive of introspection not as a singular mechanism or process, but rather as a metaphor for our diverse and heterogeneous capacities to obtain knowledge of our own minds.

Before explaining why we should dismiss the perceptual account of introspection, however, I will first survey how common and prominent this viewpoint has been and, to a large extent, continues to be, through an overview of perspectives that express, defend, or tacitly assume some version or other of the perceptual account of introspection. My purpose here is not to give a comprehensive survey or analysis of these view-points, but rather to illustrate the commonality of the perceptual model as it appears both throughout history and in contemporary thought.

1 Introspection as Inner Perception

Introspection as Inner Perception 7

Readers already familiar with perceptual accounts of introspection in philosophy and science may wish to skip ahead to the next chapter, though the material here will clarify the target of the criticisms I will make there.

Historical overview

First of all, many historically significant and influential philosophers either assumed or endorsed a perceptual account of introspection. Although some earlier figures might be noted here, such as Aristotle, Augustine, and Aquinas (Lyons, 1986 , pp. 1–2; Wallace, 2000 , p. 76), let us begin with the father of modern philosophy, Ren é Descartes, at the dawn of what William Lyons dubbed ‘the golden age of introspec-tion’ ( 1986 , p. 2ff). As you may know, Descartes sought an indubitable foundation for knowledge through first-person reflection. At the core of this pursuit was his reliance upon his introspective knowledge of his own mind. Through ‘clear and distinct’ (seemingly) observation of the contents of his own mind, Descartes constructed his epistemic and metaphysical framework for philosophy. Consider, for example, his (in)famous observation of himself as an immaterial soul:

Then I looked carefully into what I was. I saw that while I could pretend that I had no body and that there was no world and no place for me to be in, I still couldn’t pretend that I didn’t exist. I saw on the contrary that from the mere fact that I thought about doubting the truth of other things, it followed quite evidently and certainly that I existed; ... This taught me that I was a substance whose whole essence or nature is simply to think [here = ‘to be in conscious mental states’], and which doesn’t need any place, or depend on any material thing, in order to exist. Accordingly this me – this soul that makes me what I am – is entirely distinct from the body, is easier to know than the body, and would still be just what it is even if the body didn’t exist (Descartes, 1637 , Part IV).

There is much to say about this passage and the reasoning behind it, which arguably set the study of the mind on a misguided course from which we are still recovering today (see, e.g., Wheeler, 2005 ; Rowlands, 2010 ), but for our immediate purposes the thing to note is the emphasis on inner observation. The use of the words ‘looked’ and ‘saw’ is particu-larly revealing here; Descartes thought he could clearly and distinctly perceive his own nature as a conscious being, apart from his fallible


sensory-mediated perceptions of the outer world, including his own body. In this respect, Descartes conceived of introspection as a kind of inner perception, albeit a very special and epistemically privileged perception that is quite distinct from ordinary sensory perception. To be precise, the elevated epistemic status of inner observation sets Descartes’ particular brand of introspection apart from other perceptual accounts of introspection, many of which regard introspection as fallible and epistemically on par with external perception. In fact, for this reason, some distinguish Cartesian introspection from perceptual accounts of introspection (e.g., Gertler, 2011 ), but his repeated emphasis on inner observation, irrespective of its epistemic standing, clearly places him in the perceptual account camp, broadly construed.

Moving along in history, the conception of introspection as a kind of perception is more explicitly articulated by the classical empiricist John Locke. After discussing the perception of sensible objects via ideas present in the mind, Locke goes on to say:

Secondly, The other Fountain, from which Experience furnisheth the Understanding with Ideas , is the Perception of the Operations of our own Minds within us, as it is employ’d about the Ideas it has got; which Operations, when the Soul comes to reflect on, and consider, do furnish the Understanding with another set of Ideas , which could not be had from things without: and such are, Perception, Thinking, Doubting, Believing, Reasoning, Knowing, Willing, and all the different actings of our own Minds; which we being conscious of, and observing in ourselves, do from these receive into our Understandings, as distinct Ideas , as we do from Bodies affecting our Senses. This Source of Ideas , every Man has wholly in himself: And though it be not Sense, as having nothing to do with external Objects; yet it is very like it, and might properly enough be call’d internal Sense. But as I call the other Sensation , so I call this REFLECTION , the Ideas it affords being such only, as the Mind gets by reflecting on its own Operations within it self. (Locke, 1689 /1975, Book II, Chapter I , Section 4, p. 105).

So, according to Locke, we perceive the inner workings of our minds through a kind of observational faculty, which he calls ‘reflection’. Although he distinguishes this faculty from ‘sensation’, reserving the latter term for our perceptual access to the external world, Locke regards this internal operation of the mind as similar enough to ordinary sense perception to warrant the characterization of it as an ‘internal Sense’. He speaks of introspective reflection as observing the operations of the


mind, perceiving them in much the same way that we perceive external objects. The epistemic status of Lockean inner sense is thereby quite different from Cartesian introspection, but both share the idea that we conscious beings have a first-person mode of observation toward our own minds.

A similar inner sense account of introspection is found in the work of Immanuel Kant, who inherited and transformed many of the ideas of the modern period, both rationalist and empiricist. In the Critique of Pure Reason , Kant speaks repeatedly of ‘inner sense’ as a faculty by which we perceive mental phenomena. For instance, in regard to our knowledge of both internal and external objects, Kant states that

The relation of sensibility to an object and what the transcendental ground of this [objective] unity may be, are matters undoubtedly so deeply concealed that we, who after all know even ourselves only through inner sense and therefore as appearance, can never be justified in treating sensibility as being a suitable instrument of investigation for discovering anything save always still other appear-ances ... (Kant, 1781 /1929, A278/B334, p. 287)

The basic idea here is that sensory objects can only be known as appear-ances (rather than as they are ‘in themselves’), and that this applies not only to external objects but also to internal mental objects as well, including the self. So, just as we come to know of tables, chairs, trees, and people through the appearances they produce in our senses, we likewise come to know of our own mental lives by their appearing to us through some kind of inner sensory modality. While there may be some difficulties in translating Kant’s own rather idiosyncratic terminology over to our now common talk of introspection, the similarities are clear enough to show that Kant, like many philosophers and theorists before and after him, thought that we obtain knowledge of our own minds in some kind of perceptual manner, whatever the specific mechanism of this inner sense may be.

Moving along toward the twentieth century, William James is another notable philosopher (and psychologist) who characterizes introspec-tion as a kind of perception. In describing his general methodology for psychology, James famously states that:

Introspective observation is what we have to rely on first and foremost and always. The word introspection need hardly be defined – it means, of course, the looking into our own minds and reporting what we there


discover. Every one agrees that we there discover states of consciousness. (James, 1890 , vol. 1, chapter VII )

This is the most direct and confident commitment to the perceptual model of introspection we have seen thus far, perhaps illustrating the solidification of this perspective in the investigation of the mind over time. James asserts that it is so obvious that introspection is a kind of mental perception that it need not even be explained, and regards this obviousness as so fundamental that it provides his foundation for empirical psychological research.

James exhibits a very common intuition here, one that can be traced back to Descartes (if not earlier) and persists to this day, simply taking it for granted that we somehow literally observe the contents of our own minds. I will later argue that this view of introspection is not nearly as obvious as it seems if we attend closely to the phenomenology of first-person experi-ence, and that what we have here is not a literal understanding of intro-spection but rather the persistent application of perceptual metaphor to non-perceptual cognitive processes. For the time being, however, the point to notice is how obvious and basic it seems to James that introspec-tion is a kind of perception. To James, and to many others before and after him, it is simply given that we observe our own conscious mental states through a perceptual faculty called introspection.

Perceptual models of introspection in Psychology

The characterization of introspection as a kind of perception was put to practical use by a group of early psychologists that is now referred to, often in a somewhat pejorative manner, as the ‘introspectionists’ (Boring, 1953 ; Danziger, 1980 ; Lyons, 1986 ; Schwitzgebel, 2004 ). Most notably, this includes the work of Wilhelm Wundt and E. B. Titchener. Like James, these introspectionists took introspection, as the inward perception of mental states, to be a foundational method for psycho-logical research. Unlike James, however, they tried to narrow and sharpen the focus of introspection like a scientific instrument, so that it could operate more precisely than it does in everyday life. By training experimental subjects to isolate their introspective observations, the introspectionists sought to uncover the basic constituents of experi-ence, much as chemists empirically investigated the basic constitu-ents of physical substances in the construction of the periodic table. However, as those familiar with the history of psychology will know,


different introspection-based research projects came up with different, and even opposed, results. 1 Such conflict paved the way for other foun-dational psychologists to dismiss the use of introspection, and the very different behaviorist approach came to prominence as a foundation for the relatively newborn science of psychology. Even though behaviorism has also since gone by the wayside (for the most part), its rejection of ‘introspectionism’ as a psychological approach still has an impact on psychological research, with almost univocal prominence being given to third-person observation and analysis in mainstream psycholog-ical methodologies, at least as they are often officially advertised and described, tacit reliance on verbal reports notwithstanding.

Nevertheless, the characterization of introspection as a kind of perception has continued in psychology. For instance, the psychologist Nicholas Humphrey, in a book titled The Inner Eye , gives a central role to inner perception in his account of social intelligence (1986). To name another influential example, Ulric Neisser speaks of self-perception in his descriptions of ecological and interpersonal self-knowledge (1993), though he fails to distinguish between the importantly different notions that self-knowledge can be gained through a process of self-perception and that one can gain self-knowledge through ordinary sensory percep-tion (the latter being both quite plausible, and entirely compatible with a rejection of inner perception, as we will see). It is worth noting here that conceiving of introspection as inner perception and rejecting inner perception as a psychological method is not necessarily inconsistent. In particular, it is possible to hold that introspection is a kind of perception without thinking that such perception should be the basis for psycho-logical research. Perhaps, for instance, it is a particularly inaccurate or biased kind of perception and therefore cannot be a trustworthy source of information. In fact, such a view of introspection can be found in recent work in psychology, with use of the term ‘self-perception’ to char-acterize fallible and biased first-person perception of one’s own mental states (See Silvia and Gendolla, 2001 , for instance). This term often occurs with little or no conception or elaboration regarding what this ‘self-perception’ is supposed to be, in terms of actual, concrete proc-esses. Rather, it seems to be simply assumed, just as we saw above with William James and others, that people access their own psychological states in a perceptual manner. Despite the demise of introspectionism in psychology, the view that humans understand themselves through some kind of internal mental perception persists in contemporary psychology.


Inner perception in contemporary philosophy of mind

I will now turn to some prominent contemporary philosophers who explicitly articulate and defend the claim that introspection is a kind of perception. First up is David Armstrong and his influential characteri-zation of introspective consciousness as inner perception. In explaining his account of introspective consciousness, Armstrong utilizes the example of a long-distance truck driver. He states that ‘After driving for long periods of time, particularly at night, it is possible to “come to” and realize that for some time past one has been driving without being aware of what one has been doing’. (1997, p. 723). The idea is that such a person, prior to ‘coming to’, is lacking introspective consciousness. The driver is perceptually aware of the environment, at least insofar as such awareness is required for driving, but not self-reflectively aware of her awareness itself. For Armstrong, this lack of awareness is a lack of perceiving one’s own mental states. In illustration of this, he states

What is it that the long-distance truck driver lacks? I think it is an additional form of perception, or, a little more cautiously, it is some-thing that resembles perception. But unlike sense -perception, it is not directed toward our current environment and/or our current bodily state. It is perception of the mental. Such ‘inner’ perception is tradi-tionally called introspection, or introspective awareness. (1997, p. 724)

Like Locke, Armstrong distinguishes between ordinary sensory percep-tion and introspection while still regarding them as similar enough to classify introspection as a kind of ‘inner’ perception. Although he hints at some caution on this point, it is clear that Armstrong adopts a percep-tual account of introspection. In fact, he directly affirms the Kantian viewpoint we looked at earlier:

I believe that Kant suggested the correct way of thinking about intro-spection when he spoke of our awareness of our own mental states as the operation of ‘inner sense’. He took sense-perception as the model for introspection. By sense-perception we become aware of current phys-ical happenings in our environment and our body. By inner sense we become aware of current happenings in our own mind. (1968, p. 95)

So, like Kant and others, Armstrong thinks that we come to know our own mental states by perceiving them in a sensory or sensory-like manner, just as we come to know about external objects through our


various senses. However, unlike the other philosophers we have looked at so far, but like most philosophers today, Armstrong is a materialist and devotes much of his work to understanding the mind in terms of physical processes. Following this general orientation towards the mental, Armstrong suggests that introspection, as a physical process, occurs through a self-scanning process in the brain: ‘ ... it will be a process in which one part of the brain scans another part of the brain. In perception the brain scans the environment. In awareness of the perception another process in the brain scans that scanning’. (1968, p. 94). Through this materialist viewpoint on the inner perception model of introspection, Armstrong characterizes introspection as a contin-gent, fallible process manifested in the complex machinery of the brain. Just as our eyes can make mistakes in perceiving the world around us, through various physical ailments and other sources of error, so too can our brain scanner make mistakes when perceiving our own mental states. From Armstrong’s perspective, it seems that the only substan-tial conceptual difference between introspection and ordinary sense perception is in terms of what they are directed towards. One is directed inward, while the others are directed outward, but otherwise they are of the same sort of physical process with parallel epistemic status, involving the fallible perception of objects through specific mecha-nisms contingently manifested in the human mind.

The contemporary philosopher Paul Churchland holds a similar view of introspection, also connected to his account of consciousness in general (1984; 1985). He states that

... self-consciousness, on this view, is just a species of perception: self-perception . It is not perception of one’s foot with one’s eyes, for example, but is rather the perception of one’s internal states with what we may call (largely in ignorance) one’s faculty of introspec-tion. Self-consciousness is thus no more (and no less) mysterious than perception generally. It is just directed internally rather than externally. (1985, p. 74)

Like Armstrong, Churchland is a brain-centric materialist, and so at first pass it may not seem surprising that he identifies introspection as a perceptual faculty in the brain. However, Churchland is a materialist of a specific sort: an eliminative materialist. He seeks not only to explain, but to replace, the mental in terms of the physical processes imple-mented in the brain. In other words, rather than attempting to show how our ordinary psychological concepts (e.g., beliefs, experiences)


can be understood in terms of physical processes, Churchland suggests that we do away with our ordinary folk psychology altogether, ulti-mately replacing it with empirically-grounded neuroscience (1981). Considering this, I find it rather odd that he uncritically follows the common conception of introspection as a kind of perception, espe-cially when he acknowledges, as seen in the quote above, that we are generally ignorant of what particular neural processes constitute this perceptual faculty. As I will elaborate upon later, I think that this igno-rance is grounds for skepticism regarding the inner perception model. Yet, remarkably, Churchland does not even seem to consider this possi-bility, thereby falling prey to the folk psychological characterizations of the mental that he seeks to eliminate. Perhaps like James, he simply assumes it to be obvious that introspection is perceptual in nature. Whatever the case may be, however, the fact that an otherwise radical philosopher like Churchland uncritically accepts the inner perception model gives further testament to the commonality and prominence of this account.

Another noteworthy contemporary philosopher who explic-itly endorses the perceptual view of introspection is William Lycan. Following Locke, Kant, and Armstrong, Lycan argues that we have the ability to perceive internal states through some kind of monitoring process. Echoing Armstrong, he states that:

... to be actively-introspectively aware that P is for one to have an internal scanner in working order that is operating on some state that is itself psychological and delivering information about that state to one’s executive control unit. (1987, p. 72)

Lycan further claims that this introspective ability accounts for the nature of consciousness, stating that ‘consciousness is a perception-like second-order representing of our own psychological states and events’. (1996, p. 13). However, for our purposes, we can set that additional claim aside and focus on the prior idea that we have such a perception-like capacity in the first place. In this regard, Lycan is yet another advocate of the perceptual account of introspection.

So, as we can now see, the idea that introspection is a kind of percep-tion extends throughout the history of modern philosophy, into the origins of psychology, and up to present-day philosophy of mind. It is remarkably, even insidiously, common. Thus far, all of the examples we have considered came from thinkers in the Western cultural tradition. Yet, the idea that we can perceive our own mental states can be found


in non-Western systems of thought as well. For example, Buddhism contains the idea that we have a ‘sixth sense’ through which we inter-nally observe the workings of our minds (deCharms, 1998 , p. 66). This idea is often mentioned in a practical context, in describing medita-tion practices that involve observing mental events. For example, in explaining Tantric Yoga practice, it is stated that ‘The primary concern of these teachings is inward contemplation and introspection to directly perceive, in the atmosphere of meditative settledness, the functioning of the mind’. (Powers, 1995 , p. 244; see also Thera, 1962 ). The fact that the perceptual account of introspection can be found in this very different context provides yet further indication of its prominence. The idea of ‘looking within’ ourselves through the ‘mind’s eye’ is a remark-ably common, typical, and perhaps even near-universal way of under-standing the human capacity for introspection. Yet, as I will argue in the next chapter, it is fundamentally mistaken.

16

The Self-Existent made the senses turn outward. Accordingly, man looks towards what is without, and sees not what is within.

– The Katha Upanishad

How pathetically scanty my self- knowledge is compared with, say, my knowledge of my room ... . Why? There is no such thing as obser-vation of the inner world, as there is of the outer world.

– Franz Kafka

Although the understanding of introspection as a kind of inner percep-tion is common and deep-rooted in ordinary, philosophical, and scien-tific conceptions of the mind, there are good reasons to think that the inner perception model of introspection is flawed and ultimately unten-able. 1 In this chapter, I will explain some of those reasons, detailing physiological, phenomenological, functional, and evolutionary prob-lems that inner perception accounts face. More specifically, I will argue that there is a) no physiologically-identifiable faculty of introspection, b) no perceptual phenomenology of mental objects, with no distinctive qualitative character by which they appear, c) no perceptual account of the recursive functions that introspection performs, and d) no plausible evolutionary explanation of an inner perception function. Whichever of these angles one takes on the matter, the notion of inner perception does not hold up to close scrutiny. Collectively, these reasons constitute a strong case against the existence of anything like a mind’s eye, as a distinctive faculty through which we somehow literally perceive the contents of our own mind.

Considering the multiple problems that the inner perception model faces, some of which seem quite obvious once they are brought out into

2 Poking Out the Inner Eye

Poking Out the Inner Eye 17

the light of day, one might wonder how it ever could have been taken seriously in the first place. As we will see in the subsequent chapter, however, considering the idea of inner perception as a metaphor renders common usage of the concept more understandable. The concept of introspection as inner perception provides a unifying metaphor for approaching the first-person knowledge we have of our own minds, simplifying what turns out to be a rather diverse assortment of proc-esses under a single familiar framework, in keeping with a more general human tendency to understand complex and abstract phenomena through the metaphorical projection of tangible and bodily-grounded conceptual schemes. It is an understandable conception and the meta-phor works for some purposes, but it ultimately fails as a literal and explanatorily satisfying conception of what we actually undergo when we introspect, as the following considerations demonstrate.

Is there a mind’s eye in the brain?

Perhaps the most obvious point to be made against the perceptual account is that there is no identifiable sensory organ through which humans perceive the contents of their own minds. Perception, in all standard and uncontroversial cases, is connected with a particular sensory organ through which objects and events are perceived: through eyes we perceive visual phenomena, through ears we perceive audi-tory phenomena, and through skin we perceive tactile phenomena. 2 In all such cases, across the full spectrum of ordinary sensory percep-tion, there are specific physiologically-identifiable processes that enable our perceptual capacities. Accordingly, one ought to expect a similar sensory process in introspection if it is to be regarded literally as a kind of perception, as the ‘inner-sense’ terminology we saw earlier itself suggests. However, such is not the case. Through all the probing and manipulation of human brains that has occurred in the neurosciences, nothing remotely like a sensory organ that takes brain events and/or mental states themselves as input has been identified. So, it seems, there is simply no empirical physiological basis to support the idea of a real, tangible, inner perceptual faculty in human organisms.

For those who approach things from a naturalistic perspective, such that whatever we attribute to the mind must be embodied somehow in the physical world (in the brain or elsewhere), the lack of an identifiable inner sensory process may alone seem like a sufficient basis for rejecting a literal conception of introspection as a kind of perception. There are responses on behalf of the literalist account, however. For example,


David Armstrong, one of the philosophical inner perception theorists we looked at earlier, denies the need for a specific inner-sense organ. He believes that there is some sort of scanning function performed by the brain upon brain states, yet denies that this scanning function requires some specific organ to do the job. In support of this, he offers the following analogy:

It is sometimes argued that introspection cannot be compared to sense-perception because no sense-organ is involved. We say ‘I see with my eyes’, but there is nothing with which I can say that I discover I am thinking. Now I do not believe that this objection would carry much weight, even if the difference from sense-perception could be made out. But in any case there is one sort of sense-perception where we do not say that we perceive with anything: bodily perception. When I become aware that I am hot, or that my limbs are moving, and I do not gain this knowledge by touch, there is no organ that I can say that I perceive these things with . Yet any psychology text-book would take these to be cases of sense-perception. (1968, pp. 95–96)

Armstrong’s analogy with bodily perception here is fundamentally misguided. The problem is that, in contrast to purported inner percep-tion, we know that there are specific physiological processes that are involved in bodily perception, as any textbook that concerns itself with the topic will surely show. In fact, there are actually a number of different physiological processes whereby information about one’s own body is processed, such as heat and pain receptors, balance detection, detection of muscle position and movement, and so on. In all such cases (which are often lumped together under the general category of prop-rioception), there are specific sorts of nervous tissues that can be func-tionally and physiologically identified with the perceptual processes at issue. Now, these specific physiological processes may or may not be counted as sensory organs, depending on how the concept of a sensory organ is characterized, but nevertheless they are physical processes that can be precisely identified with the mode of perception at work, regard-less of whether or not they constitute sensory organs. We have nothing remotely like this at hand when it comes to introspection, as a special and unique kind of perception. There are no identifiable nervous tissues that serve the specific purpose of detecting mental states (whatever that may mean ... more on this in the next section). So, if there is no specific physiological process that we can associate with introspection, how can


we legitimately suppose that there is some actual perceptual process involved? Seen in this light, the perceptual model of introspection is simply an ad hoc hypothesis with no empirical founding. To be clear, I am not denying that introspection, whatever it may be, has a physi-ological basis. In fact, as a materialist, I think that all mental processes are identifiable with physical processes of some sort or other, whether they are in a brain or elsewhere. Rather, my point here is only that there is no physiological process that can be uniquely identified with intro-spection as a kind of inner perceptual faculty.

To elaborate on this, let us consider inner perception functionally, in terms of the perceptual function it is purported to perform, as Armstrong himself and many other philosophers are inclined to do with regard to the mind in general. Introspection, characterized as a perceptual process in functional terms, is a faculty that performs the function of observing mental states, in some manner or other (what exactly it might observe is a matter we will turn to shortly, but it need not be specified here). Now, in order for a function to actually occur in the natural world, there must be some embodied process through which it occurs. In other words, if a particular function is performed there is, by necessity, some particular concrete physical event that constitutes its performance. Yet, this seems to be what Armstrong is denying in the case of inner percep-tion. He says that introspection occurs through some sort of brain scan-ning process, but then denies that there is any actual brain scanner in the brain. This is a fundamental inconsistency in Armstrong’s account, contrary to his avowed materialism and functionalism. His function-alism, as it turns out, isn’t materialist enough. If introspection is to be identified with some perceptual process in the brain, then some such perceptual process and its associated physiological mechanism ought to be identified. Since no such brain-observing process can be identified, the perceptual account is fundamentally ungrounded in the concrete natural world we all live in.

Another response here is to suggest that inner perception occurs through a brain process, but one that we cannot currently identify. Perhaps inner perception is a distributed higher-order neural network that is not precisely localizable in neurophysiological terms. Arguing along these lines, in response to the criticism that there is no identifi-able inner perception mechanism in the brain, B. Alan Wallace states:

This point should come as no surprise, however, in light of the fact that most of the higher order mental functions are little under-stood by contemporary neurophysiology and there is no cogent


neuroscientific understanding of the production of conscious-ness. ... within the brain sciences there is presently a general consensus that, while there is evidence for some sort of localization, there are no precise and different locations or even different sorts of brain tissue corresponding to different thoughts, sensations, or even different types of thoughts or sensations. Thus, the fact that no organs of introspection have been discovered can hardly be counted as grounds for rejecting its existence. (Wallace, 2000 , p. 87)

Although Wallace acknowledges that no organ of introspection has been identified, he thinks that we can still regard introspection as a kind of perceptual faculty, presuming that we just have not yet been able to identify the physical basis for it in the brain. As he sees it, the complex and distributed nature of higher-order brain states precludes the identification of precise introspective mechanisms. This alone, however, is not a good reason for thinking that introspection is a kind of perception. It is merely an appeal to ignorance, clearly insufficient as positive support. Just as a lack of information regarding extraterrestrial beings does nothing to support their existence, a lack of information regarding an inner-sense mechanism does nothing to support the idea that there is an introspective perceptual faculty. So, Wallace’s response should carry little weight as a positive basis for asserting the existence of an internal perceptual faculty.

Now, some might reply that the appeal to ignorance at play here equally applies to the denial of an inner perceptual faculty. The fact that we have not found an introspective perceptual faculty in the brain does not necessarily mean that there is not one. Fair enough. Perhaps the scientific exploration of the brain has simply not developed far enough. Perhaps there is some distributed neural network that oper-ates as an inner perceptual faculty; we just have not yet found it. Of course, future research always holds the possibility of uncovering some previously unknown or unrecognized object or process, but holding out for some future discovery to justify a positive assertion now is a risky and groundless enterprise. If we are going to give explanatory weight to introspection as a genuine perceptual faculty, then we need, at a minimum, some positive understanding of what that faculty is and how it works. We simply do not have any such positive understanding of a faculty in the brain that serves the function of perceiving mental states.

Admittedly, however, just noting the lack of an identifiable organ or physiological process is not, by itself, a convincing basis to dismiss


the perceptual account of introspection. It provides some prima facie ground for doubt, but more is needed to conclusively reject the account. In the following sections, I provide what I regard as the further justifica-tion needed for conclusive dismissal. Taken as a whole, these considera-tions make a strong case for rejecting the inner perception model of introspection, clearing the ground for a more nuanced understanding of introspection that is both philosophically satisfying and in accord with the available empirical data.

What are the objects of mental perception?

Turning from physiology to phenomenology, let us now consider intro-spective experience from the first-person standpoint of a conscious subject. As has been discussed in much detail by phenomenologists such as Husserl, Merleau-Ponty, and others, perception can be characterized in terms of how it is experienced from the perceiving subject’s point of view. In perception ‘the world shows up for us’, (2009, p. 4) as the contemporary philosopher Alva No ë is fond of putting it. Considering the saliency of phenomenology in understanding the nature of percep-tion, we can ask whether a distinctively perceptual phenomenology occurs in introspection as well, as the idea of inner perception would seem to suggest. Is there a way in which the ‘inner world’ shows up for us, analogous to ordinary perception of the outer world? If we confront this question pre-theoretically, bracketing the background conceptualization of introspection as a mode of inner perception, it seems clear to me that the answer is no. There is no distinctive percep-tual phenomenology identifiable with introspection, analogous to the phenomenology of external perception. To illustrate this, I will consider two basic aspects of ordinary perceptual experience of the world: the appearance of objects and their phenomenal properties. The existence of perceived mental objects will occupy us in this section and we will turn to the question of introspective phenomenal properties in the following section. In both cases, however, the underlying point is the same: There is no distinctive phenomenology to be identified with inner perception.

Patently, one key aspect of perception is the appearance of objects in perception. Our eyes see swaying trees and running dogs, our ears hear human voices and bird chirps, our noses smell the exhaust of a passing car, and so on. 3 Considering this, we can pose the following question: If there is a literal perceptual faculty in the mind, such as a ‘mind’s eye’, then what sorts of objects does it perceive? Presumably, the answer


must be brain states and/or mental states in some form or other, so let us consider each of these possibilities, seeking to determine what, if anything, we might perceive through introspection.

If the supposed internal perceptual faculty perceives brain states, then these brain states must be occluded or ‘scrambled’ in some way or other, as they do not appear to us in introspection as brain states (Lyons, 1986 , pp. 57ff). Brain states are incredibly complicated electro-chemical events among virtually innumerable neural networks encased inside one’s skull. However, this is definitely not what we perceive, if we perceive anything at all, through introspection. The thought ‘I am thinking’, for instance, does not appear in experience as a particular neural event, or even as any discernible physical thing at all, as Descartes noted and made (too) much of several centuries ago. So, if we perceive a brain state when we are aware of having such a thought, then that brain state must be filtered through some sort of process that transforms it into something that appears quite different, to such an extent that it is unrecognizable as such. Otherwise, philosophers would not have had to spill ink over the mind/body problem all these years, and Descartes himself could have readily identified mental states with brain states. So, if we have the capacity to perceive brain states, it must be through some mechanism that alters their appearance so radically that they do not seem like brain processes at all.

The idea of a perceptual brain scrambler might fly in a Philip K. Dick novel, but not as a literal account of introspection. 4 It is highly improb-able that there would be some such scrambling mechanism in the brain, serving the specific function of altering perceptions beyond any recog-nition as a physical event. Outside the construction of purposefully skeptical scenarios, such as Descartes’ evil demon hypothesis, there is no reason to suppose that nature, or God, or whatever, would impose such a device upon human cognition. So, this brain scrambling idea is nothing more than an ad hoc hypothesis, having no merit or evidence beyond filling in the gaps of an implausible theory.

To further support my point here, consider the other side of the coin. Not only is it implausible to think that we perceive brain states, it is equally implausible to hold that perceiving brain states counts as intro-spection. Suppose, for instance, that you cut open my head or place me in an fMRI in such a way that I can directly observe my own brain states. It seems absurd to say that I would be introspecting in this case. Even if I had a complete understanding of the brain, being able to accurately correlate specific neural events with specific subjective experiences or cognitive states, I would not be experiencing or understanding my


mentality in the same first-personal way that I do through introspec-tion. I would be observing my mind in the third-person, in the same manner that anyone else could. No matter what kind of perception we imagine here, whether it be looking directly at a brain, observing an fMRI or other mechanical monitoring process, or even listening to auditory representations of neural firings, it does not seem remotely like introspection in any way. If no other way of perceiving brain states seems at all similar to introspection, we are left with little reason for thinking that introspection itself is the perception of brain states.

One reply to this line of criticism might be to point out the possibility that we do not perceive brain states, but rather only secondary repre-sentations of brain states. Arguably, when we perceive ordinary external objects, such as tables or trees, we do not perceive their intrinsic phys-ical states (i.e., their atomic structure), but rather various representa-tional features of those states, such as colors and shapes. Perhaps these representations need not be understood as ‘scrambled’, illusory, or deceptive. Instead, they can be understood as simply the way physical events appear through specific sensory modalities. So, analogously, perhaps we perceive our brain states through representational features of those states. But what would these representations be, if this were to be our mode of access to the contents of our own minds? Presumably, they would be mental states, as we commonly understand them: pains, thoughts, beliefs, desires, and hopes. So, while we can rule out the direct perception of brain states for the reasons given above, perhaps there is room left to say that we perceive brain states as mental states, in this secondary, representational manner. Let us turn, then, to consider whether we perceive mental states. 5

While there are numerous kinds of mental states, let us narrow our scope a bit and focus on two prototypical examples of mental states: emotions and beliefs. If we can perceive mental states, then surely these two staples of mental life would be included among the kinds of things one can perceive. So, can these be counted as objects of perception, in the same way that trees and cars are objects that we perceive through our senses?

It does not seem to me that they can. First of all, beliefs and emotions do not exhibit the distinctive perceptual unity possessed by externally perceived objects, such that they continue to appear as one and the same object through change over time, a trait sometimes referred to as ‘perceptual constancy’. For instance, when watching a dog run back and forth from you while playing fetch, the dog continues to appear as a singular constant object throughout the extended event, distinct


from other objects in your perceptual array and despite changes in both its appearance and your continuously shifting perceptual attention. Things are not like this in introspection. It is notoriously difficult to identify what exactly a belief is and precisely where it is in our experi-ence, such that it is a unified perceived object across one’s experience. Arguably, the same could be said of emotions. What are emotions, as objects of experience? Where are they located? What are their boundary conditions, such that we can distinguish an emotion from other aspects of one’s experience? Answers to such questions are not forthcoming in our experience, but if emotions had perceptual properties then it would seem that we should be able to readily describe emotions in terms of characteristics that could answer these questions in a clear common-sense manner, just as we can readily describe cars, trees, dogs, and other things in terms of their perceivable qualities (e.g., shapes, colors, textures, kinds of movement). Considering these examples, we can see that mental states lack the easily identifiable perceptual proper-ties possessed by uncontroversial objects of perception. Mental states do not persist through time as single, isolatable objects of perception, as do pencils, seashells, and other ordinary objects. It is difficult, if not impossible, to say when and where they begin or end, and to say how they are separable from, and/or related to, one another on the basis of their observable characteristics.

To bring this home with a more concrete example, consider your belief that you are reading right now. You probably were not explicitly aware of this belief, but now that I have brought it to your attention, surely you believe that you are reading at this moment. But now ask yourself, what is this belief? Where is it? Where does it begin? Where does it end? If you turn your attention to something else for a moment, what happens to the belief that you are reading? Does it disappear, or is it still present in some way or other? If you turn your attention back to the fact that you are reading, is it the same belief that was present before? Or is it a similar but identifiably different belief? Is there any clear way to answer these questions through identifiable perceptual characteristics? Do these questions even make sense? I hope that my point is clear. Beliefs, whatever they may be, are amorphous, protean, and often experientially intangible. They are not the sort of thing that can be a perceptual object, if they can be counted as objects at all.

Many have argued that beliefs must be understood relationally, in terms of the functions or conceptual roles they perform through inter-acting with other mental states and/or the things which they are about, rather than as objects that can be defined and understood in terms


of inherent properties. I tend to agree with this perspective, though it would take us off target for me to explain why here. 6 If this is the right way to think of beliefs, however, or if it is even a plausible option, then beliefs could not be perceptual objects, by their very nature. They are intentional relations between beings and the objects with which they interact, rather than intrinsically observable objects in themselves. Consider again some belief that you have about what you are reading. From this perspective, that belief is the belief that it is in virtue of how it functions in your cognitive economy, e.g., the role it plays in your understanding of what you are reading, or its role in your ability to relate what you are reading to other information you have. Beliefs, then, are not objects that can be perceived. They are postulated, inferred, assumed, or hypothesized, based on the informational processes in which they function.

To take these worries about perceptual objects a bit further, let me outline a further possible perspective here. Perhaps mental states, such as beliefs and emotions, should not be thought of as objects, but rather as subjective events that enable the possibility of having perceptual objects. Following the phenomenological tradition of Husserl, many have suggested that we think of the self as a subject rather than an object. (For a recent illustration of this idea, see Zahavi, 2002 .) From this perspective, the focus is on the self as subject, as the possibility for experience, rather than as an object of experience itself. Along these lines, perhaps mental states ought to be thought of as events that we are constituted by, as cognitive and experiential subjects, rather than as things that we experience as objects. In other words, perhaps we do not know about mental states as objects in the way that we do the objects of the so-called external world. Instead, we know them by being them.

Though he is immersed in the analytic philosophical tradition, rather than the phenomenological tradition following Husserl, John Searle makes a similar point, arguing against the idea of introspection as a kind of inner observation on the grounds that conscious mental states are inherently subjective. He states that:

... where conscious subjectivity is concerned, there is no distinc-tion between the observation and the thing observed, between the perception and the object perceived. The model of vision works on the presupposition that there is a distinction between the thing seen and the seeing of it. But for ‘introspection’ there is simply no way to make this separation. Any introspection I have of my own conscious state is itself that conscious state. (1992, p. 97)


I think that there is something to Searle’s point here, but it goes a bit too far. There is a significant incongruity between ordinary objects and mental states. The former can be objects of perception while the latter cannot. We can perceptually observe trees, birds, chairs, and so on, but when it comes to beliefs, desires, personality traits, and such, we are at pains to say in what way these things could even count as percep-tual objects. Contra Searle, however, this is not to say that a partic-ular mental state cannot be understood separately from the subjective experience of the person who is having it. As I will explain in more detail later, mental states can become objects by becoming represented through conceptual structures. In addition to having, or being consti-tuted by, mental states, we learn about them by conceptualizing and thinking about them, just as we do with other, extramental things. In effect, this turns mental states into objects of a sort, just as analyzing physical movements leads to positing forces such as gravity as objects of inquiry. But, importantly, these are not objects that we perceive. Rather, they are conceptual abstractions, allowing us to make sense of certain phenomena through an interpretational framework we project onto them. To use an analogy between physical events and theoret-ical concepts in physics, perhaps beliefs, emotions, and other mental states are more akin to gravity than to observable falling objects. They provide the background conditions for the occurrence of experience itself, rather than being objects of experience themselves.

Perhaps an example will help: Suppose you pause a moment to again reflect upon what you are reading, this time turning your attention to your thoughts about the content of what you have been reading. In doing so, you might quite naturally think in terms of beliefs, such as ‘I believe that this argument is mistaken’ or ‘I can’t believe that this guy believes that!’ Now, in having such thoughts, do you perceive anything that constitutes the thoughts themselves? In other words, do you know what you believe by somehow observing beliefs in your experience? Surely, the beliefs are objects of your attention in some way, insofar as you are consciously thinking about them, but from consideration of these examples I think it is clear that they are not objects in virtue of being perceived. We simply do not perceive beliefs. Instead, we use the concept of belief to interpret and understand thought. In doing so, we utilize beliefs as objects, not in the sense of things that we perceive, but rather in the sense of things that we posit to make sense of the world. So, while I reject the idea that mental states are perceived as objects through some introspective faculty, mental states can be, and are, treated as objects through our various conceptual capacities. However, to be


clear, these are not objects that we know about in a perceptual manner. They are objects that we deploy in a theoretical manner to understand ourselves and the other thinking agents in the world around us, just as scientists deploy theoretical entities like forces and subatomic particles to understand the phenomena they investigate.

In this section, I have argued that we perceive neither brain states nor mental states, as objects of perception in some type of inner observa-tional faculty. In summation, I want to point to a similar concern posed by G ü ven G ü zeldere, in criticizing higher-order perception accounts of consciousness. He states:

Locke, Armstrong, Churchland, and Lycan all talk about conscious-ness as the awareness, or perception, or monitoring, or scanning of one’s mental states. There is an ambiguity inherent in all these state-ments concerning the nature of the proper object of such internal, perception-like awareness. What exactly is being perceived in the ‘perception of what passes in one’s own mind’: the content of the mental state that happens to be ‘passing through’ one’s mind at the time or the mental state itself? Or the content of another thought to the effect that one is having such a mental state? There is surpris-ingly little attention paid to spelling out the answer to this question in any detail. (1997, p. 792)

Although G ü zeldere is concerned here with the explanation of consciousness, we can set that issue aside and focus on the way the accounts he is discussing treat introspection as a kind of mental percep-tion. In this respect, the point to take home is that such accounts lack any clear and unambiguous explanation of what exactly is involved in such perception. As G ü zeldere states, none of these accounts cash out precisely how this perception takes place or what it involves. Rather, it is simply assumed that mental states can be objects of perception. Presumably, the authors of these accounts think that it will be obvious to their readers what is perceived and how the perception takes place. I have shown in this section that this presumption cannot be made. There is a reason that perceptual accounts of introspection have little to say about what is actually perceived, and it is not because what is actually perceived is obvious. Rather, it is because there is no clear and understandable way to construe mental states as percep-tual objects. We simply do not perceive beliefs, pains, desires, fears, personality traits, or any other mental state as an object appearing in our experience.


Do mental states exhibit perceptual qualia?

Another feature that perceptual experiences have in common is that they have distinctive phenomenal properties. They all ‘appear’ in their own idiosyncratic way, each having a unique manner in which there is ‘something that it is like’ to undergo such a perceptual experience. Philosophers often use the term ‘qualia’ to refer to these phenomenal properties. Think again of visual objects and how they appear in experi-ence, in terms of color, shape, and movement, or the way sounds produce their particular kinds of experiences. These instances of perception each have a distinctive phenomenal feel that is intrinsic to that particular type of perceptual experience. So, again we can ask how introspection, construed as a kind of perception, holds up in regard to this aspect of perception. Do mental states have comparable phenomenal properties, or ways of seeming? Do they have perceptual qualia? 7

On the positive side, emotions, and perhaps other kinds of mental states, might be said to have distinctive sorts of feels that are intrinsic to their nature. Consider experiences of anger or love, for example. There is a unique experiential character to these emotions. But do such experi-ences exhibit the same type of phenomenal feels that constitute percep-tual qualia? Arguably, they do not. Although they do have phenomenal qualities, emotions are not manifested in experience in the same vivid, unique way that perceptual phenomena are. There is no single way of seeming that is characteristic across the emotions, in the same way that, say, the experience of color exists across the spectrum of possibilities in visual perception. This lack of distinctive perceptual qualia is likely due to the fact that emotions are not perceptual objects at all, as illustrated in the previous section. Emotions, insofar as they may pertain to percep-tion at all, are modes of perceiving, rather than perceivable objects in themselves. Fear, for example, is not something we perceive in ourselves, but rather something that we feel in association with perceiving or thinking about something else that induces the feeling in some way or other. Consider love as another example. When we love someone, the experience of that love is manifested through our perception and thoughts of the person, and not as a distinct phenomenal property that we perceive in itself. In other words, we know of the love through our experiences with the person, not through perceptual experience of the emotion itself. Considering examples such as these, we can separate the phenomenology of emotions (and arguably other mental states) from perceptual phenomenology, acknowledging the apparent lack of percep-tual qualia in our experience of our mental lives.


The case against introspective perceptual qualia is even stronger when we consider the case of beliefs. Arguably, beliefs have no way of feeling at all. Perhaps there is some experiential sense of confidence associ-ated with having a belief, but a belief is not itself identifiable as a belief through some phenomenal way of seeming. Even David Chalmers, an adamant champion of conscious phenomenal states, is skeptical of the idea that beliefs have some distinctive sort of qualia or phenomenal appearance through which they may be said to be beliefs. Here is what he says about it:

Certainly, there is often conscious experience in the vicinity of belief: there is something it is like when one has an occurent (i.e., conscious) belief, and most nonoccurent beliefs can at least bring about a conscious belief. The crucial questions, though, are whether this conscious quality is what makes the state a belief, and whether it is what gives it the content it has. This may be more plausible for some beliefs than for others: for example, one might argue that a conscious quality is required to truly have beliefs about one’s expe-riences , and perhaps also certain sorts of experiences are required to have certain sorts of perceptual beliefs about the external world (perhaps one needs red experiences to believe that an object is red?). In other cases, this seems more problematic. For example, when I think that Don Bradman is the greatest cricketer of all time, it seems plausible to say that I would have had the same belief even if I had had a very different conscious experience associated with it. The phenomenology of the belief is relatively faint, and it is hard to see how it could be this phenomenal quality that makes the belief a belief about Bradman. What seems more central to the belief’s content is the connection between the belief and Bradman, and the role it plays in my cognitive system. (1996, p. 20)

So, while there might be qualitative, phenomenal experiences associated with beliefs, beliefs are not themselves constituted by or identifiable through some phenomenal way of seeming. Although Chalmers is not concerned with the nature of introspection in making this point (his concern is to distinguish between psychological and phenomenological processes and to place beliefs in the former category), it bears an important implication for our topic: We do not identify beliefs by the way they phenomenally appear to us, but rather by the roles they play in our cognition.

Later, I will elaborate on this point in considerable detail and offer an account of how we come to have knowledge of our own beliefs in a


conceptual, rather than perceptual, manner. So, readers who are skep-tical of this brief account of belief should withhold final assessment until my full account of introspective knowledge of belief is in place. For now, however, I will simply appeal to my own experience and count on the reader having the same sorts of experiences (or, to be more accu-rate, the lack thereof) that I do. In my experience, beliefs simply do not have a distinctive phenomenal feel, in virtue of which they are identi-fied as beliefs. Of course, to be fair, I could be missing something here. Perhaps I have had my inner eye plucked out, and I am like a blind person denying the vivid nature of colors, but it sure seems to me that beliefs do not have the same sorts of phenomenal qualities that char-acterize perceptual experience. They do not present themselves in the same vivid, qualitatively identifiable way that perceptual objects do. There are only our ordinary perceptual experiences of external objects and our conceptual reflections upon them. There are no perceptions of these mental processes as phenomenal objects in themselves. As the point is commonly put in analytic philosophy of mind, (using visual metaphors, it is worth noting) they are ‘transparent’ or ‘diaphanous’ (Moore, 1903 ; Harman, 1990 ).

As a final point on the topic of perceptual qualia in introspection, consider the fact that even if there are distinctive qualitative experi-ences intrinsically tied to particular mental states and our ability to iden-tify them as the type of mental states they are, there are no perceptual qualia that are uniquely characteristic of mental experiences in general. Consider vision (or any other uncontroversial perceptual modality) in contrast. There are unique phenomenological properties intrinsic to our experience of vision manifested across all visual experiences. Visual experiences, in normal cases of human vision, involve colors and shapes of various kinds, for example. If there is indeed a perceptual faculty through which we observe our own mental lives, then it would seem that it, too, should have a distinctive phenomenology with types of percep-tual qualia that are uniquely tied to it as a distinct perceptual modality. Yet, this is not what we find in our experience of our own minds. Even within the category of emotion alone, as described above, there is no single set of perceptual qualia that can be uniquely associated with it across the board. Insofar as there are experiential qualia associated with emotions, they are each uniquely associated with individual emotions (the experience of anger, of love, of fear, and others) and have nothing identifiably common across the range of emotional possibilities. This is even more manifest when we consider all types of mental states together: beliefs, emotions, speculations, imaginings, personality traits, aspirations,


intentions, dispositions, and others. There simply are no uniquely char-acterizable qualia that occur across all of these kinds of phenomena, on a par with colors, shapes, sounds, tastes, smells, and so on, as we would expect from a single perceptual modality. On the basis of these phenom-enological considerations, we can again conclude that there is no internal perceptual faculty through which we observe the contents of our minds. When we ask for concrete details concerning the particular perceptual qualia uniquely associated with the supposed perceptual modality, we come up empty-handed. The perceptual qualia simply do not exist.

Perception, evolution, and knowledge of minds

Another plausible angle to explore is to understand perception from the perspective of evolutionary psychology. If we think of the human organism as a product of natural selection, we can understand the various specific processes of the organism in terms of the evolutionary adaptations that have brought them about. More specifically, percep-tion can be understood in terms of the functional benefits it confers upon organisms in their pursuit of survival and reproduction (Gaulin & McBurney, 2001 , pp. 91–109; Palmer & Palmer, 2002 , pp. 60–65). For example, from an evolutionary standpoint, we can understand vision as a unified functional process, in terms of what it does for organisms that have such processes. 8 Broadly speaking, it allows organisms to detect certain salient features of objects in the environment, such as the color and shape of a specific piece of fruit, the movements of a potential predator, or the face of a particular conspecific. In performing such functions, the eye works with specific cognitive processes as a functional unit that enables an organism to obtain information about its environment. This functional unit can be plausibly regarded as an evolutionary adaptation. It is a complex collection of tightly knit proc-esses that all work together to promote the survival and reproductive success of the organisms that have it. In the case of perception, broadly speaking, the adaptive advantage arises from the ability to process salient information from the surrounding environment. Setting the details aside, various types of perception may be characterized in terms of how they function to provide information to an organism about the state of the world it lives in. Of course, much more could be said to fully illustrate and defend this view of perception (as well as the more general presumptions regarding adaptations and functions), but this brief over-view presents a plausible and useful way of thinking about the nature of perception in general, in terms of adaptive functional processes.


This leads to yet another basis for rejecting the idea that introspection is a kind of perception: There appears to be no identifiable functional/adaptive process that serves the purpose of perceiving one’s own mental states. This is related to my initial objection regarding the absence of an introspection organ, but shifts the focus away from the organ toward the function it supposedly performs. In order to get a handle on this objection, let us first look at an account that suggests the opposite. Nicholas Humphrey offers the following hypothesis:

Now imagine that at some time in history a new kind of sense organ evolves, the inner eye whose field of view is not the outside world but the brain itself. Like other sense organs, the inner eye provides a picture of its information field that is partial and selective; but equally like other sense organs it has been designed by evolution so that its picture is a useful one, a ‘user-friendly’ description which tells the subject just so much as he requires to know in a form that he is predisposed to understand – allowing him by a kind of magical translation to see his own brain-states as conscious states of mind. ... Suppose that is indeed what consciousness amounts to, and that we human beings are the only animals in nature to have evolved this kind of inner eye. What would it mean for our ability to do psychology? To begin with, it would mean that each individual human being would have almost literally a headstart in reading his own mind. No more fussing about like a behaviourist psychologist with ‘intelligent guesses’ about what lies behind our own behaviour. We would know immediately where the deeper explanation lies – in our own brain-states, which our inner eye reveals. But in prac-tice it would mean very much more: for the explanation we have of our own behaviour could then form the basis for explaining other people’s , too. We could, in effect, imagine what it’s like to be them, because we know what it’s like to be ourselves. (1986, pp. 70–71)

Lest you think that Humphrey is just drawing a metaphorical picture here, he goes on to make this inner eye hypothesis the concrete basis for understanding social intelligence. In essence, Humphrey argues that the ‘inner eye’ is an evolutionary adaptation, hypothesizing that it would confer a significant fitness benefit on those who have one. In contrast, I suggest a different account. I think it is more plausible to hypothesize that our introspective capacities are not adaptively benefi-cial in themselves. Rather, they are by-products (i.e., spandrels) of other adaptive processes that make them possible.


To illustrate this, let us compare the identifiable benefits of knowing ourselves versus the benefits of knowing others. In terms of evolu-tionary fitness, it is not particularly beneficial for me to have accurate representations of my own mental states. Knowing that I am a coop-erative person, for instance, would not add anything beneficial to my interpersonal interactions. Any benefit would already be conferred by my actual cooperativeness as I engage with others in the world, regard-less of whether I accurately represent that feature of myself to myself.

Similar reasoning could apply to other types of mental states, such as beliefs, desires, and pains. Being introspectively aware that I believe there is some edible fruit in front of me, for example, would not seem to add any fitness benefit to me, as the belief itself would already confer any benefit that it may have. Considering this, it seems that there is no real evolutionary pressure for developing some inner observational process of one’s own mind through natural selection. Being able to understand the mental states of others, however, is clearly beneficial. Correctly believing that another person is cooperative, for instance, would quite obviously confer significant benefits to me. It would enable me to engage in mutually beneficial interactions with that other person (or perhaps to take advantage of the person, if I were self-interestedly inclined to do so). 9 Similarly, the ability to accurately represent the beliefs and desires of other people would be very fitness-enhancing in that it would enable the anticipation of various significant behaviors (e.g., potentially negative and positive social interactions). So, if there is an identifiable adaptive benefit here concerning knowledge of minds, it is in regard to our understanding of others, and not ourselves. In other words, the evolutionary pressures for perceptual and cognitive adapta-tions are geared toward an ability to represent and think about things in one’s external environment. There is no impetus toward internal perception, and hence no identifiable adaptive causal basis to support Humphrey’s ‘inner eye’ hypothesis.

This perspective overlaps with the thinking of a number of psycholo-gists and philosophers who have postulated the presence of an adaptive ‘theory of mind’ module/capacity in humans and some non-human primates. We will return to this topic in more detail later, but for now we can regard the ‘theory of mind’ as a cognitive module adapted to facilitate social understanding and interaction (Premack & Woodruff, 1978 ; Byrne & Whiten, 1988 ; Carruthers & Smith, 1996 ; Ristau, 1998 ). Roughly, the idea is that selective pressures for the ability to anticipate the actions of others in one’s social environment have led to a specific cognitive capacity to attribute beliefs and desires to others. Although


this idea is in need of further clarification and development. (For a critique, see Heyes, 1998 .) the evolutionary angle it develops is quite plausible and has a tremendous amount of explanatory power. At the very least, it seems safe to say that evolutionary pressures helped to form our abilities to detect and interpret external events, including the behaviors and/or mental states of others. But the idea that there is an adaptive advantage that might lead to the selection for an ability to directly perceive our own internal states seems much more doubtful. Considering this, introspection probably draws from the adaptive ability to understand others, rather than being an adaptive ability in its own right.

This idea is further corroborated by evidence in developmental psychology. For instance, it appears that children come to have an understanding of their own false beliefs at the same time that they understand false belief states in others – at around the age of three and a half years. As will be discussed in more detail later, this suggests that understanding false beliefs in both oneself and in others has the same cognitive origins (Gopnik, 1993 ; Gopnik & Meltzoff, 1997 ). If Humphrey and others who postulate an inner eye are correct, it would be rather surprising that this is the case. Instead, arguably, we ought to see the development of understanding one’s own beliefs before the development of understanding the beliefs of others, because it is presumably that sort of inferential ability that would have prompted the evolutionary selection of an internal observation mechanism. So, on the ‘inner eye’ model, especially as Humphrey describes it, it seems that children would ‘see’ the nature of beliefs within themselves and then later use that understanding to infer the existence of similar states in others. Since this is not so, at least in the case of children’s under-standing of false beliefs, the evidence weighs negatively against the inner eye hypothesis.

Another relevant consideration here is that introspection, or at least certain prevalent forms of it, is relatively rare in comparison with outward-directed perception and thought. Arguably, most human attention is directed toward the external world: navigation, object identification and manipulation, social interaction, and so on. People just don’t spend a lot of time focusing on their own mental states in general. Certainly, it does happen (or there would be no phenomenon to explain in this book!), but it is not nearly as prominent as the vast diversity of external concerns in human life. There is, in fact, some empirical support for this. Mihaly Csikszentmihalyi has done a signifi-cant amount of research regarding how people spend their time affects


how they feel, and one thing he has found in this regard is that people generally don’t like to be left alone, with only their own thoughts to occupy their attention (1997, pp. 89ff). To this effect, he states that:

... people in general report much lower moods when alone than when they are with others. They feel less happy, less cheerful, less strong, and more bored, more passive, more lonely. The only dimension of experience that tends to be higher alone is concentration. When first hearing these patterns, many thoughtful persons are incredulous: ‘This cannot be true’, they say, ‘ I love to be alone and seek out soli-tude when I can’. In fact it is possible to learn to like solitude, but it does not come easily. If one is an artist, scientist, or writer; or if one has a hobby, or a rich inner life, then being alone is not only enjoy-able but necessary. Relatively few individuals, however, master the mental tools that will make this possible. (Ibid., p. 90)

Evolutionarily, this makes sense. Survival and reproduction largely depend upon how an organism relates to the world, including the social world of other organisms like itself. So, it seems that we ought to be naturally focused on external things, such as food, predators, prospec-tive mates, and social competitors and cooperators. Anticipation, recog-nition, and prediction of these things are what matter to survival and reproduction, and hence are what drive evolutionary adaptation. It comes as no surprise, then, that our cognitive resources and disposi-tional orientations are geared toward these external things, and not toward our own inner nature and existence.

As further support for this perspective, consider the fact that inward-focused religious traditions and practices, such as Buddhist meditation, have to impose strong disciplinary measures upon atten-tion in order to cultivate inner-directed practices. Drawing our atten-tion to our own mental states in solitude takes a lot of effort, with constant struggle against inclinations to attend to the external world. Taking this into consideration, coupled with the ideas presented above, it seems that introspection is not our forte. It is not something that we do easily or automatically, as is the case with all of our other forms of perception, or adaptive processes more generally. So, it is implausible to think that we have some special part within us that is adapted and dedicated to the task.

In summary, there are no good reasons to suppose that there is a perceptual faculty specially adapted to perceive one’s own mental states. Instead, it is more plausible to think that our ability to understand


ourselves is a by-product of our adaptive, externally directed, abilities to understand our environment, including other people. Admittedly, these evolutionary considerations are somewhat speculative. In constructing adaptive explanations, especially rather loose and abstract ones such as these, there is the danger of constructing ‘just so’ stories that do not reflect actual developmental history, but rather only our imagina-tive capacities. Still, I think that these considerations give added plausi-bility to my rejection of inner perception. The inner perception model of introspection faces serious difficulties with regard to any supposed evolutionary origins it might have and, relatedly, has some problems with accommodating some basic observations regarding our tendency to attend to the external world rather than our own minds. Conjoining these concerns with the previous arguments I have given, the case against the perceptual account of introspection is quite strong.

Perception and recursion

One final reason for rejecting the perceptual account of introspec-tion pertains to the fact that introspection is recursive, in that we can introspect our introspections. Thus far, we have focused on aspects of perception that introspection lacks, in terms of physiology, phenom-enology, and function, but it is also worth considering properties of introspection that are not identifiable in perception. A detailed posi-tive account of introspection is forthcoming in the later chapters of this book, but for now we can note one particularly salient feature of introspection that is unavailable to perception: recursion. Simply put, introspective recursion is the application of introspection to itself. For instance, I am currently reflecting on a past event I seem to remember: As a child, I once saw what I thought was a rabid dog running through my neighborhood, while waiting for my school bus to arrive. In intro-spectively attending to this memory, I not only can recall the memory but also reflect upon the recall process itself. I can wonder whether this memory is veridical or not, for instance. Did it actually happen, or is it a falsely reconstructed memory? I can further think about this wondering, reflecting upon why I care whether my memory is veridical or not. Kicking it up yet another level, I can write about it, as I am doing now. Perhaps these iterative reflections can only plausibly go up a few levels, so to speak, due to the limits of our cognitive abilities, but it is sufficient to show that introspection is recursive. We can introspec-tively attend to not only our first-order mental states but also our intro-spective states themselves. In contrast, no perceptual faculty can do


this. We cannot hear or otherwise perceive the hearing of the sounds that we hear; we only hear the sounds. Our eyes cannot see their own visual perceptions; there are just the perceptions that they, in conjunction with the brain and surrounding environment, produce. Contrary to a popular phrase, we cannot feel our pain. There is just the pain itself, as a feeling. Of course, we can think about, and reflect upon, the pain, or any other perceptual state, as will be discussed in much detail later, but that is not a case of recursive perception. Rather, it is applying some other cognitive process to the pain, conceptual rather than perceptual in nature. So, to succinctly state and summarize the argument here, perceptions cannot themselves be perceived, but introspective states can be introspected. Therefore, introspection is unlike perception in this fundamental way, illustrating yet another reason to reject the idea that introspection is a kind of perception.

Introspection and attention

Before wrapping up my thoughts regarding the perceptual account of introspection and moving on, it will be worthwhile to consider one more option for such an account. Although this option ultimately fails to support a perceptual understanding of introspection, it will start heading us in the right direction, toward some possibilities that will be developed in later chapters.

According to some proponents of the introspection as perception account, inner perception occurs through attention being directed toward our own mental phenomena. William Lycan, in particular, states, ‘As I would put it, consciousness is the functioning of internal atten-tion mechanisms directed at lower-order psychological states and events’ (1996, p. 14). From this perspective, one might deny the existence of an explicit inner perceptual device, just as I have done in the preceding sections, but still maintain that introspection is perceptual in nature nonetheless, in terms of higher-order attentional processes similar to (or perhaps even identical to) those that play a role in our everyday perception of the world. However, this viewpoint can be understood in two rather different ways: either there is some single window of atten-tion through which we access the contents of our own minds, perhaps analogous to a spotlight that brings attention to particular events or objects over others, or there are a number of parallel and/or intercon-nected attentional processes happening in the mind on a sub-personal level that play a role in some way or other in forming our introspective abilities. So, let us consider each of these possibilities. We will find that


the former is implausible and that the latter, rather than supporting a perceptual account of introspection, actually leads us toward another quite different, but more explanatorily satisfying, perspective.

Insofar as the attention account entails positing some single atten-tional device in the brain, there is a significant problem that such an account confronts. The problem is a version of the homunculus worry that regularly creeps up in cognitive science and the philosophy of mind, in that it postulates some kind of inner agent within the mind (in this case, an inner observer of sorts that functions by attending to the diverse functions of the mind). The room for such homunculi is limited, both in terms of actual, concrete processes in the mind and in terms of explanatory devices in the sciences of the mind. I tend to lean toward the idea that there is really no room for them at all; agents are things to be explained themselves, and not explanatorily fruitful posits in the attempt to understand mental abilities and cognitive processes. So, positing some attentional agent in the brain in order to explain how we attend to our own minds really does not get us anywhere. It is tanta-mount to replacing one problem with another.

Lycan himself considers, accepts, and responds to this concern. Following Dennett, Lycan describes the problem as ‘Cartesian materi-alism’, in that some kind of inner theater is postulated where it ‘all comes together’; where some audience-like agent sits back and observes the events in the mind (1996, pp. 30ff; Dennett, 1991 ). It is worth noting here that this Cartesian conception of an inner observer in the head has been the target of much, and much-deserved, criticism in philos-ophy and cognitive science. From Gilbert Ryle’s early logical behaviorist critique of the Cartesian ghost in the machine (1949), which was a big influence on Dennett’s further critique of the Cartesian theater, to recent 4E (embodied, embedded, extended, and enactive) anti-Cartesianism (e.g., Wheeler, 2005 ; Menary, 2007 ; Rowlands, 2010 ), the idea that there is some inner observer inside the mind has been vigorously criti-cized in the philosophy of mind and cognitive science across the past several decades. Lycan accepts this as a legitimate worry, but then goes on to say:

But it should be clear that the inner-sense view is not per se committed to Cartesian materialism. For even if an internal scanner resembles an internal audience in some ways, the ‘audience’ need not be seated in a Cartesian theater: There need be no single , executive scanner, and no one scanner or monitor need view the entire array of first-order mental states accessible to consciousness. Accordingly, there


need be neither a ‘turnstile of consciousness’ nor one central inner stage on which the contents of consciousness are displayed in fixed temporal order. An internal monitor is an attention mechanism, which presumably can be directed upon representational subsystems and stages of same. No doubt internal monitors work selectively and piecemeal, and their operations depend on control windows and other elements of conative context. On these points, the inner-sense theory has already parted with Cartesian materialism. (1996, p. 32)

This is indeed a move away from the homunculus problem embedded in the Cartesian perspective. If we dismiss the idea of a single atten-tional spotlight where all the contents of the mind come together to be perceived by a unitary observer, and replace this with attention mechanisms on a subsystem level, we no longer need to worry about a hidden homunculus doing our explanatory work. However, rather than supporting the perceptual account of introspection, such that intro-spection is regarded as a unique perceptual modality unto itself, this tactic is actually a significant move away from it.

First of all, there are functional and physiological reasons for sepa-rating attention processes from perception processes, despite the fact that they work together in various ways to produce our mental lives as we experience them. Most notably, there is pertinent neurological evidence that perception and attention (or, more specifically, primary sensory processing and secondary attentional processing) involve iden-tifiably different areas of the brain (Fuster, 2002 ; Picton et al., 2002 ). Very roughly, the posterior regions of the primate neocortex are devoted to processing sensory/perceptual data, while the frontal regions seem to involve the more supervisory, monitor-like cognitive domains associ-ated with attention (Fuster, 2002 , esp. pp. 97, 100, and 102). If atten-tion and perception functions occur in noticeably different areas of the brain, as these data suggest, we simply should not regard the explana-tion of introspection through attention mechanisms as supporting the inner perception model. These are very different kinds of processes and should not be confused with one another.

Secondly, fragmenting the mechanisms involved in introspection, as Lycan suggests, is a step away from the common-sense nature of the perceptual account, and a step toward the type of cognitive account I will defend in Chapter 5 , in explaining important and fundamental aspects of our introspective abilities through more basic non-introspec-tive cognitive processes. One of the main intuitive appeals (perhaps even the main appeal) of postulating a perceptual or perception-like


faculty in the mind is that it fits with our intuitive sense of unity in our conscious experience. If we do, in fact, observe our mental lives, it is from the perspective of the single, conscious ‘me’ that we are all inti-mately familiar with in our subjective experience. However, if we think of introspection not as a single mode of perception, but rather as a collection of smaller, less agent-like, and sub-personal attention mecha-nisms, then it ceases to fit with this commonsensical way of thinking about ourselves. To be clear, I am not suggesting here that we think of ourselves and our introspective abilities as a single, unified thing (To the contrary!). The point is that our intuitive unity of consciousness is part of the idea that we perceive our mental lives through some kind of ‘inner sense’, as maintained by Locke, Kant, James, and others. If we divorce the perceptual account of introspection from this intuitive idea, then we are moving away from the idea that introspection constitutes a unique kind of perceptual access to our own minds. Of course, I see this as a step in the right direction. As will be discussed in the next chapter, introspection is a diverse, heterogeneous collection of processes, and not a single process or even a single type of process. So, despite the fact that Lycan persists in calling his account a version of the Lockean inner-sense account, I think that his move toward fragmentation is a good idea. Attention does indeed need to be considered as an aspect of introspection, as will be explained in some detail later, but not in terms of a single attentional device, and especially not in terms of a unique kind of inner perception. As I will next explain, applying the idea of perception here is merely metaphorical and is not the development of a literal account of what happens in introspection.

41

So, if introspection is not a kind of perception, what is it? What are we doing when we introspect? The primary goal of the remainder of this book is to develop a satisfying answer to this question, providing a viable understanding of introspection and its epistemic properties. Before turning to the details, however, I want to emphasize here in this transi-tional chapter the fact that introspection is a diverse phenomenon, held together not by a single kind of cognitive process or epistemic status, but rather by a metaphorical unity applied to the subjective first-person orientation we have towards our own minds. Key to understanding the diversity of introspection is the realization that the concept of intro-spection itself, as a looking within, is a metaphor. If we want an accu-rate characterization of introspection, we cannot take the concept itself literally. It is, rather, a unifying perceptual metaphor projected onto a diverse set of epistemic relationships we hold to ourselves. Once this is understood, the concept of introspection is opened up for revision, to be reconsidered and reconceived. Regarded as a metaphor, introspec-tion need not be confined to a singular mechanism or process. Instead, it can be treated as a unifying concept for multiple states and processes at work in our first-person epistemic grasp of our own minds. The artic-ulation of this point here will provide a framework for understanding how the various aspects of introspection discussed in subsequent chap-ters fall together under this one umbrella concept of ‘introspection’.

The prominence of perceptual metaphor

In light of the arguments developed in the previous chapter, we can safely dismiss the idea that introspection is a kind of perception. But if introspection is not a kind of perception, then why do so many

3 Introspection as a Metaphor


theorists of the mind, both past and present, conceive of introspection in a perceptual manner? To answer this question, I find it informative and useful to take a look at the work of George Lakoff and Mark Johnson on the prominent roles that metaphors play in cognition (1980; 1999). Through application of their general account of cognitive metaphors to the particular case of introspection, I suggest that the enduring attrac-tiveness of the perceptual account of introspection can be explained by the prominence of perceptual metaphors in human cognition. We metaphorically project our concrete embodied familiarity with percep-tion onto our general understanding and conceptualization of knowl-edge, and this metaphorical projection of perceptual concepts onto the domain of knowledge is what underlies our tendency to understand our first-person self-knowledge in a perceptual manner.

Emphasizing an embodied approach to philosophy and cognitive science, Lakoff and Johnson claim that our higher cognitive abilities are essentially dependent upon our body-based sensorimotor processes. Roughly, the idea is that our bodily engagement with the world forms the basis of our concepts and ways of thinking about how the world works. We form our basic conceptual structures through body-centric engagement, as we navigate our way in life via sensorimotor interaction with the world. In turn, these basic conceptual structures are abstracted from these origi-nating embodied domains and applied metaphorically to other domains, allowing us to bootstrap ourselves into higher-level abstract conceptuali-zations of our experiences. Here is how Lakoff and Johnson describe it:

Our subjective mental life is enormous in scope and richness. We make subjective judgements about such abstract things as impor-tance, similarity, difficulty, and morality, and we have subjective experiences of desire, affection, intimacy, and achievement. Yet, as rich as these experiences are, much of the way we conceptualize them, reason about them, and visualize them comes from other domains of experience. These other domains are mostly sensori-motor domains, as when we conceptualize understanding an idea (subjective experience) in terms of grasping an object (sensorimotor experience) and failing to understand an idea as having it go right by us or over our heads. The cognitive mechanism for such conceptuali-zations is conceptual metaphor, which allows us to use the physical logic of grasping to reason about understanding. (1999, p. 45)

From this perspective, metaphors are much more than literary devices. They serve as basic conceptual structures that play a constitutive role in

Introspection as a Metaphor 43

human cognition. According to Lakoff and Johnson, this cognitive use of metaphor is pervasive, encompassing the full range of human experi-ence and thought, from emotions and personal relationships to polit-ical orientations and metaphysical frameworks. The explanatory reach of Lakoff and Johnson’s theory is worth considering, but, to be clear, I do not wish to make any commitments here regarding the extent to which Lakoff and Johnson’s perspective can account for human cogni-tion in general. Rather, my purpose is to apply it to the much more limited domain of introspection. Whether there are other domains of thought that utilize embodied metaphorical projections, the under-standing of introspection in terms of perception clearly involves this type of cognition.

In order to see this, let us first look at the metaphorical use of percep-tion to understand knowledge more generally. Lakoff and Johnson note that

We get most of our knowledge through vision. This most common of everyday experiences leads us to conceptualize knowing as seeing. Similarly, other concepts related to knowing are conceptualized in terms of corresponding concepts related to seeing. In general, we take an important part of our logic of knowledge from our logic of vision. Here is the mapping that projects our logic of vision onto our logic of knowledge.

The Mind Is a Body Thinking Is Perceiving Ideas Are Things Perceived Knowing Is Seeing Communicating Is Showing Attempting to Gain Knowledge Is Searching Becoming Aware Is Noticing An Aid to Knowing Is a Light Source Being Able to Know Is Being Able to See Being Ignorant Is Being Unable to See Impediments to Knowledge Are Impediments to Vision Deception Is Purposefully Impeding Vision Knowing from a ‘Perspective’ Is Seeing From a Point of View Explaining in Detail Is Drawing a Picture Directing Attention Is Pointing Paying Attention Is Looking at


Being Receptive Is Hearing Taking Seriously Is Listening Sensing Is Smelling Emotional Reaction Is Feeling Personal Preference Is Taste (1999, p. 238)

Reflecting upon these common metaphorical constructions, it is clear that knowledge is often conceptualized in sensorimotor terms throughout everyday thought and discourse. We can all think of particular instances of the above metaphorical paradigms in our own thoughts and words, as well as those of others. I, for instance, have repeatedly written in this book of ‘looking’ at arguments and considering ‘viewpoints’ or ‘perspectives’. Surely most, if not all, speakers of contemporary collo-quial English have said or heard the words ‘I see what you mean’ innu-merable times, typically without giving even the slightest notice to the metaphorical nature of the statement. It is obvious in everyday life that such language is a prominent way of conceptualizing knowledge and understanding. Yet, none of these sorts of conceptualizations can be said to be what is literally going on in our minds. We do not actu-ally see meanings, and philosophical theories are not viewpoints in the literal sense of being perceptual outlooks onto the world (even though they may affect how we understand and respond to what we perceive, however).

I wager that this sort of metaphorical conceptualization is what is really going on when people think of introspection in terms of inner perception. We do not actually see or otherwise perceive the diverse contents of our own mental lives. We do not literally look within ourselves in our self-reflective moments throughout life. These are just ways of conceptualizing self-knowledge in terms of something with which we are already familiar. Getting an actual grasp of what literally goes on in our minds when we introspect is really quite difficult, and can even lead us into confusing and seemingly paradoxical territory ( see , for instance, Hofstadter, 1979 ; Berm ú dez, 1998 ). So, it should really be no surprise that we use our intuitive understanding of perception as a metaphorical projection to get a grasp on what happens in introspec-tion. However, as I have shown in the preceding chapter, it would be a mistake to take this understanding as a literal account. If we want to understand the ways in which we actually acquire self-knowledge, we need to set the perceptual metaphor aside and do the hard work needed to obtain an accurate understanding of what we are really doing when we introspect.


What is introspection?

If introspection is not inner perception, what is it? Of course, one possi-bility is that it simply does not exist. If there is no looking within, why not simply eliminate the concept of introspection altogether? William Lyons’ The Disappearance of Introspection ( 1986 ) heads in this direction. The term ‘introspection’, as Lyons notes (p. 1), is derived from the Latin words for ‘to look’ and ‘within’ ( spicere and intra , respectively). But if there is no literal referent to the concept of looking within, as both Lyons and I agree, why should we continue to use the term? Lyons’ suggestion here is that ‘introspection’ is on the way out as a viable concept, gradually disappearing from the intellectual landscape as we obtain a more accurate understanding through the sciences of the mind. From this perspective, the concept of introspection is a relic of an outdated folk psychology, alongside other antiquated notions like ‘belief’, ‘desire’, ‘intention’, to be replaced with scientifically-informed neurophysiological terms that more accurately capture the workings of the mind (p. 155).

This eliminativist response could perhaps work if everyone unilat-erally agreed to the rejection of the term, but such an event is both unlikely and unnecessary. Top-down rational revision of language is a tough (if not impossible) row to hoe, especially if a term is deeply entrenched in both colloquial and theoretical usage, as is the case with introspection. The resurgence of work on and appeal to introspection since the publication of Lyons’ book is itself a case in point. If introspec-tion had, in fact, been on the cusp of disappearance from philosoph-ical and scientific investigation of the mind, as Lyons suggested, then further attention to the topic ought to have been on the decline, just as the term ‘phlogiston’ fell off the map of scientific discourse towards the end of the eighteenth century. But this is most definitely not the case. Attention to, and reliance upon, introspection has grown significantly since Lyons’ book, especially with the rise of serious and widespread study of consciousness in recent years. In fact, understanding the nature, epistemic status, and methodological role of introspection is at the heart of much contemporary debate among both philosophers and scientists (e.g., Varela & Shear, 1999 ; Jack & Roepstorff, 2003 ; Hurlburt & Schwitzgebel, 2007 ; Petitmengin, 2009 ). The elimination of ‘intro-spection’ from discourse on the mind does not appear to be on the horizon anytime in the foreseeable future.

Nor, however, is it really desirable in the first place. Despite the fact that introspection should not be understood literally as inner perception,


the term and its associated metaphorical concept can play legitimate roles in the right contexts. It is already in place in our common under-standing, serving the role of referencing our first-person understanding of our own minds, and there is no reason why it should not continue to do so, as long as we recognize its non-literal nature. Compare the concept of introspection with the concept of a sunrise. The sun, of course, does not literally rise as it appears on the horizon. Yet, we still talk that way, irrespective of our post-Copernican understanding of our solar system. Despite reasonable grounds for elimination, we continue to use the term ‘sunrise’ to refer to this daily event in our lives, even in our official record keeping. We simply find it useful to refer to and conceive of it in that way for everyday purposes and it works. So, just as the idea of a sunrise persists in everyday colloquial English to refer to the daily rotation of the earth, so too can the term introspection continue to be used to refer to the first-person stance that we hold with regard to our own minds. As Lakoff and Johnson help us realize, meta-phors play a significant role in human cognition. Along these lines, it is normal and acceptable to use the perceptual metaphor behind the concept of introspection to understand knowledge of our own mental states.

So, how then are we to understand the meaning of ‘introspection’, in terms of a literal non-metaphorical characterization of the concept? Here is the working definition I offer: Introspection is the process of seeking and/or acquiring knowledge of one’s own mind, from one’s own subjective first-person standpoint.

Exactly what this conception of introspection involves will be unpacked throughout the remainder of this book, but a few prelimi-nary observations are in order here. First, we must include both seeking and acquiring in the definition to accommodate the fact that intro-spection sometimes succeeds in the acquisition of knowledge but also sometimes does not. Introspection can result in one having genuine knowledge about oneself, in which case it constitutes the acquisition of knowledge. However, as will be detailed later, introspection can fail to provide knowledge as well. Introspection can lead to false conceptions about oneself, or fail to generate any conceptions at all. Importantly, introspection does not automatically entail success, and thus we must include the possibility of knowledge failure in its definition. Moreover, introspection might be acquired without being sought. It is possible to know oneself in some manner or other without necessarily having intended to seek out that knowledge. In fact, I might introspectively


know something about myself that I’d rather not know. So, just as intro-spection cannot be limited to the actual acquisition of knowledge, it cannot be limited to the seeking of knowledge either.

It is also important to include in the definition of introspection the fact that it is sought or acquired from a subjective first-person standpoint. Knowledge about one’s mind from a generically objective third-person standpoint, in contrast, is not typically regarded as intro-spective. I might know that I have a mental disorder, through an fMRI scan or behavioral diagnosis, for instance, without that knowledge qualifying as an instance of introspection per se. In fact, notoriously, I might be introspectively oblivious to my mental disorder, even in the face of overwhelming external evidence. Obtaining knowledge of a mental disorder through third-person means, however, might come to take on an introspective role, if for instance it allows me to recog-nize aspects of myself to which I was previously inattentive. There are important and notable relations between first-person introspection and objective externally-available information and procedures, as we will see, so a strict separation between the two may not be possible, but it is nonetheless characteristic of introspection to occur from a subjec-tive orientation, via one’s first-person experience of or access to one’s mind. There will be much more to say about this territory later, starting with the unique subjectivity involved in knowing one’s own conscious experiences that is the focus of the next chapter, so let it suffice now to note that the subjective first-person standpoint that is characteristic of introspection can set it apart from other epistemic orientations in important and interesting ways, hence its inclusion in the definition above.

Finally, we must not take the definition of introspection as a process to denote it as a single specific process, distinctly aimed at obtaining knowledge of mental states. Understanding introspection as the process of seeking or acquiring self-knowledge through first-personal means does not entail that there is a distinct mechanism that is THE process that constitutes introspection. Instead, as we will see, there are multiple states and processes that can be introspective in some manner or other, some of which are not distinctively introspective in nature and some of which are quite different from others, both in terms of the mechanisms they involve and the epistemic characteristics they exhibit. So, the term ‘process’ in the definition I offer should be under-stood as broadly and generically as possible, inclusive of diverse kinds of processes.


The diversity of introspection

Philosophers, scientists, and people in general have a tendency towards unifying phenomena under a single explanatory framework, whether that framework involves positing a single deity to explain the entirety of existence, a single principle to prescribe ethical behavior, or a single formula to encapsulate a broad spectrum of phenomena within a scien-tific theory. Of course, this tendency is based on processes that we need in order to understand ourselves and the world we live in, but it can also cause us to overlook the inherent complexity of some phenomena. I contend that the latter is the case with introspection. Contrary to our tendency to think of introspection as a single faculty through which we access and become conscious of the contents of our own minds, we must realize that there is no single cognitive phenomenon we can point to and say ‘that’s introspection’. As explained above, the idea of introspection itself is a metaphorical projection of perception onto our first-person access to our own minds. The concept of introspection orig-inates in a unifying categorization of the diverse sorts of things that we can know about ourselves. What makes introspection a unitary concept is the fact that it refers to the subjective first-person epistemic orienta-tion that we hold with regard to ourselves, but this orientation can take multiple forms and utilize diverse processes.

First, and most primitively, there is the phenomenal awareness involved in one’s own conscious states, such as an immediate feeling of a pain. Insofar as we are embodied conscious beings, we know what it is like to be ourselves and have the experiences we have, simply in virtue of the fact that we are ourselves. As I will argue in the next chapter, this constitutes a unique kind of self-knowledge that lies at the core of introspection. However, despite its unique and essential role in self-knowledge, this phenomenal knowledge only constitutes a small portion of what we call introspection. To understand the full spectrum of our introspective abilities, especially the full-blown introspective moments when we are actively engaged in thinking about ourselves, we must turn to other higher-level processes and abilities. Here we can recognize the role of the attention processes involved in one’s own expe-rience and cognition, those that we draw upon when, for instance, we are attending to some particular auditory sensation, such as a linguistic utterance, or remembering some past event in one’s own life. Such proc-esses enable us to attend to the contents of our own minds, thereby initiating the second-order reflections involved in much of our intro-spective efforts. Coupled with these attentional abilities are our abilities


to represent things conceptually, including things in our minds and traits we may attribute to ourselves. By applying our philosophical and scientific understanding of these basic sorts of cognitive abilities (abili-ties we have outside the domain of introspection) to the domain of introspection, we can understand important facets of our introspective abilities. For instance, we can understand self-attributions of beliefs and desires in terms of our more basic commonsense conceptual abilities, thereby obtaining some insight into how we understand and explain ourselves, both to ourselves and to others. Another important facet of introspection pertains to our linguistic abilities, most particularly our ability to engage in inner speech. Through this phenomenon, we can understand a diverse range of introspective capabilities, ranging from our ability to know our own immediate thought processes to our ability to construct rational narratives about the overarching course of our lives, with an equally diverse range of epistemic properties, from self-determining truth values to self-deceptive capabilities.

The details of all these various aspects of introspection will be cashed out in the chapters to come, but this brief overview alone is sufficient to make apparent the cognitive and epistemic diversity of introspection. We obtain information about ourselves through a rather heterogeneous collection of psychological processes, processes which vary significantly in their epistemic qualities. It is important to emphasize this epistemic diversity, hence my reiteration of this point throughout this book. All too often, introspection is mischaracterized as a single kind of process, and with this comes the erroneous idea that there is a single epistemic dimension to introspection. Some people regard it as infallible or epis-temically privileged in some way, while others denigrate introspec-tion as significantly biased and/or untrustworthy. However, whether it is the insistence upon infallibility in Cartesian foundationalism or the cursory dismissal of introspection often found in the beginning of psychology textbooks, all such unitary assessments of introspection are misguided. These singular accounts may capture some salient features of introspection, but we cannot characterize the nature and epistemic qualities of introspection under a single categorical assessment or formulaic expression. Introspection as a whole is neither infallible nor epistemically deplorable. It can be either, or somewhere in between, depending upon what aspect or kind of introspection is at work in a particular case. Introspective information comes from a heterogeneous collection of processes, each with their own unique epistemic pros and cons. So, we have to consider which domain of introspection we are talking about if we are to say something useful or accurate about it


and its epistemic properties in a given context. Indeed, this is how I will approach introspection in the following chapters. I will discern, illustrate, and analyze importantly different modes of introspection, in attempt to better understand them and assess their unique epistemic qualities. The cumulative result will be a more nuanced and accurate framework for understanding how and what we know, and can fail to know, about our own minds from our own first-person perspectives.

51

At the most basic level, our introspective knowledge of our own minds originates with the knowledge we have of our conscious experiences through those very experiences themselves. Simply put, we know ourselves by being ourselves, not as objects but rather through the qual-itative character of our experiences as living subjects. Many philoso-phers have brought attention to the qualitative, experiential character of consciousness, often referred to as phenomenal consciousness, and I will draw on this attention here to explicate an immediate kind of knowledge that we have in virtue of being constituted by our conscious states as embodied beings in the world. In other words, our lived experi-ences themselves are a kind of knowledge, which may be referred to as ‘phenomenal knowledge’. As a kind of self-knowledge that is intrinsic to lived experience itself, phenomenal knowledge can be understood as an elemental component of introspection, with unique epistemic proper-ties that are distinct from those of other kinds of knowledge, including other kinds of introspective knowledge.

The view of phenomenal knowledge that I explain and defend here runs counter to most recent accounts of phenomenal knowledge in analytic philosophy of mind, which generally attempt to relate or reduce phenomenal knowledge to some other recognized kind of knowledge. In direct contrast, I claim that phenomenal knowledge is a unique type of knowledge unto itself, distinct from the other types of knowledge typically discussed in mainstream epistemology. I will characterize this kind of knowledge in terms of it being constituted by conscious experi-ence itself, but in a way that falls between the cracks of the more conten-tious issues regarding the metaphysical status of consciousness, which has been and continues to be the focal point of most philosophical work concerned with phenomenal knowledge. In fact, in partial justification

4 Knowing Our Own Consciousness


of this omission, as well as to circumscribe the proper epistemic domain of this unique type of knowledge, I will claim and argue that conscious experiences in themselves do not reveal to us their metaphysical nature. Rather than taking a stand on the nature of consciousness, my purpose is to bring to light the epistemic aspects of phenomenal consciousness, particularly in regard to the knowledge that experiencing agents obtain of themselves through conscious experience. Of course, the metaphysics and epistemology of consciousness may have significant interconnec-tions with each other, 1 but we need not always attend to both and can justifiably parse them apart for our purposes here.

To achieve an understanding of the role of phenomenal consciousness in introspection, I will begin by introducing phenomenal conscious states as they have been characterized in contemporary philosophy of mind. I will then analyze the kind of knowledge that such states establish, pointing out the flaws that occur among various attempts to understand phenomenal knowledge in terms of other kinds of knowledge. Following this critique, I develop my alternative existential constitution model of phenomenal knowledge and defend the idea that this constitutes a unique and genuine kind of introspective knowledge. Finally, I will offer some clarifications regarding the nature and scope of this kind of introspective knowledge and explain how skepticism toward introspective self-knowledge erroneously ignores the unique epistemic characteristics of phenomenal knowledge.

Knowing what it is like

To get a grasp on the kind of self-knowledge that is involved in the immediate phenomenal experience of one’s own conscious states, let us overview two paradigmatic illustrations given by the philosophers Thomas Nagel and Frank Jackson, both of whom suggest that there is something special about conscious states in virtue of their subjective, phenomenal qualities. Readers already familiar with this work may wish to skip ahead to the next section, but a brief refresher will not hurt and will also help indicate what aspects of this work I will and will not be utilizing in subsequent sections.

In his highly influential chapter ‘What is it like to be a bat?’ Thomas Nagel draws attention to the subjective, felt quality of conscious mental states (1974). The use of the phrase ‘what it is like’ has become a common way of referring to the phenomenal aspect of conscious experience in both analytic and phenomenological philosophy of mind (e.g., Block et al., 1997 ; Thompson, 2003 ; Gallagher & Zahavi, 2008 ), so it is worth

Knowing Our Own Consciousness 53

looking at what exactly Nagel had in mind when he brought the phrase into philosophical discourse. He states, ‘an organism has conscious mental states if and only if there is something that it is like to be that organism – something it is like for the organism’ (1974, p. 519). So, conscious states, as Nagel characterizes them, are identifiable with the experiences that an organism has as a subject of experience. There is some phenomenal quality to the organism’s experience, such that there is something it is like for the organism to have the experience, to be in that particular state. Notice also the fact that Nagel puts the term ‘ be ’ in italics, drawing attention to the existential dimension of conscious experience. Although Nagel himself does not put it this way, one crucial feature of conscious experience is the fact that organisms are, at least in part, constituted by them as embodied subjects in the world. This will be important later when I characterize the kind of knowledge that such conscious states produce.

In conjunction with his characterization of consciousness, Nagel suggests that consciousness fails to be understood through objective, physical knowledge, most particularly the kind of knowledge pursued by science. Indeed, Nagel’s primary point is to illustrate the inability of current scientific knowledge to account for conscious experience. To this effect, he suggests that no amount of knowledge concerning the cogni-tive and physiological processes of a bat can tell us what it is like, on the inside (so to speak), for a bat to experience echolocation. Only a being that undergoes the experience of echolocation as a bat can know what it is like to have such an experience. Objective physical information simply cannot portray to us the subjective, experiential dimension of conscious states, or so it seems to Nagel. From this basis, Nagel concludes that consciousness cannot be reductively explained in terms of physical processes, at least not without radical changes in our understanding of physical reality. However, despite having some agreement with Nagel, I want to separate myself from this further claim concerning the physical explanation of consciousness. In fact, I will later offer a criticism of such inferences, inferences that immediately extrapolate from phenom-enal experiences themselves to propositional claims about their nature. For the time being, however, I want to focus only on the qualitative, experiential character of consciousness. Irrespective of whether they can be accounted for in physical terms, conscious experiences exhibit phenomenal qualities such that there is something it is like to have, or be constituted by, them. Regarding this portrayal of conscious mental states, I think Nagel is right on track. There are qualitative aspects to feeling pains, seeing colors, hearing sounds, tasting flavors, and having


emotions, such that we can say there is something it is like to be in such states.

Although he explicitly distinguishes his arguments from Nagel’s, Frank Jackson offers some similar concerns regarding the unique expe-riential character of consciousness. To illustrate his position, Jackson proposes the hypothetical example of a future neuroscientist, Mary (1982; 1986) . Mary knows all the physical facts regarding the neuro-physiology of vision, understanding completely and exactly what goes on in the brain when people see. But there is one important catch about Mary’s situation: She has been confined to a black-and-white environ-ment from birth. All of her surroundings are cast in black, white, and varying shades of gray, and she has somehow been prevented from seeing the color of red at all, presumably including even in her dreams and imagination. 2 Given this scenario, Jackson goes on to consider what might happen if Mary were suddenly allowed to experience color:

What will happen when Mary is released from her black and white room or is given a color television monitor? Will she learn anything or not? It seems just obvious that she will learn something about the world and our visual experience of it. But then it is inescapable that her previous knowledge was incomplete. But she had all the physical information. Ergo there is more to have than that, and ‘physicalism’ is false. (1982, p. 558)

Like Nagel, Jackson’s primary motivation is to show that the phenom-enal aspect of experience escapes physical understanding. When Mary finally comes to experience colors, she purportedly learns something that no amount of physical information could possibly give her. There is some kind of new knowledge acquired when she first experiences color vision, knowledge of what the experience itself is like. On this basis, Jackson concludes that there must be more to phenomenal states than physical processes, since complete scientific knowledge of physical proc-esses does not provide phenomenal knowledge of the qualitative char-acter of conscious experience. However, as previously mentioned, I want to set aside metaphysical claims about the relationship between phenom-enal mental states and physical states. In fact, I think that Jackson is much too quick to conclude on the basis of the Mary example that phenomenal states must be non-physical in some manner. 3 This will be touched on later, but for now I simply want to draw attention to the fact that Mary does indeed learn something new when she experiences color vision for the first time, irrespective of any metaphysical implications


this knowledge may or may not have. Although Mary thoroughly under-stood how color vision happens in functional and physiological terms, she did not know what it is like to actually undergo color vision when she was confined to her black-and-white environment. Only by actually having a color vision experience does Mary come to know the qualita-tive, phenomenal character of the experience. Although I disagree with the conclusion about physicalism, I think that Jackson’s thought experi-ment here is quite useful in drawing attention to the unique phenom-enal characteristics of conscious experiences. There is indeed something that it is like to have the particular phenomenal experiences we enjoy as conscious beings, such that one can only know these experiences by actually having them. This constitutes a very basic and essential compo-nent of self-knowledge, in that one knows what one’s own conscious experiences are like simply by, and only by, having them.

The inadequacy of standard models of phenomenal knowledge

If we obtain knowledge through the qualitative phenomenal character of conscious experience, then what kind of knowledge is this? Following in the wake of Nagel and Jackson’s work on consciousness, a number of viewpoints have been expressed on this topic, the overwhelming majority of which attempt to understand phenomenal knowledge in terms of another, presumably better understood, type of knowledge, such as propositional, ability, or acquaintance knowledge (Ludlow et al., 2004 ; Alter & Walter, 2006 ; Nida-Rümelin, 2010 ). I, however, will take a different route and argue that phenomenal knowledge is a distinct kind of knowledge unto itself, fundamentally different from other kinds of knowledge. As I will characterize it here, phenomenal knowledge is not knowledge we acquire in the same way that we acquire other common kinds of knowledge, by learning a propositional fact, developing an ability, or becoming acquainted with an object. It is, rather, a kind of knowledge we have as subjective loci of experience, constituted by the having of experience itself. I call this characterization of phenomenal knowledge the existential constitution model of phenomenal knowl-edge, in reference to the unique property that the knowledge obtained is constituted by conscious experience itself, in virtue of our concrete existence as conscious subjects in the world, rather than in virtue of learning something about an experience as an object.

Before explaining this model of phenomenal knowledge, however, it will be instructive to first explain why other models fail to provide a


satisfying account of phenomenal knowledge. There is a vast body of work on this topic (Ibid.), so it would take us astray to consider all of the options here in detail. However, a brief overview of why the main contending models are inadequate will both provide some background context and help motivate my reasons for developing my alternative existential constitution model. I will thus explain why phenomenal knowledge is not adequately characterized in terms of knowing prop-ositions about conscious experiences, knowing how to recognize or identify kinds of experiences, or by being acquainted with objects in experience, which together constitute the main options aired in the literature thus far.

First, consider propositional knowledge. Propositional knowledge occurs when a propositional fact is known, such as knowing that the capital of Arkansas is Little Rock or that the sun is a star. Propositional knowledge has been the primary object of analysis in mainstream epis-temology, so it is not surprising that philosophers have tried to under-stand phenomenal knowledge in propositional terms. In fact, many simply assume that knowledge must be propositional in some form or other and carry this assumption over into the analysis of phenomenal knowledge. Jackson, for example, characterizes phenomenal knowledge in terms of non-physical facts and this characterization has shaped how the issue is treated in much of the literature (Jackson, 1982 ; 1986 ; Ludlow et al., 2004 ). Contrary to this background theoretical orienta-tion towards propositional content, however, I think it is quite clear that no amount of propositional knowledge is sufficient for phenomenal knowledge, for any propositional content can be stated without thereby conveying phenomenal content. For instance, no amount of proposi-tional description, by itself, can conceivably teach a blind person what it is like to see red. Propositional description can teach facts about red, of course, but none of those facts, in themselves, constitute knowledge of the phenomenal character of redness itself. This, if anything, is what Jackson’s Mary example ought to teach us: no conceivable propositional content, however detailed or complete it may be, can tell her what her experience will be like when she sees red for the first time, even if that experience is itself a physical and scientifically understandable event. Assuming that phenomenal knowledge is propositional leads Jackson to the implausible and vacuous idea of non-physical facts, when dropping the propositional assumption about knowledge instead would accom-modate his key example just as well.

Motivated in part by this observation, some theorists have attempted to characterize phenomenal knowledge in terms of abilities rather than


propositions (e.g., Lewis, 1988 ). From this perspective, what Mary learns when she sees red for the first time is not some mysterious proposi-tional fact but rather the ability to recognize the color red. This over-comes the theoretical bias toward propositions described above, but it will not work either, as having the ability to recognize red does not capture the phenomenal character of the experience itself. Of course, ability and experience often occur together in everyday life, but the two are distinctly different in nature and conceptually distinguishable from each other. There is no clearly identifiable relationship between cogni-tive ability and phenomenal experience, such that the former might provide an adequate conceptualization of the latter. Conceivably, we can imagine a conscious being who experiences red without thereby being able to recognize it again on subsequent occasions, perhaps due to a lack of memory or other cognitive capacity. On the flipside, we can conceive of a being who has the ability to recognize red but who has no phenomenal experience of the color (a non-conscious color-detecting machine, for example). Either way, perceptual ability and experience can be decoupled from each other, rendering the ability account of phenomenal knowledge untenable.

We need not rely on conceptual analysis alone to see the flaws in the ability account of phenomenal knowledge. The well-known phenom-enon of blindsight is an apt empirical counterexample to consider here, illustrating how ability and phenomenal experience can come apart in human beings. Blindsight patients are people who are experientially blind, due to certain specific kinds of brain damage, but who are none-theless able to make accurate perceptual judgments, without conscious experience of what they perceive (Weiskrantz, 1986 ). For example, when asked to place envelopes through mail slots, blindsight patients consistently deny that they see or know the angle of the slot and yet are capable of accurately aligning envelopes through slots at various angles when compelled to do so, thereby exhibiting identifiable perceptual abilities without any corresponding phenomenal knowledge regarding the ability. Given that perceptual ability and phenomenal experience can come apart in this manner, it is plausible to conclude that phenom-enal knowledge cannot be identified with perceptual ability per se, even if perceptual ability may be somehow otherwise related to phenomenal experience. Considering this, we may justifiably reject ability knowl-edge as an adequate model for understanding the nature of phenom-enal knowledge.

This leaves acquaintance knowledge as the last remaining option for understanding phenomenal knowledge in terms of another recognized


kind of knowledge. Simply put, acquaintance knowledge is knowledge of an object with which one is familiar through direct experiential contact, such as when we say we know a person or place. In line with this, acquaint-ance theorists maintain that phenomenal knowledge is acquired through a particularly special and direct relationship we hold to our own minds, such that we know our own consciousness by being directly acquainted with our conscious experiences themselves, rather than through a medi-ating proposition or ability. This approach to phenomenal knowledge has gained some notable ascendancy in recent work, with multiple philoso-phers from divergent theoretical perspectives converging toward the idea of acquaintance in accounting for phenomenal knowledge. 4 To illustrate, here is a sampling of philosophers who defend an acquaintance model of phenomenal knowledge in some form or other:

Knowing what an experience is like is knowledge essentially resting on acquaintance by introspection. ... There is a kind of knowledge of the phenomenal properties of your own mental state that is not available to other persons unless they stand in an introspective acquaintance relation to one of their own mental states, and that state shares the phenomenal properties of your mental state. ... Some kinds of knowl-edge require distinctive forms of engagement between the knower and the known. That is why, although qualia are physical, people cannot know all there is to know about them unless they experience them for themselves. (Bigelow & Pargetter, 1990 , pp. 146–147)

The claim is that there is a kind of knowledge of a phenomenal quality, knowledge by acquaintance, which can consist in attentively experiencing the quality rather than possessing information or abili-ties. This is not an exotic epistemic state. It is neither ineffable nor unmistakable. It is the familiar sort of knowledge to which we refer when we discuss knowing people and places as well as experiences. (Conee, 1994 , p. 147)

... we bear a special relation to the phenomenal properties instan-tiated in our experience ... This relation would seem to be a particu-larly intimate one, made possible by the fact that experiences lie at the heart of the mind rather than standing at a distance from it; and it seems to be a relation that carries the potential for conceptual and epistemic consequences. We might call this relation acquaint-ance . ... Acquaintance can be regarded as a basic sort of epistemic rela-tion between a subject and an instance of property: I am most directly acquainted with this instance of phenomenal greenness. (Chalmers, 2003 , pp. 248–250)


In talking of my knowing the phenomenal character of a given experience, I am talking of knowing a certain thing – something you too can know. Surely the natural and obvious view to take is that you and I know this thing by being acquainted with it. We experience phenomenal character, and thereby we know it. In so knowing it, we do not know a truth. We do not merely have abilities. We know a thing. (Tye, 2009 , p. 117)

Although ultimately unsatisfactory, for reasons I will provide shortly, the acquaintance approach to phenomenal knowledge exhibited by these contemporary philosophers has some commendable features that have moved discussion of the topic in the right direction. First, like the ability approach, it opens up the possibility of understanding phenomenal knowledge as non-propositional in nature. Knowing what it is like to have a particular kind of conscious experience does not in itself involve assent to a claim about such experiences. For example, in the case of Jackson’s Mary, she knows all of the facts already. When she comes to know what it is like to see red, there is no further proposition X such that Mary learns for the first time that X is the case. Instead, on the acquaintance model, she becomes acquainted with the experience of red in a way that only a person who actually experiences red can, and that is the integral feature of her coming to know what it is like to see red.

As another example, consider knowing the phenomenal character of what it is like to undergo a distinct kind of bodily sensation, such as experiencing an orgasm or a sneeze. I take it as an obvious point that knowing what it is like to have an orgasm does not consist in learning propositional assertions concerning the experience. Of course, there can also be propositional content involved in one’s understanding of a phenomenal state, such as ‘Wow, that’s even better than chocolate ice cream!’ but any such proposition is tangential to the actual experi-ence of the phenomenal state itself, which is what forms the content of knowing what a particular conscious experience is like. The acquaint-ance model acknowledges this non-propositional quality of phenom-enal knowledge, attempting to accommodate it through the notion of direct acquaintance with experience itself, unmediated by proposi-tional content. In this respect, the acquaintance model of phenomenal knowledge is a move in the right direction.

Moreover, the realization that phenomenal knowledge is non-propositional in nature helps us understand where Nagel and Jackson go wrong in drawing metaphysical conclusions from the phenomenal


character of experience. The experience involved in a conscious state does not in itself provide us with any factual knowledge about that state. An experience does not tell the experiencing agent that it is a physical event or a non-physical event. It is simply silent, so to speak, on the matter. To make any such judgment, the agent must engage in reflective, propositional thought about the experience, which requires the mediated application of fallible concepts (more on this in the next chapter). If the agent is operating under certain assumptions about the nature of physical reality, then she may conclude that her experience cannot be a physical event. Alternatively, however, she may have a different understanding of things that leads to the conclusion that the experience is a physical event. Either way, the agent is going beyond the experience itself in making any such judgment about it. Now, I am willing to grant Nagel and Jackson the point that, given our current understanding of physical reality, it is indeed puzzling how conscious experiences can come about through physical processes. However, it is just as likely (if not more likely) that this puzzlement is due to prob-lems in our understanding of physicality as it is that consciousness is a non-physical event. Conscious experiences in themselves, however mysterious they may seem, simply do not preclude the possibility that they are physical events. Perhaps that is just what physical reality is like, when known from the unique perspective of being a particular kind of physical event. As Bertrand Russell put it, ‘w e know nothing about the intrinsic quality of physical events except when these are mental events that we directly experience’ (1956, p. 153). In other words, phenom-enal experience may just be what it is like to be the particular kind of physical being we are, rather than a distinct non-physical property only contingently related to our embodied experience. Contra Nagel and Jackson, phenomenal experience does not embody non-physical facts that set itself apart from the scientifically knowable physical world. Phenomenal consciousness itself, as a kind of non-propositional content, leaves intact the possibility that it is a property or event in the natural world. By drawing attention to this non-propositional nature of phenomenal knowledge, the acquaintance model helps us see that phenomenal knowledge itself does not reveal any special metaphysical facts about its nature.

Nevertheless, the concept of acquaintance ultimately fails to capture the nature of phenomenal knowledge. The key problem is that acquaint-ance posits a relationship between a knowing subject and a distinct known object, such as when we say we are acquainted with a person or a place, but in cases of phenomenal knowledge, there is no identifiable


object of acquaintance that forms the content of one’s knowledge. In knowing the phenomenal character of conscious experience, there is no dichotomy between the one who is acquainted and that with which she is acquainted. Instead, the knower knows what her own phenom-enal states are like because she is in those states herself. She knows the phenomenal character of a conscious experience as the subject of expe-rience, rather than knowing the experience as an object with which she is acquainted.

C onsider again the example of an orgasm. When you experience an orgasm and thereby know what it is like to undergo the experience, you have this knowledge in virtue of being the one having the orgasm. At the moment of knowing what the experience is like, when the phenomenal character of the experience is known, you are in the state as an expe-riential subject rather than encountering it as object. In terms of your experience in itself, there is no separate orgasmic object with which you are acquainted. Rather, you yourself are in an orgasmic state such that you know the qualitative character of the state by being in it.

As another example, consider the experience of being depressed. When one consciously experiences depression, thereby knowing what it is like to be depressed, depression is not known as an object of acquaintance, in the same way that one knows another person or becomes acquainted with a particular place. Depression, rather, is known through the qualitative character of what one undergoes as an experiencing subject, without a dichotomous acquaintance relationship between oneself and one’s experience. Quite simply, then, the experi-ence of depression cannot be accurately characterized as an object of acquaintance, thereby illustrating how the acquaintance relationship between an epistemic subject and a known object fails to capture the unique epistemic character of phenomenal knowledge.

To be clear, this is not to say that we cannot also know phenomenal states as objects, via other epistemic means. In fact, as I will explain in the next chapter, we do need to conceptualize our experiences as objects in some manner in order for us to think about or cognitively reflect upon them. Clearly, we human beings have the capacity to obtain knowledge of our own mental states by forming and applying concepts, treating our experiences as objects to be known in propo-sitional terms. But, this kind of introspective self-knowledge extends beyond the experiential nature of phenomenal knowledge presently under consideration. Thus, our conscious experiences in themselves should not be thought of in this object-based manner, as an object of acquaintance or otherwise. We do not obtain phenomenal knowledge


as an object, but rather through being phenomenally conscious subjects ourselves.

The existential constitution model of phenomenal knowledge

In consideration of the problems outlined above with various attempts to understand phenomenal knowledge in terms of some other recog-nized kind of knowledge, it is clear that we must turn elsewhere for an accurate understanding of phenomenal knowledge. Our knowledge of our own conscious experiences involves knowing the qualitative phenomenal character of consciousness in a manner that cannot be characterized as knowing a fact, having an ability, or being acquainted with an object. Instead, we know the phenomenal character of what our conscious states are like by actually being in them, as conscious beings embodied in the world. This is a unique kind of knowledge that must be understood on its own terms rather than being reduced to another kind of knowledge.

To capture the nature of phenomenal knowledge, I propose the exis-tential constitution model of phenomenal knowledge, which asserts the following:

Phenomenal knowledge, understood as knowing the qualitative phenomenal character of a conscious state, consists in the actual embodied experience of the state in the world, in virtue of one’s being the existential subject of experience that undergoes the conscious state. The knowledge is constituted by the experience of the state itself.

We learned earlier that introspective knowledge of our own mental states is not something that appears to us as an object in an act of percep-tual observation. The above critique of existing models of phenomenal knowledge shows further that conscious experience is not an object that we come to know by getting acquainted with it, learning how to recognize it, or learning propositional facts about it. It is not an object but rather an experiential event. Accordingly, my proposal is that we think of our phenomenal knowledge of our own conscious experiences as a unique epistemic state that we acquire in virtue of the fact that we ourselves are composed of the known experiences as subjects. When I come to know what it is like to experience a color, a bodily sensation, an emotion, or any other phenomenally conscious state, this knowledge is


constituted by me actually being in the state and undergoing the expe-rience as a conscious being. I know the experience by being in it, rather than being related to it in a dichotomous relationship between myself and the experience.

I label this conceptualization of introspective knowledge the ‘exis-tential constitution’ model of phenomenal knowledge on the basis that our first-person knowledge of our own experiences is composed of the experiences themselves. We know our experiences through our being constituted by them, as conscious beings existing through time. One key feature that distinguishes this model from other attempts to conceptualize first-person knowledge of conscious experience is that it refrains from characterizing the kind of knowledge in question in terms of the knowing subject obtaining an epistemic relation with a distinguishable known object. In the case of phenomenal knowledge, the knower and the known are one and the same, as a conscious subject undergoing an experience of some sort or other. In this respect, this type of knowledge is quite distinct from other types of knowledge (propositional knowledge that X is the case, acquaintance knowledge with some entity or property X). It is a type of knowledge that does not match or fall under existing epistemological paradigms and thus must be understood according to its own unique nature if it is to be conceptualized accurately. Because it occurs through our being consti-tuted by the experiences we undergo as concretely existing conscious beings, it is apt to understand it in terms of existential constitution. The emphasis here is on our concrete embodied existence as conscious human beings. I choose the notion of existential constitution to point out the fact that phenomenal knowledge is fundamentally rooted in our existence as conscious beings, in virtue of its being existentially composed of our experiences themselves as we undergo them.

The epistemic status of phenomenal knowledge, as understood through the existential constitution model, is quite different from other types of knowledge. The epistemic properties of this kind of knowl-edge will be revealed further as we go along, both in what remains of this chapter and through comparison with other kinds of introspec-tive knowledge later in the book, but a few key properties are worth underlining here. As already discussed, phenomenal knowledge stands in contrast to standard propositional knowledge by having no identifi-able propositional content. Instances of phenomenal knowledge, under-stood as embodied phenomenal experiences themselves, are neither true nor false. Phenomenal knowledge is thereby infallible, by default, due to the simple fact that it contains no content that could even


possibly be wrong. By the same token, however, it does not offer any propositional content that reveals anything true either. Phenomenal knowledge cannot be wrong, but also cannot be right. It thereby stands altogether outside the usual epistemic norms of correctness (e.g., true belief, successful ability, and authentic acquaintance), as emphasized in the existential constitution model.

Phenomenal knowledge is an epistemic misfit in another respect as well, in terms of how it is acquired. Typically, knowledge is something that can be transferred from one person to another: items of propo-sitional knowledge, both a priori truths and empirically discovered facts, can be stated and explained in language, abilities can be taught through demonstration and training, and objects of acquaintance can be introduced by one person to another. Phenomenal knowledge, in contrast, can only be acquired through the experiential states of which it itself is composed. By its very nature, as knowledge of the phenom-enal character of what a conscious state is like, this kind of knowledge is obtained through embodied experience of the known state. The exis-tential constitution model, unlike the other models discussed earlier, accommodates this unique epistemic property by directly identifying phenomenal knowledge with conscious experience itself, unmediated by propositional content, ability, or relational acquaintance.

Although it offers a new approach to understanding phenomenal knowledge, the existential constitution model is not without precedent in philosophy. In particular, it shares an emphasis on our concrete embodied experience with existential and phenomenological perspec-tives in continental philosophy, as exhibited in the work of Husserl, Heidegger, Sartre, Merleau-Ponty, and others. As with the existential constitution model, these philosophers give priority to the first-person experiential aspect of consciousness, as subjectively experienced through our concrete embedded existence in the world. It is notable that this orientation has received renewed attention in contempo-rary work on mind and cognition, as exhibited in the work on 4EA (embodied, embedded, extended, enacted, affective) cognition by Andy Clark, Shaun Gallagher, Dan Zahavi, Richard Menary, Alva Noë, Michael Wheeler, Michelle Maiese, and others. Although there is signif-icant divergence concerning the details, these philosophers all empha-size the fact that adequate and comprehensive understanding of mind and cognition requires understanding ourselves as concrete embodied beings in the world, beings who experience themselves as conscious subjects fundamentally integrated into the world. As Maiese puts it, in articulation of what she calls ‘the Essential Embodiment Thesis’,


... human consciousness is necessarily instantiated in all the vital neurobiological systems, organs, and processes of our living bodies. The conscious mind of a creature like us is not a thing , but rather essentially a set of more or less fully realized spontaneous capacities in a desiring, affective, motile, suitably neurobiologically complex, egocentrically centered, spatiotemporally oriented, living organism that actively engages with the environment. If EE [Essential Embodiment] is correct, then our conscious lives are partially consti-tuted by , and not just instrumentally dependent upon , the facts of our human animal embodiment (2011, p. 235).

Obviously, this embodied approach extends beyond our focus on intro-spection here, but the shared emphasis on being constituted by our concrete existence as embodied beings in the world is a noteworthy point of explanatory convergence. Both the existential constitution model and the embodied approach to mind recognize the primacy of our subjective experience as beings in the world, such that our concrete existence is seen as constitutive of our consciousness. What the exis-tential constitution model adds to the mix is the recognition that our embodied conscious experience is itself constitutive of a kind of knowl-edge we have of ourselves, namely our phenomenal knowledge of what it is like to be ourselves, as embodied conscious beings in the world. In this respect, the existential constitution model of phenomenal knowl-edge can be seen as another positive step towards an accurate under-standing of ourselves, with emphasis on the unique knowledge we have of our own consciousness through our lived experience in the world.

Objections and clarifications

I find the understanding of phenomenal knowledge described above to be both intuitively appealing and explanatorily useful. It makes sense of the fact that we have qualitative, phenomenal experiences that seem ineffable through any other means besides actually having them, and helps us understand what makes this experiential kind of knowledge epistemically unique. However, I anticipate that others may not find this perspective as acceptable and explanatorily satisfying as I do. So, I will now address some potential objections to the idea that knowing what it is like provides a unique kind of introspective knowledge. In doing so, I will also offer some clarifications that will further elaborate upon my view of phenomenal knowledge, hopefully making it more appealing and understandable.


One possible concern is that phenomenal knowledge, as I have described it here, is not really a kind of knowledge at all. The standard view is that knowledge consists of justified true beliefs (or, more recently, some variant thereof, such as reliably-produced or virtuously-formed true beliefs). However, if my view of phenomenal knowledge is correct, such knowledge has no propositional content. Consequently, it could not possibly consist of true (or false) beliefs, and it is unclear as to what it would even mean for a phenomenal state in itself to be justified or unjustified. An experience itself is neither true nor false, and deter-mining whether or not it is justified is, if anything, an ethical issue rather than an epistemological one. So, in what sense can an experience itself be considered an instance of knowledge? If knowledge requires, by definition, true belief content, and phenomenal knowledge contains no such content itself, then it cannot be regarded properly as a kind of knowledge.

This issue is only definitional, however. We merely need to recog-nize that there are different senses of the word ‘knowledge’ as we commonly use it in English. If we strictly define knowledge as justified true belief, such that having propositional content is a necessary condi-tion of knowledge, then it does indeed exclude the kind of knowledge that I have portrayed. Yet, it seems perfectly natural for us to speak of knowing what phenomenal states are like in the sense I have described. If, for example, someone were to ask you, ‘Do you know what the taste of wasabi is like?’ amidst an everyday conversation about food, I antici-pate that you would say ‘yes’ or ‘no’ without any hesitation or confusion about the meaning of the word ‘know’ in this usage. People talk about knowing experiences, in the phenomenal, qualitative sense that I have been describing, all the time. So, on this basis, we can at least say that there is some sense in which knowing what experiential states are like is a kind of knowledge, and that it is this kind of knowledge that we have in mind with the concept of phenomenal knowledge. Knowing what an experience is like gives us a kind of knowledge. In experience, we know something of ourselves as embodied beings in the world. When we experience something for the first time, we thereby obtain a genuine kind of knowledge that we did not have before, prior to the experience. By simply broadening the scope of knowledge beyond the strict sense of justified true belief, we make room for saying that we know what phenomenal states are like.

Even if we do not count knowing what something is like as a proper instance of knowledge, however, it still serves a clear epistemic purpose. Phenomenally conscious states are not themselves beliefs, but they do


play integral roles in our belief systems. Whether we are operating from a foundationalist, coherentist, or some other epistemological frame-work, beliefs and experiences are intimately related. At a minimum, many of our beliefs are about our experiences. For instance, your beliefs regarding the tastes of foods, such as what foods taste similar or dissim-ilar, and what foods you like and dislike, are intimately tied to your phenomenal experiences with foods. From a broadly foundationalist perspective, we could add that your experiences provide justification for your beliefs, such as when you form the belief that some dish is spicy on the basis of a ‘hot’ sensation in your mouth. Of course, there is much more to say about the relationship between experiences and beliefs, but for the present point all we need to take note of is the fact that there is some epistemic relation or other between them. So, the phenomenal experience involved in the kinds of conscious states I have been discussing has an epistemic component. I prefer to characterize this as phenomenal knowledge, as referred to by the colloquial phrase ‘knowing what X is like’ and further understood through my existen-tial constitution model, but even if you are hesitant to regard this as a proper kind of knowledge, there is still room to regard phenomenal experience in epistemic terms, in that it underlies or relates to belief systems in some manner.

Another potential objection to my view is that phenomenal knowl-edge is not introspective in nature and thus has no proper place in an account of introspection. Introspection is often regarded as being self-reflective in some way. For instance, as we saw earlier, the most common understanding of introspection is that it involves ‘looking within’ oneself. All worries about the literal accuracy of this under-standing aside, one might plausibly think that introspection, whatever else it may be, is inherently self-directed: I’m not introspecting unless I’m explicitly directing my thoughts inward, towards myself. If so, then it may seem as though introspection excludes phenomenal experiences in themselves. A particular experience of red, for instance, need not be self-reflective in any way. When Mary sees a red fire truck for the first time, she may not be thinking about herself at all. So, how then could this possibly count as a kind of introspection?

There are a couple of things that need to be said here. First of all, I grant that there are numerous ways of introspecting that involve going beyond phenomenal experiences in some way. In fact, in the chapters that follow, I will describe several different modes of introspection that are clearly and explicitly more self-reflective than phenomenal experi-ence. Again, I think that introspection is a heterogeneous phenomenon,


which does not admit of any singular explanation. So, we should not be surprised if any one kind of introspection does not match up with all of our intuitions about what introspection involves. Admittedly, phenomenal states are not introspective in the sense that introspec-tion involves self-reflective thought. Nevertheless, there is still a sense in which we can legitimately regard phenomenal states as providing a kind of rudimentary introspective knowledge. Phenomenal experience, in the sense that it involves knowing what something is like, provides some intrinsic understanding of the experiencing organism’s own conscious states. Consider again Nagel’s statement that ‘an organism has conscious mental states if and only if there is something that it is like to be that organism – something it is like for the organism’ (1974, p. 519). The feature of being something it is like for the organism is crucial here. While conscious experience in itself is not self-reflective, it is intrinsi-cally self-regarding in virtue of this very basic, and easily overlooked, aspect. When an organism knows what an experience is like, it thereby automatically knows something about itself, namely, what its experi-ences are like in a concrete embodied manner. It knows itself in the most basic and intimate way possible, in virtue of being itself. So, in light of this feature of conscious experience, what might be described as the embodied first-person orientation of the existential constitution model of phenomenal knowledge, we can say that phenomenal experi-ences have an intrinsically introspective element to them. While Mary may not be explicitly entertaining thoughts about herself when she sees red for the first time, the experience itself gives her knowledge of herself. She knows what her own phenomenal experience of seeing red is like.

José Luis Bermúdez has made some related points concerning self-consciousness that will be helpful to consider here. According to Bermúdez, both perception and proprioception contain inherently self-specifying content (1998, pp. 103–162). Following J.J. Gibson’s theory of ecological optics, Bermúdez illustrates in detail how an organism tacitly obtains information about itself through the constantly shifting perceptual array it encounters while moving about in an environment. For instance, when you walk from one room to another, the perceptual changes you visually encounter inherently specify to you your loca-tion in relation to the objects around you. Similarly, the various modes of proprioception (detection of muscle positions, balance detection through the vestibular system in the inner ear, etc.) provide inherently self-specifying information to organisms that have them. For example, our limbs contain various nerve receptors that indicate their positions to us. I will take up proprioception again in the next chapter. The


important point here is that, by its very nature, proprioception provides information with self-specifying content, without explicit representa-tion of the self as an epistemic object. In recognition of this, Bermúdez regards proprioception, along with visual perception, as a primitive form of self-consciousness, detailing the various ways in which it provides an organism with information about itself (1998, pp. 131–162).

To be clear, I am not concerned with the informational content of these states here. Instead, the key point is that the informational content is intrinsically self-specifying in nature and that phenomenal states themselves have a similar self-specifying nature that results from being embodied and situated in the world. The experience itself provides immediate and intimate knowledge about the experiencing agent to that same agent, in a direct non-dichotomous and non-mediated manner. By its very nature, such a phenomenal state confers self-understanding in the most primitive manner possible to an experiencing subject. As discussed earlier, it is crucial to recognize that knowing what a phenom-enal experience is like intrinsically involves being constituted by the experience itself. The experiencing agent of a phenomenal state is itself made up of that state. So, in this respect, knowing what it is like is an intrinsically self-specifying form of knowledge. While this is clearly not the reflective sort of self-knowledge that we often have in mind when we speak of introspection, it is nonetheless a very basic and important kind of self-knowledge that is unique to the first-person orientation we hold toward our own minds. Phenomenal knowledge is, in this respect, a kind of introspection.

Putting skepticism of self-knowledge in its place

With the existential constitution model of phenomenal knowledge established, we can now make some important clarifications regarding the epistemic status of introspection. In particular, I want to address here the sceptical view that introspection is inherently fallible and untrustworthy. While there are some grounds for scepticism regarding certain domains of self-knowledge, as we will see later, broad scepticism towards introspection in general ignores the unique epistemic features of phenomenal knowledge recognized in the existential constitution model. I will confront this over-generalized scepticism towards intro-spection by discussing it in a place where it is particularly evident: Hilary Kornblith’s inverted Cartesian scepticism, such that objective knowledge of the external world is given priority over our unreliable introspective grasp of our own minds (1998; 2002).


With a clever inversion of Nagel’s take on consciousness, Kornblith’s ‘What is it like to be me?’ argues that we do not have privileged intro-spective knowledge of our own minds. Although this work brings some important deficits of introspection to light, Kornblith misconstrues the nature of knowing what one’s own conscious experiences are like, thus leading to an over-generalized scepticism about self-knowledge. As established by the overarching thesis of this book, there are importantly different types of introspection, which have correspondingly different epistemic qualities. So, the prospects for genuine self-knowledge through introspection will vary in accordance with the type of introspection in question. Kornblith, however, does not acknowledge any such qualifica-tion, and this leads to his over-stated scepticism.

To illustrate his perspective on introspection, Kornblith presents us with a man, Jack, who lacks an accurate understanding of his rela-tionships to others and of his own character traits and emotions. For instance, Kornblith states that

Jack is very defensive. He is quite insecure, and he believes, incor-rectly, that people are frequently talking down to him. ... Jack does not believe that he is defensive, insecure, or self-involved. Jack’s unreflective opinion of himself is quite different; it is terribly inac-curate. ... Jack’s understanding of his emotions is no more accurate than his understanding of his character. I have already said that Jack becomes angry and defensive when he is criticized in any way. But Jack sincerely believes that he does not become angry when criti-cized ... (1998, pp. 50–51)

Obviously, Jack is lacking in self-knowledge. Given this description, there is no question that his understanding of himself is quite wrong, and in multiple ways. From this basis, Kornblith goes on to ask, ‘What is it like to be Jack?’ and concludes that Jack himself would fail to know what it is like to be himself. Jack does not know what it is like to be Jack because his introspective capacities are badly askew.

This conclusion is overstated, however. Of course Jack is oblivious to and/or wrong about the facts concerning his emotions and character traits, but these facts are distinguishable from the phenomenal char-acter of what it is like to be Jack. Crucially, what it is like to be Jack is what it is like to be someone who experiences his particular emotions and character traits as he does, including the experience of unknow-ingly misunderstanding them! The phenomenal knowledge of what it is like to be Jack, understood as the kind of qualitative non-propositional


knowledge described by the existential constitution model, is distinct from propositional knowledge concerning the facts about Jack. It is the knowledge Jack has in virtue of his being the conscious being he is, irrespective of whatever false beliefs he may have constructed about his experiences. Knowing what it is like to be Jack is not the same as knowing what kind of person he is. Rather, it is knowing what the experience of such a person is like, from the subjective experiential standpoint of actually being that person. There is a crucial difference in kinds of introspection here, to which Kornblith does not give adequate attention.

Kornblith acknowledges that there is something that it is like to be Jack, but denies that such experience, in itself, constitutes any kind of self-knowledge. In order to have self-knowledge, Kornblith holds that one must be able to conceptualize such experience and form beliefs about it. To this effect, he states that ‘the experience of being oneself, without any belief about that experience, is no kind of knowledge at all’ (1998, p. 53). This perspective follows the presumption that knowledge is essentially propositional, ‘knowing that’. Knowledge, in this sense, requires assent to some propositional belief that constitutes the content of what one knows. If, however, one accepts this restricted definition of knowledge, then it would seem that one must also refrain from talk of ‘ knowing what it is like’, as this terminology inherently implies the excluded sense of knowledge. Crucially, Kornblith does not keep to this requirement. By discussing ‘knowing what it is like’ and simultaneously claiming that experience itself is not a type of knowledge, Kornblith is making a tacit but fundamental contradiction in his overall account. This creates confusion in the attempt to analyze one’s ability to ‘know what it is like’, as it literally entails that ‘what it is like’ is no type of knowledge at all! If knowledge requires propositional beliefs, then it simply does not make sense to discuss ‘knowing what it is like’.

One related point that needs attention here is Kornblith’s positive account of how one can know ‘what it is like’. Kornblith claims that if we are to know what it is like to be Jack, ‘We will need to engage in theory construction. We will need to try to figure out, given his behavior, including his verbal behaviour, what the mental life of such a person must be like’ (1998, p. 51). Underlying this proposal is the assumption that there is no qualitative difference between first-person and third-person knowledge. To this effect, Kornblith states that

Because the difference between a first-person perspective and a third-person perspective is a matter of causal proximity, it is comparable to


the difference between looking at a person from across a table and from across a street: closer is, frequently, better. (1998, p. 55)

So, according to Kornblith, third-person objective knowledge can poten-tially tell us (as long as it can get ‘close enough’) all there is to know about what it is like to be a given person, such as Jack. Such knowl-edge would not be different in kind from that which is obtained from a first-person perspective, by the person in question.

In contrast, we can see through the existential constitution model of phenomenal knowledge that the first-person subjective perspec-tive is different in kind from the third-person objective perspective emphasized by Kornblith. These are qualitatively different epistemic conditions. The first-person, subjective perspective is characterized by knowing what X is like, while the third-person, objective perspec-tive is characterized by knowing that X is Y, where Y is a propositional predicate attributed to X. The former kind of knowledge is intrinsic to the knowing subject, while the latter is essentially relational, involving the knowing subject’s conceptual relationship to the known object. By denying this qualitative distinction and assuming that it is conceivable to give a third-person account of first-person experience, Kornblith is simply ignoring the unique epistemic nature of knowing what phenom-enal states are like.

This is not to say that Kornblith’s scepticism about self-knowledge is entirely misguided, however. In his portrayal of scepticism, Kornblith mostly seems to be concerned with first-person knowledge of one’s own personality and character traits, such as friendliness, defensiveness, opti-mism, and so on. There is indeed significant empirical work indicating that people can be very wrong about these things (Nisbett & Wilson, 1977 ; Taylor & Brown, 1988 ; Wilson, 2002 ), and I think Kornblith is right to draw on these findings in order to understand the epistemology of introspection. I will discuss some of these empirical findings later, when we get to the conceptually mediated introspection of one’s own beliefs, desires, and personality traits. For now, however, the main point to grasp is that these abstract psychological concepts are qualitatively different from phenomenal, conscious experiences, and that these different kinds of mental states are known about through very different means. As I have explained above, knowing what a phenomenal state is like is something that is immediately and intrinsically embodied by the experiencing agent, in virtue of the fact that the agent herself consists of the state. However, an agent comes to understand her general person-ality and character traits over time, through complex belief-mediated


conceptualizations of her experiences and actions. This will be explained in more detail later, but for now it is sufficient to note that general beliefs about oneself arise through the mediated application of fallible concepts. So, unlike phenomenal self-knowledge, knowledge of one’s own general psychological traits, such as one’s beliefs and person-ality traits, are mediated and therefore prone to error and misconcep-tion. To the degree that people tend to make false conceptualizations of themselves, some scepticism about self-knowledge is warranted.

However, in regard to the domain of phenomenal knowledge of one’s own conscious states, scepticism is entirely inapplicable. When an agent has a phenomenally conscious experience, such as a perception of redness, a taste of chocolate, or an experience of sadness, the agent thereby knows what the experience is like, immediately and intrinsi-cally, in virtue of being in that conscious state. There is simply no room for error here, as the conscious experience itself constitutes the kind of knowledge in question. There is no room for accuracy either, though, it should be noted. In fact, given the non-propositional nature of this kind of knowledge, it does not even make sense to ask whether it is accurate or not. Phenomenal knowledge cannot be right or wrong, for the simple reason that it contains no propositional content. Its content is not a description of an experience, such that it can be true or false of the experience, but rather the experience itself. Now, of course, the agent can go wrong in identifying or characterizing the experience, such as when a person misidentifies a surprising tap on the back as a pain, but this takes us into other, conceptually mediated, domains of knowledge. The conscious experience itself is what it is, and is intrinsi-cally known as such.

Conclusion

In this chapter, I have presented and defended the idea that conscious experiences constitute a unique kind of phenomenal knowledge. We have knowledge of our own conscious experiences, in the sense of knowing what they are like, because we ourselves are composed of those experiences. We know our own conscious states by being them. This is the most immediate and primitive kind of self-knowledge possible. In characterizing this phenomenal knowledge through the existential constitution model, I have identified its distinctive epistemic properties that set it apart from other kinds of knowledge. The unique epistemic position obtained by being a conscious, experiencing agent needs to be recognized if we are to obtain a complete and accurate understanding


of introspection, and I hope I have helped to bring about this recogni-tion here.

However, it is also important to note the limits of this kind of self-knowledge. Phenomenal states in themselves have a very narrow and specific epistemic domain, consisting only of their intrinsic experiential qualities. They do not reveal anything about their metaphysical nature or causal structure, nor their conceptual relations to other states, either in the mind or out in the world. All of these things require further cognitive processes, whereby we can abstract from our experiences, conceptualize them, and construct propositional content about them. Identifying a phenomenal state as painful or pleasurable, reflecting upon an emotional experience and what it reveals about one’s character, thinking about one’s beliefs, and considering the course of one’s life are all prototypical instances of introspection, and they all go beyond the immediate experience of phenomenal states in one way or another. So, phenomenal states constitute only a small portion of the broad range of introspective states overall. In the following chapters, I will survey these further introspective domains, illustrating a variety of processes that extend beyond the intrinsic knowledge of experience to develop the more robust and self-reflective kinds of knowledge that are typically involved in introspection.

75

In the previous chapter, I explained an immediate kind of introspective knowledge that occurs through the intrinsic phenomenal quality of a conscious experience itself. However, many introspective states are medi-ated through various higher-level cognitive processes. The typical image of introspection involves an inward-focused, soul-searching person who ruminates, analyzes, and reflects upon her own mental life. This takes us beyond the immediate knowledge of a conscious phenomenal state toward more complicated reflective states that engage the extensive cognitive capacities of the human mind. In this chapter, I lay out a framework for understanding this more robust sense of introspection by drawing upon the fundamental processes involved in human cognition. The central idea is that we engage in higher-level introspection by utilizing the mind’s cognitive capacities to represent and think about our own minds. There is no need to posit any special cognitive processes or faculties that are specifically dedicated to introspection. The cognitive processes that we already know about (at least to some extent) can be utilized to understand what goes on when we introspect in a reflective manner.

The full spectrum of cognitive abilities is relevant here: memory, attention, imagination, learning, decision-making, problem-solving, conceptualization, language, and so on. There are many different kinds of cognition, enabling a broad range of cognitive capacities involved in human life. As Richard Menary puts it, ‘I t is quite natural to be pluralistic about cognitive processes and vehicles; as such, there is no single genuine “cognitive kind.”’ (2007, p. 13). Given the diversity of cognition itself, however, to fully survey all of the cognitive capacities available to humans and specify their roles in introspection would require a book in itself. So, instead of attempting a comprehensive account of all the cognitive processes relevant to introspection, my goal here is to offer a big-picture

5 Introspection through Cognition


framework that outlines how we can understand introspection through the general cognitive abilities of the human mind. In doing so, I will focus on three commonly recognized aspects of cognition that are particularly important in accounting for introspection: representation, conceptualiza-tion, and attention. These fundamental, and overlapping, capacities of the human mind are central ingredients in our capacity to understand the world we live in, but they also enable us to understand ourselves as well.

This is not to say that these attributes are together sufficient for cognitively mediated introspection, however. Memory, as another basic component of cognition, arguably plays an integral role in our intro-spective abilities as well. William Lyons’ ( 1986 ) account of introspec-tion emphasizes memory, for example, and does a fairly good job of bring out the role of memory in much of what we regard as introspec-tion, which is partly why I do not expand on it further here. Some other cognitive abilities relevant to introspection will be discussed later in subsequent chapters, including folk psychology/mindreading, language-enabled propositional thought, and social cognition, but as I mentioned, I will not attempt to cover them all.

In order to round things out and explore the boundaries of introspec-tion, however, I will include another more controversial aspect of cognition here, in addition to the three cognitive staples of representation, conceptu-alization, and attention. One of the more interesting proposals about the mind in recent years is the ‘extended mind’ hypothesis, which claims that cognitive processes extend beyond the human body, out into the external environment, including particularly the various artifacts we use to cognize, such as notes written down with pen and paper, computers and other information-processing devices, and perhaps even language itself (Clark & Chalmers, 1998 ; Menary, 2007 ; Clark, 2008 ). If cognition can extend out into our environmental interactions, integrating external artifacts into cognitive processes, then perhaps cognitively mediated introspective proc-esses can extend outside of our heads as well. Being in agreement with the idea that cognition can extend into our environmental interactions, I will explore the possibility of extended introspection later in this chapter. Nothing else I say is dependent upon the possibility, but I will suggest that the seemingly paradoxical idea that we can introspect by turning to our external environment is more plausible than it may at first seem.

Representation, metarepresentation, and misrepresentation

According to many philosophers of mind, the mind is essentially a representational device. For example, Fred Dretske proposes, as part

Introspection through Cognition 77

of his ‘Representational Thesis’ about the mind, that ‘All mental facts are representational facts’. (1995, p. xiii). In laying out the foundation of this representational thesis, Dretske offers a very clear and useful formulation of what constitutes representation in general:

The fundamental idea is that a system, S, represents a property, F, if and only if S has the function of indicating (providing informa-tion about) the F of a certain domain of objects. The way S performs its function (when it performs it) is by occupying different states s 1 , s 2, ... .s n corresponding to the different determinate values f 1 , f 2 , ... .f n , of F. (1995, p. 2)

So, for example, a thermometer is a representational system in that it has the function of indicating the temperature of certain objects within a specified range. The states of the thermometer provide infor-mation about certain states of objects. Applying this to minds, the basic idea is that mental processes serve the function of indicating the states of various objects or events. Vision, for instance, has the function of indicating certain properties (e.g., movement, size, shape, color) in an organism’s external environment. Visual states provide information about various states of affairs, and are thereby representational.

This representationalist understanding of mental states brings up many issues of controversy in philosophy, including important ques-tions about the nature of representational vehicles and content (e.g., Does it consist of computable symbolic concepts, as Jerry Fodor and others have maintained, or do non-conceptual embodied representations play essential roles in cognition?) and questions about the explanatory value of representations in our understanding of the mind (Does representa-tionalism explain everything about the mind, including consciousness, or might it instead lead to erroneous abstract conceptions of the mind that do not do justice to the concrete reality of embodied existence? Might the concept of representation be dispensed with altogether in our understanding of the mind?), but for our purposes here a relatively neutral understanding of representation will suffice, without strong commitments regarding these further issues. Nothing I say assumes that all aspects of cognition are representational in nature, leaving open the possibility of non-representational cognition. I also will avoid strong assumptions about the vehicles of representation, setting aside as much as possible whatever particular mechanisms lie beneath the hood, so to speak. All I will assume is that there are representational processes involved in cognition, irrespective of whatever forms and prominence


they may take in the mind more generally. Considering this, those who are critical of representationalist approaches to the mind may still find what I offer here acceptable, and perhaps even complementary to some of their concerns.

Dretske’s more robust approach to representation, and his application of this approach to introspection, is clear and useful, so I will be drawing upon it throughout this chapter. However, I want to make an important caveat in regard to a disagreement I have with Dretske’s overall account. Doing so will help carve out a more neutral reliance upon representa-tionalist explanation, as well as clarify how the territory I will cover in this chapter relates to the issues discussed in the previous chapter concerning phenomenal knowledge. According to Dretske, the repre-sentational understanding of mentality extends to the kinds of states we looked at in the previous chapter: phenomenal conscious experiences. From his perspective, the way to understand phenomenally conscious states, such as the experience of pains, orgasms, or colors, is in regard to their being informational indicators about things, events, or states of affairs. Dretske claims that this representational approach to phenom-enal experiences explains their phenomenal character. For instance, in discussing the simplified, hypothetical example of a ‘mono-representa-tional parasite’, which embodies the sole representational function of representing the specific temperature of 18ºC (indicative of a host for the parasite), Dretske states that:

If you know what it is to be 18ºC, you know how the host ‘feels’ to the parasite. You know what the parasite’s experience is like as it ‘senses’ the host. If knowing what it is like to be such a parasite is knowing how things seem to it, how it represents the objects it perceives, you do not have to be a parasite to know what it is like to be one. All you have to know is what temperature it is. If you know enough to know what it is to be at a temperature of 18º C, you know all there is to know about the quality of the parasite’s experience. (1995, p. 83)

So, according to Dretske, the representational facts about a mental state provide us with a complete understanding of what it is like to be in that state. As I see it, this claim is too strong. First of all, it is at least conceivable (and probably true) that there are representations that have no qualitative experience at all. Let us stick with the example of temper-ature and consider a thermometer. A thermometer represents tempera-tures, but it seems unlikely that there is anything it is qualitatively like


for the thermometer to be in one particular state or another. Of course, we can imagine that lower temperatures feel cold while higher temper-atures feel hot to the thermometer, but this is nothing more than a hypothetical projection of our own anthropocentric experiences. It does not rule out the possibility that there is simply nothing that it is like to be a thermometer, or Dretske’s parasite. Secondly, it is also conceivable (if not actually the case) that the same objective feature may seem qualitatively different to different representational systems. For instance, we can imagine that Dretske’s parasite may experience the temperature of its host in a number of different ways. Perhaps it feels like a pleasant tickle, or perhaps everything outside 18ºC feels painful in some respect while the 18ºC temperature of the host is simply an eased absence of pain. Through the consideration of such possibilities, we can see that the qualitative experiences of a representational system cannot be strictly identified with the features of objects or events that the representational system represents. Even though we may know what is represented, the qualitative experience of the representation is still undetermined by Dretske’s representationalism alone. So, Dretske’s claim that if we know what a representational system represents, then we know everything there is to know about the qualitative experiences of the system, including what they are like (in the Nagelian sense), is false. Representational facts do not tell us everything there is to know about mentality.

This is not to say that qualitative experiences cannot be under-stood as representations, nor that what representations represent has no bearing upon what they are like experientially. Pains, color experi-ences, and auditory sensations are representational states and can be plausibly understood within a representational framework, but that understanding alone does not tell us what it is like experientially to be an organism that is composed of such representational states. There is a fundamental difference between understanding the functional nature of a representation and actually experiencing a state that can be understood in representational terms. This distinction is crucial here and must be duly taken into account. Otherwise, it can be errone-ously conflated, leading to Dretske’s incorrect and dismissive portrayal of phenomenal experience solely in representational terms. In fact, this is partly why I am treating these two aspects of the mind in two separate chapters. The qualitative experience of being in a represen-tational state and the functional/representational nature of that state may ultimately boil down to the same thing metaphysically speaking (a token brain state, perhaps), but these are importantly different aspects


of that thing, with very different epistemic qualities. Qua experience, the state is known in the existential sense I explained in the previous chapter, while qua representation the state is known in terms of the informational content it contains and the function(s) that information performs. Importantly, and as I emphasized in the previous chapter, the former does not fit into the standard epistemic analyzes of whether or not it is veridical and/or reliable, as the kind of knowledge it embodies is not propositional, or informational, in nature. However, qua represen-tation, mental states may well fall under the standard epistemic evalu-ations of being true or false, reliable or defective, and so on. Often, and perhaps even usually, representational content and phenomenal character occur together in our mental lives, but this does not entail that they have all the same epistemic properties. In fact, it provides considerable clarity to treat them apart with regard to their distinct epistemic characters, which get confusingly conflated with each other. Qua phenomenal character, mental states are known in terms of what it is like for a conscious subject to have them, while qua representational content mental states are known in terms of the intentional content they represent. The purpose of this chapter is to address the latter, with regard to the roles representations can play in a human mind’s ability to know itself.

Now that we have this clarification out of the way, we can return to the nature of representation and how it pertains to introspection. We humans have a broad array of representational abilities, from our senses and the various representational capacities they involve to the concep-tual representation of abstract entities and relations, such as justice, knowledge, causality, minds, and the speed of light. These representa-tional abilities are not inherently introspective in nature. In fact, the majority of our representations are other-directed, representing the variety of things we encounter in our environment. However, we do represent aspects of ourselves as well, and in several different ways.

First of all, we represent states of our own body. Pains, for instance, represent certain instances of bodily damage. 1 Another mode of body representation is proprioception, our general sense of bodily location. There is some disagreement regarding the nature of proprioception, but it is clear that we internally represent the location and states of many of our body parts (Sacks, 1970 ; Ramachandran & Blakeslee, 1998 ; Damasio, 1999 ; Churchland, 2002 , pp. 59–126). To experience this for yourself, you can simply close your eyes and move your arms about. Through nervous tissues serving the function of proprioceptive repre-sentation, you will have information about the location and position


of your arms. In doing so, you are engaging in a very basic form of self-representation. Obviously, the representation of our bodily states to ourselves is vitally important to our embodied existence, underlying our capacities to move limbs or maintain balance. In fact, this basic capacity to represent the body is known to be a localized function of the primary cortex in the human brain, in an area known as the senso-rimotor cortex (although, interestingly, there appears to be some degree of plasticity within this representational capacity. See Ramachandran and Blakeslee ( 1998 ) for more on this.).

So, proprioception, along with other forms of representing one’s own body and location, is a foundational aspect of our capacity to know ourselves (Berm ú dez, 1998 ). However, some may argue that repre-senting one’s own body is not a kind of introspection. Introspection, as it is typically understood, pertains to mental states. A representa-tion of a bodily state, such as the position of one’s own arm, is itself a mental state, but that which it represents, arguably, is not. So, while bodily self-representation plays a fundamental role in self-knowledge broadly construed to include our embodied existence as flesh-and-blood animals, we need to move on to other kinds of representation, especially the representation of representations themselves, in order to provide a more satisfactory and comprehensive account of representational intro-spection, whereby we represent our own mental states to ourselves from our own first-person standpoint.

It will be useful here to make some conceptual distinctions between the different ways in which we can represent ourselves. Representations, broadly speaking, can be about anything, their crucial feature being that they are about something, whatever that something may be. So, we can classify different kinds of representations according to differ-ences in kinds of things that are represented (rather than, say, classi-fying them by their underlying mechanical structure). As discussed above, there are bodily representations, such as proprioception, pain, and sensual pleasures. The defining feature of such representations is that they represent a physical state of one’s own body. Alternatively, there are representations of mental states, such as beliefs and desires. (The distinction between mental and physical that I am drawing upon here is merely pragmatic and is not meant to suggest that these are fundamentally different kinds of things. Arguably, mental states are just physical states of a certain kind). As discussed earlier, a central feature of many mental states is that they are representational. So, consid-ering the representation of these representational states leads us to the idea of metarepesentation, or the capacity to represent representations


themselves as representations (Dretske, 1995 , p. 43; Sterelny, 1998 ; Sperber, 2000 ). In regard to the broad range of representational states, metarepresentation can itself be broken down into different kinds: representations of beliefs, desires, sensory representations, memories, inferences, linguistic tokens, and so on. Specific features of some of these kinds of representations will be discussed in detail later. For now, however, I will consider metarepresentation more broadly, as the kind of representation that involves the recursive representation of mental states as mental states.

Before turning to this, however, there is one further kind of represen-tation worth mentioning, while we are drawing distinctions: the repre-sentation of oneself as a unified agent, person, and/or self. In addition to representing various particular aspects and states of ourselves, we also represent ourselves as a whole, in terms of thinking, experiencing, and acting beings. As in the case of metarepresentation, there are argu-ably a number of different subdivisions that fall under this category, such as representing oneself as a biological organism, a rational agent, a disembodied soul, or one of numerous kinds of social entities (e.g., parent, teacher, friend, voter). As we will see in the next section, these different modes of representing oneself as a particular kind of being utilize different kinds of self-concepts, which may overlap in complex and interesting ways.

First, however, let us return to metarepresentation and the central role it plays in introspection. In essence, metarepresentation is the representation of a representation as a representation. So, for instance, if I think to myself that I am perceiving a computer screen right now, a component of that thought is the representation of my perception of the computer screen. In having this thought, I am representing my perceptual representation of the computer screen as a representation, thereby producing a metarepresentation. This capacity to represent representations is a fundamental aspect of introspection. Insofar as introspection involves the representation of one’s own mental states (which, as we have already seen, can often be understood as representa-tions themselves), it inherently involves metarepresentation.

This more or less echoes Dretske’s account of introspection. He states that:

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, metarepresentational .

... Introspective knowledge, being a form of representation, is, therefore, a metarepresentation – a representation of something (a thought, an experience) as a thought or an experience or (more specifically) a thought about this or an experience of that. If E is an experience (sensory representation) of blue, then introspective knowledge of this experience is a conceptual representation of it as an experience of blue (or of color). (1995, pp. 43–44)

So, when I introspect, say, that I have a pain in my left foot, I do so by representing the pain as a pain, through the mediation of a represen-tational concept that characterizes the experience of pain as a pain. Along these lines, Dretske characterizes introspection as a species of ‘displaced perception’, in regard to the fact that we do not know our own mental states by perceiving them, but rather by conceptually representing our perceptions (1995, pp. 41 ff.). For example, I know that I weigh 158 pounds not by directly perceiving my weight but rather by perceiving a scale and having a conceptual understanding of what scales represent, such that my perception of the scale is displaced into a ‘perception’ of my weight via my conceptual understanding of the information presented by the scale. Analogously, I introspectively know that I see a computer screen not by perceiving my perception of the computer screen but rather by seeing the computer screen and having the conceptual capacity to characterize this perception as an instance of seeing. I represent my perceptual state to myself by displacing it into a conceptual framework that represents it as a representational percep-tual state.

According to Dretske, such metarepresentational processes are simply all there is to introspection. As previously explained, I regard intro-spection as a heterogeneous collection of importantly different kinds of processes, so I do not follow Dretske in this exclusionary single-faceted account of introspection. For instance, in regard to knowledge of pain, I not only know that I am in pain through the metarepresentational process explained above, I also experientially know what the pain is like, in the sense explained in the previous chapter. However, I emphatically agree with Dretske on the point that we do obtain introspective knowl-edge in this manner, and I think that this point is crucial to a complete understanding of introspection. We do have the capacity to repre-sent our representations, and this enables the ability to represent, and thereby know, aspects of ourselves. Moreover, this metarepresentational


capacity, through the application of concepts and conceptual struc-tures to ourselves, provides the foundation for more abstract domains of introspection, taking us beyond the relatively basic knowledge of our own immediate experiences (e.g., pains, visual perception) to the highly conceptual knowledge of our character traits and social personas.

I will turn to the role that concepts play in introspection momentarily, but first there is an important epistemic point about representation that needs to be addressed. Wherever there is representation, there is also the possibility of misrepresentation. Representations, whatever they may be and however they may come about, are mediated through processes that can have errors of various sorts. Representations can misrepresent. Experiences and information can be misconstrued or misinterpreted, and concepts can be misunderstood or applied incorrectly. Taking note of these possibilities, coupled with our understanding of introspection in terms of metarepresentation, we can justifiably infer that introspec-tive metarepresentation is fallible. Just as we can misrepresent a tree swaying in the wind at night as a person stumbling down the street, we can misrepresent a surprising tap on the back as a pain or confuse an emotional experience of fear as anger.

This runs counter to the common intuition that we cannot possibly be wrong about our own mental states, especially in regard to occurent states like pains and perceptual experiences. It may seem as though I cannot be wrong about being in pain, or about having a red visual experience, but insofar as introspection involves mediated representa-tion, this seeming is wrong. We can fail to accurately identify a sensa-tion, misrepresenting it as a pain. We can incorrectly conceptualize a red experience as an orange one, and then continue on to believe that we saw something orange when in fact the color experienced was red. Representation makes misrepresentation possible, so it is possible for us to misrepresent our representations when engaging in metarepresenta-tional introspection.

Conceptualizing our own mental states

Introspection through metarepresentation is typically enabled through concepts of representational states. For instance, in order to represent the pain in my left foot as a pain, I must first have some concept of pain or other (e.g., recognitional, theoretical), which I apply to the sensa-tion in representing it as a pain. Without a representational concept to apply to the experience of pain, I simply would be unable to make the second-order step of constructing a representation of my pain as a pain.


Once again, it will be instructive to look at what Dretske says about the matter:

Not having a concept of representation does not prevent one from representing things, but it certainly prevents one from believing (hence, knowing) that one is doing it. Until a child understands representation, it cannot conceptually represent (hence, cannot believe; hence cannot know) that anything – including itself – is representing (and, therefore, possibly misrepresenting) something as F. What prevents small children and animals from introspecting is not the lack of a mysterious power that adults have to look inward at their own representations. They already have all the information they need, and they needn’t look inward to get it. What they lack is the power to give conceptual embodiment to what they are getting information about. (1995, pp. 59–60)

Setting aside the issue of whether or not children and animals actu-ally have the conceptual abilities in question (contra Dretske, there may be conceptual resources beyond the overt language-driven concepts of human adults), the key point here is that representing something as X requires having a concept of X. This is most clearly illustrated in cases of culturally dependent concepts (concepts that derive from human social processes, rather than possibly innate cognitive structures). For instance, consider the representation of an automobile as an automo-bile. In order to see an automobile and recognize it as such, one must have not only sensory experiences of the automobile but also a concept of what an automobile is and the capacity to identify something that falls within that conceptual category. This is something that you and I do on a regular basis (presuming that you are a typical adult living in a cultural situation that is similar to mine, of course), but someone or something else that lacks the concept of an automobile would be inca-pable of doing so (an animal, or an adult human that has been entirely isolated from the post-industrial world, for instance).

With this point in mind, let us return to Dretske’s case of an organism that lacks any concept of representation. Such an organism simply would not be able to conceptualize its representations as representa-tions, and so could not engage in the kind of introspection that we have been considering. This point in itself may seem to be so obvious that it need not even be established, but it helps us identify what it is that enables introspection through metarepresentation: concepts. Concepts, whatever they may be, are the vehicles through which we


organize our experiences and gather information into coherent, under-standable bundles. Without concepts to structure them, our experiences would only amount to a ‘booming, buzzing confusion’, to borrow from William James’s description of an infant’s unstructured experience of reality (1890). This applies to our understanding of our own minds no less than it does to our understanding of the external world. Without concepts of representation, we can know the intrinsic phenomenal qualities of our experiences but we cannot form second-order repre-sentations of the experiences, such that they may become the objects of thought and understanding. So, considering the fundamental roles played by concepts in human cognition, a look at concepts will give us some insight into the nature of introspection.

Before jumping into the role of concepts in introspection, however, I want to address a potential worry that could creep up here. Pointing toward concepts to explain introspection may seem somewhat vacuous, especially to those who have explored the quagmire of views concerning the nature of concepts. There are a number of very different perspectives on what concepts are, from Jerry Fodor’s claim that concepts are atom-istic mental particulars, to the more common, but still controversial, view that concepts are structured and interconnected representational complexes, not to mention the view that they are non-mental abstract objects, such as Platonic forms (Fodor, 1998 ; Margolis & Laurence, 1999 ). So, without a clearer idea of what concepts are, appealing to concepts to explain introspection may not seem like a very progressive option. I do not have a definitive view of concepts to offer, so I will refrain from pressing any particular account of concepts here, with the exception of assuming that they are broadly mental in nature. Fortunately, however, we can still appeal to concepts without necessarily stating precisely what they are. Whatever they may be, concepts play central roles in human cognition and it is this functional level with which we are primarily concerned. Accordingly, we can discuss concepts in terms of what they do, rather than in terms of what they are. Of course, some views define what concepts are in terms of what they do, and this seems to me to be a fairly plausible route to take in understanding the nature of concepts, but for our purposes we can discuss concepts on this level without excluding the possibility that the nature of concepts includes more than their func-tional roles. Whatever they may be, concepts enable us to engage in an amazing variety of cognitive processes, and it is this enabling capacity of concepts that I will address and appeal to in what follows.

With that said, let us get back to the heart of the matter. One central feature of concepts is that they allow us to engage in abstract thought,


such that we can think about things beyond our immediate experience. Consider, for example, the concept of a cause. We utilize the notion of causation to understand a wide variety of things on a daily basis, from the satiation of hunger by eating food, to the pressing of a gas pedal to make an automobile move forward. Yet, as Hume famously pointed out, we never observe causes, not even within ourselves. The concept of a cause does not come from, but rather extends beyond, our immediate experience, and thereby brings us to a new level of cognitive ability. It enables us to make sense of our experiences, by discerning abstract rela-tions and patterns among them.

The very same sort of thing occurs when we employ concepts of mental representations to understand our own minds (and, of course, the minds of others). When we have the capacity to conceptually repre-sent our mental states as mental states, such as when we conceptualize an auditory sensation as an instance of hearing or a feeling of hunger as a desire, we obtain a new level of cognition that enables us to understand our own mental lives. The abstract cognitive abilities enabled through concepts are especially important in this domain because, as I have already argued, mental states are not observable objects. Contrary to the perceptual model of introspection, we do not observe pains, percep-tions, desires, beliefs, decisions, and so on. Instead, we form represen-tational concepts of them, such that they can be cognitively treated as objects. So, for instance, in believing that I believe I am currently sitting in front of a computer monitor, I utilize the concept of belief to abstractly represent my belief as an object that I can think about. Such metarepresentation through the employment of concepts is what enables us to engage in higher-level introspection, where we cognitively reflect upon our own mental lives.

Abstract, unobservable objects are typically regarded as theoretical entities, and this leads us to an important point about the role of concepts in introspection: The concepts of unobservable mental enti-ties that we use in introspection are implicated in a broadly theoret-ical understanding of mentality. In other words, the utilization of one of these concepts involves some abstract, theoretical understanding of how the things that fall under the concept work. For instance, the concept of desire is typically connected to the abstract understanding that an entity with a desire for X will try to achieve X, all things consid-ered. So, when I conceptualize a sensation as a desire, I do so with the understanding that the object of desire is something that I want to obtain. Similar explanatory frameworks surround our other concepts of mental phenomena, such as beliefs, pains, emotions, and decisions.


This theoretical understanding of mental phenomena forms a loosely knit body of knowledge, known as ‘folk psychology’ in the philosophy of mind and ‘theory of mind’ in the cognitive sciences. Folk psychology/theory of mind and the role it plays in self-understanding (belief/desire explanations, in particular) will be the focus of the next chapter, so for now I will only make a few general points regarding its conceptual/theoretical basis.

First of all, some clarification regarding the theoretical status of mental concepts is due here. In making this connection between concepts and theories, I am not necessarily suggesting that concepts themselves should be understood in terms of theories. (For a review of this idea, and a comparison with other views on the nature of concepts, see Margolis & Laurence, 1999 .) Concepts may be understood in this manner, or they may be understood in some other way (or perhaps most plausibly, different kinds of concepts may require different kinds of explanations). As I said earlier, I am not making any claims here about what concepts are. Instead, what I am claiming is that concepts, whatever they may be in themselves, are intrinsically tied to theoretical understandings of the phenomena they encapsulate. Consider a concept of pain, for instance. Concepts of pain could be cognitively manifested in a number of different ways. But whatever it is, a concept of pain is clearly tied to some theoretical understanding of what pains themselves are, and the roles they play in behavior, such that feelings of pain can be understood as instances of pain and individually distinguished from other kinds of feelings in virtue of that theoretical understanding.

Secondly, in characterizing our understanding of mental concepts in terms of theoretical structures, I am not necessarily suggesting that these theoretical structures are explicit, formalized theories, such as we see in contemporary science. Some have drawn parallels between the ordinary conceptual frameworks of humans and the practices of scien-tists (see Gopnik & Meltzoff, 1997 , for instance), and I think that this is a worthwhile idea to explore, but I do not want to commit myself to a strong version of this claim here. Instead, in saying that our concepts of mental phenomena involve theories, I only mean to suggest that there are abstract explanatory structures that we utilize to interpret and under-stand mental phenomena and their relations to behavior. This clearly has some overlap with the roles of theories in science. Scientists posit and use theories to explain and predict phenomena as well, and it is this general feature that is behind my use of the concept of a theory. However, scientists’ particular methods in forming and using theories (developing explicit equations to codify a theory, for instance) are arguably quite


different than what commonly goes on in a person’s mind while intro-specting (or while thinking about the mentality of others). So, when a person applies the concept of desire to her own feelings of hunger, and thereby understands her experience within a particular theoretical framework regarding what desires are and do, she is not necessarily utilizing an explicit theory of desire, on a par with an explicitly formu-lated scientific theory, but she is using an abstract explanatory frame-work of some sort to understand that aspect of her own psychology.

Finally, although the labels ‘folk psychology’ and ‘theory of mind’ suggest a single, cohesive body of understanding, our theoretical under-standing of mental phenomena may involve multiple, and perhaps even conflicting, conceptual frameworks. In regard to the prominent belief/desire explanatory framework that typically falls under these labels, it is plausible to think of folk psychology/theory of mind as a somewhat unified theory. Again, this will be the topic of the following chapter, where I will further analyze the role this framework serves in self-understanding, but in regard to the broader domain of mental concepts in general, our overall theoretical understanding is much more amorphous. In fact, there are different ways of conceptualizing at least some mental phenomena that can actually be at odds with one another. This is perhaps most apparent in regard to the self. The concept of a self is notoriously difficult to nail down to a single acceptable characteriza-tion, as those familiar with the literature on the topic will recognize. Consider, for example, the conflicting intuitions at work in Bernard Williams’ well-known article ‘The Self and the Future’ ( 1970 ). Williams poses a variety of thought experiments which conjure up opposing views on the essential nature of one’s self and how it persists through time. In some cases, it seems as though one’s bodily continuity is most important to preserving the self, while in others one’s self seems most closely associated with psychological characteristics, irrespective of bodily changes. There is much to say on this topic, with the nature of the self being a central issue throughout the history of philosophy to the present day, but for our purposes the point to grasp is that there are different conceptions of what a self is, and that these conceptions can come apart in certain contexts. We conceptualize the self in a variety of different ways: as a body, an experiential subject, a rational agent, and a social entity, which itself can be further broken down into a variety of conceptions: citizen, employee, spouse, and others. These different concepts of the self often overlap, but they can diverge from one another, in which case we can see how they implicate different theoretical frameworks for understanding the nature of the self.


In summary, the primary point is that concepts of mental phenomena, along with their associated theoretical frameworks, provide the cognitive basis for higher-level introspection. The use of concepts to understand our own minds is not fundamentally different from our ordinary conceptu-alization of the world around us. There are concepts that play special roles in introspection (e.g., belief, desire, emotion, self), and there may be some self-oriented biases in the ways concepts are applied to oneself, but the same basic cognitive processes are at work in both one’s conceptual under-standing of the external world and oneself. The crux of this cognitive ability is the employment of abstract conceptual structures that allow us to make sense of our experiences. By applying concepts to our own mental phenomena, we bootstrap ourselves out of our immediate experiences into an abstract realm of concepts and theoretical structures, whereby we can understand and reflect upon what goes on in our own minds.

Attention and introspection

No account of introspection could be complete without addressing the role of attention. Attention involves the processing and organization of perceptual information, actively directing our cognitive resources toward particular phenomena. Attention also guides our more abstract representational abilities, involving the application of conceptual frameworks to our occurent thoughts and experiences. With these roles comes the capacity to direct our cognitive resources toward our own mind. For example, in order to become introspectively aware of a visual experience as an experience that you are having (suppose you are introspectively attending to your experience of reading this text, for instance), you must attend to your experience and through that atten-tion develop a conceptual representation of your experience as a visual experience of some sort. Without the capacity to attend to a specific phenomenon like this, no aspect of your experience could become an object of self-reflective understanding. So, clearly, attention plays a significant role in our capacity to introspect.

But what is attention? As is the case with many psychological phenomena, there are multiple modes and levels of understanding the nature of attention. Consider the following diverse aspects of atten-tion: 1) the first-person experience of attending to things, 2) ordinary behavioral attributions of attention (for example, observing that one student is actively attending to a lecture, while another is attending to Facebook), 3) the cognitive understanding of the functions involved in and performed by attentive processes, and 4) the neurological


understanding of the brain mechanisms involved in attention. I will focus primarily on the first of these aspects of attention, but the others must be given their proper recognition. In fact, in some cases, the third and fourth aspects may point out problems for the first two. This is an interesting issue in itself, so allow me to digress for a moment before addressing the role of attention in introspection.

We have an ordinary, common-sense understanding of attention that pervades our everyday discourse about the activity of our own minds and the minds of others. As William James put it, much as he did with introspection, ‘Every one knows what attention is. It is the taking possession by the mind, in clear and vivid form, of one out of what seem several simultaneously possible objects or trains of thought. Focalization, concentration of consciousness are its essence’. (1890, p. 403). This seems intuitively compelling, as far as it goes. In our expe-rience, our minds select some objects or thoughts to the exclusion of others, and through this selective process we have what seems to be a single, focused domain of conscious experience that we guide from one object to another. The metaphor of a spotlight is apt here, suggesting the idea that we direct our attention upon particular objects in the same way that shining a light upon objects highlights them, making them a focal point (see Baars, 1997 , pp. 95–111, for an application of this metaphor in understanding attention). But on closer analysis, it is not at all clear what is going on when we do this. For instance, there is evidence of multiple kinds of attention within the human mind, with distinctly different modes of operation (LaBerge, 1995 ; Hardcastle, 1998 ; Pashler, 1998 ; Fuster, 2002 ; Itti et al., 2005 ). This runs counter to our ordinary understanding of attention as a single, isolated focus upon some phenomenon. Arguably, there are even attention processes that we are not aware of, operating behind the scenes, so to speak, in the background construction of our experiences and thoughts. Considering such things, our ordinary understanding of attention might not be adequate, or even on target, when it comes to understanding the actual cognitive processes involved in attention.

This concern should seem familiar by now, as it is similar to the more general point that I have been making about the diversity of introspec-tion. Although we commonly understand introspection in perceptual terms (due to the utility of perceptual metaphors in getting a handle on our understanding of ourselves), this is not an adequate account of what really goes on when we introspect. Similarly, one may argue that our common-sense, metaphor-driven understanding of attention does not accurately map onto what is really going on in our minds.


There are, however, important differences between attention and introspection. As explained earlier, there is no identifiable perceptual mechanism whereby we literally perceive objects in our own minds, nor are there identifiable mental entities to serve as the objects of such perception. Regarding attention, on the other hand, things are quite different. There are identifiable areas in the brain that can be associ-ated with attention, and there are data that illustrate a variety of atten-tive processes in human cognition (ibid.). Moreover, since attention is a second-order cognitive process, there are no problems regarding the determination of what objects it takes as input (which, recall, was a devastating problem for the perceptual account of introspection). Rather than gathering information from determinable objects, as is the case with perception, attention is involved in processing already existing information in one’s cognitive economy. So, considering the fact that these aspects of attention are much more readily identifiable than they are in the case of supposed mental perception, appeal to attention as an explanatorily useful cognitive mechanism has at least some prima facie plausibility.

Unfortunately, I cannot pursue these issues further here. (It would take another book, Rethinking Attention , to do so!) So, suffice it to say that we can plausibly discuss attention as a legitimate category of human cogni-tion, while still noting that we need to be careful that we do not take our common-sense understanding of it too literally. The actual proc-esses involved in attention are probably somewhat different from what they seem to be to our first-person conscious experience of attention and our ordinary understanding of it, but this need not prohibit us from discussing attention without a detailed account of its nature, so long as we keep in mind the tenuous status of our understanding. Unlike the very dubious phenomenon of mental perception involved in a literal account of introspection, attention is a real identifiable phenomenon that we can both experience and empirically investigate.

With that said, let us return again to the role of attention in intro-spection, being mindful of the empirically revisable nature of what is discussed. It is worth noting that there are different aspects of attention that pertain to different kinds of introspection. To illustrate this, let us consider the concrete example of attending to a pain. First of all, in having a pain, one can attend to the feeling of pain itself. Arguably, this kind of attention plays a role in bringing about the phenomenal knowl-edge of consciousness discussed earlier, but I am not going to enter into the difficult issue of how attention and consciousness relate to one another here. (For accounts that relate attention and consciousness


from empirical standpoints, see Crick, 1994 ; Baars, 1997 .) Suffice it to say that, whatever consciousness may be, paying attention to a state can play a role in making one phenomenally conscious of that state. A passage drawn from the perspective of Buddhist meditation illustrates the idea:

If a sense object exercises a stimulus that is sufficiently strong, atten-tion is roused in its basic form as an initial ‘taking notice’ of the object, as the first ‘turning towards’ it. Because of this, conscious-ness breaks through the dark stream of subconsciousness (- a func-tion that, according to Abhiddhanna (Buddhist psychology), is performed innumerable times during each second of waking life). This function of germinal mindfulness, or initial attention, is still a rather primitive process, but it is of decisive importance, being the first emergence of consciousness from its unconscious subsoil. (Thera, 1962 , p. 24)

In addition to attending to an experience itself, one can pay attention to that which the experience represents (a scratched leg, for example in the case of a pain). With this mode of attention, the representational capacities discussed above come into play. Third, one can attend to the pain itself as a representation. As discussed earlier, this draws upon the metarepresentational capacity to conceptualize one’s experience, by placing a particular experience into a conceptual category that explains or makes sense of the experience. Finally, one can also attend to oneself as the experiential agent that is having the pain, perhaps attending to the situation that one is in and the possible courses of action to be taken in response (removing one’s leg from a thorny bush, for example).

All of these different kinds of attention are relevant to various kinds of introspection. A great deal more could be said about this, mapping out the various connections between these mental processes, and perhaps my drawing attention to the issues will enable some progress to be made on the topic. For our purposes, however, the interesting thing to note about these attention capacities is the general point that they can serve multiple cognitive purposes. Attention (whether it be the basic ‘bottom up’ processes that filter and organize our basic conscious experience or the higher-level processes through which our abstract conceptual abilities are employed) can be directed toward either internal or external stimuli (or perhaps simultaneously to both), depending upon what aspects of the attended object are being attended to. In attending to a visual experience, for example, I can either attend


to my experience of the visual field, reflecting upon the fact that I am having the experience, or I can turn my attention outward toward that which I am seeing, thinking about the various environmental factors involved in the production of the experience. Attention is an amaz-ingly plastic capacity. Whether it involves multiple disparate mecha-nisms or the redirection of a particular mechanism (or some of both), the human capacity to attend to a multiplicity of phenomena dramati-cally expands the range of things that we can think about, including a variety of aspects of ourselves. Coupled with the ability to conceptually represent aspects of our mental lives, our ability to attend to various sorts of phenomena provides a foundation for introspective cognition. This is the primary point that I want to make about the role of attention in introspection. The very capacity to attend to various phenomena, however it is actually manifested in the mind, enables us to direct our cognitive resources recursively, back upon themselves. It is crucial to note that this introspective capacity originates from the general atten-tion capacities of the human mind and does not require appeal to any special capacity that is dedicated to attending to ourselves. The ability to attend to ourselves simply falls out of the ability to attend to phenomena in general.

As with representation and conceptualization, it is important to note that the use of attention to engage in introspection opens the door for epistemic error. Attention helps us focus upon aspects of our own minds, thereby making them potential objects of self-knowledge, but we can also fail to attend to aspects of our own minds (such as failing to iden-tify the presence of a particular emotion) and perhaps even purpose-fully turn our attention away from aspects of ourselves that would otherwise be readily noticeable. So, the selective nature of attention is a double-edged sword, such that it both enables and inhibits our capacity to know our own minds. The fact that we can fail to attend to aspects of our own mind reveals that, contrary to some accounts, the mind is not entirely transparent and accessible to itself. Acknowledgement of this point is important to an accurate understanding of the epistemology of introspection.

Introspection and the extended mind

Introspection, as we have seen, is typically regarded as inward reflection, such that the contents of one’s mind are known in some first-person manner or other. It may seem, then, that introspection is essentially internal, operating purely within the confines of one’s own skull. But


this is not necessarily the case. Conjoining the cognitive account of introspection I have presented here with the extended mind hypoth-esis that cognition reaches out into our interaction with the external environment (e.g., Clark & Chalmers, 1998 ; Menary, 2007 ; Clark, 2008 ), we encounter the possibility of extended introspection: introspectively engaged cognitive processes that are not contained in the head but rather are embodied through external events. Just as a handwritten or electronically texted shopping list may serve as a cognitive constituent in the process of stocking one’s kitchen, so too may external cognitive vehicles play a role in the introspective process of knowing one’s own mind.

What form might extended introspection take? For starters, we can consider the shopping list example itself as a possibility. Typically, a shopping list facilitates externally directed processes, but it might be used introspectively as well. Suppose, for example, you have ‘chocolate donuts’ written on your shopping list, next to other various breakfast items you need to pick up at the store. As you shop in consultation with your list, you pause to rethink the chocolate donut choice, thinking to yourself that it is not an item you should buy after all. The chocolate donut listing next to other breakfast items also prompts you to recall that you made your list before eating breakfast, with your irrational desire to eat chocolate donuts fully intact while making the list. This realization prompts you to further rethink your habit of writing your list before breakfast, thereby forming the future intention to make your list after breakfast instead, enabling more reasonable choices, less subject to the whims of your early morning desires. The point is that the external-ized content on the shopping list might itself be a constituent in the introspective process of reflecting upon your desires and habits, thereby enabling you to know something about yourself that may have gone unrecognized otherwise. In this case, the words on the list are them-selves part of the introspective process. This might not resonate with the profound soul-searching insights that some associate with intro-spection, but it is arguably an instance of introspection nonetheless, however mundane it may be.

In describing his own take on the extended mind approach, which he calls ‘cognitive integration’, Richard Menary states that

the co-ordination of bodily processes of the organism with salient features of the environment, often created or maintained by the organism, allow it to perform cognitive functions that it otherwise would be unable to; or allows it to perform functions in a way that


is distinctly different and is an improvement upon the way that the organism performs those function via bodily processes alone. (2007, p. 3)

Applying the point to introspection, this presents us with the possibility that introspection through external vehicles might not only count as a genuine kind of cognition but may also be an improvement upon the use of cognitive capacities contained in the head alone. The shopping list example above could be an illustration of this, but to press the point further, let us consider another example.

One notable kind of introspection is the process of autobiographical writing, either formally, as in the case of a book, or informally, as in the case of a personal diary. In such cases, the externalized language recorded on the page or screen enables the author to engage in intro-spective reflection upon herself and her life, providing a representa-tional medium that she can attend to in furthering her introspective processes beyond the content immediately available within. This kind of introspection, aided through the use of written language as an external cognitive artifact, can offer some advantages that internal representa-tional vehicles alone are unable or unlikely to achieve. The internal stream of conscious thoughts and experiences of a human being is a continuously changing flux of content, subject to protean shifts of attention and the fallible results of memory, whereas written language enables a kind of permanence and rigidity that can transcend these features of experience. Compare, for example, a person’s conscious thoughts and memories about her childhood with her diary she wrote as a child. If she is introspectively reflecting upon her childhood, perhaps in attempt to better know herself through the formative events of her past, the diary could quite plausibly provide more trustworthy content than her internal recall. Memory is notoriously fallible, subject to biased and unreliable reconstruction from one’s current viewpoint, while a diary is a direct record of one’s thoughts and experiences as they were conceived at the time of writing. We can also just plain forget things, but a diary’s content is faithfully preserved exactly as long as it is. So, the introspecting agent might discover something about herself in her childhood diary that otherwise would have been forgotten or misconstrued by her adult self. In this way, written language, as an external component of cognition, can serve as an introspectively utiliz-able representation. This is not to say that reading one’s childhood diary is an infallible path to knowledge; externally facilitated introspection is fallible, too, of course, but it can embody features that set it apart from


other kinds of introspection and justify its inclusion as a genuine kind of introspection.

Language plays a notable role in the examples of external introspec-tion I have described. Perhaps other modes of external cognition can play introspective roles as well, such as introspectively relevant imagery in visual media, but language seems particularly amenable to introspec-tive use through the vast cognitive abilities it enables. In fact, language itself may be considered a socially constructed cognitive artifact, in addition to the various particular cognitive tasks that utilize it (Clark, 1998 ), and it is also arguably an integral component of the conceptual processes described earlier in this chapter, such that the conceptualiza-tion of one’s own mental states occurs through the vehicle of language (in which case concept-based introspection may be aptly characterized as a type of external introspection itself). These are interesting possi-bilities to explore, but I will not press them further here. 2 Suffice it to say that to whatever degree cognition in general can extend out into our interaction with the world, whether through language or some other vehicle, so too may cognitively mediated introspection take place outside the head.

Conclusion

In this chapter, I have discussed how our ordinary cognitive processes are utilized in our ability to introspect. Arguably, from an evolutionary perspective, we have these cognitive abilities because they help us to take in and utilize information about our external environment, but as we have seen, they can also help us understand aspects of ourselves. We are not built to introspect per se, but the cognitive abilities we have enable us to do so, albeit in a fallible and not necessarily always functional way. Perhaps, then, the introspective abilities that these basic cognitive processes enable can be understood as evolutionary spandrels, in that they are by-products of basic adaptive capacities created through the evolutionary processes involved in human ancestry (similar in spirit to Owen Flanagan’s ( 2000 ) treatment of dreams as the ‘spandrels of sleep’). Of course, placing these considerations into an evolutionary framework is highly speculative, and I do not want to put too much weight upon it here, but it is a plausible way of fitting some of our introspective abili-ties into a broader understanding of human nature in general.

In any case, the more central point to take home here is that, however it may have come about, the capacity of a human being to reflect upon her own mental states does not require the existence of any special


cognitive machinery that is specially geared toward that task. To explain our introspective capacities, our basic cognitive machinery will do just fine. I have focused on the roles of three basic aspects of cogni-tion: representation, conceptualization, and attention, together with the more controversial, but still relevant, domain of external cognition, but the full spectrum of cognitive processes at work in introspection is undoubtedly much more complicated than what has been covered in this general overview. I will move on in the upcoming chapters to discuss some more specific aspects of the cognitive processes involved in introspection, as well as the epistemic implications they have, filling in some of the important details to the general explanatory framework established here.

99

Now that we have a framework for understanding the cognitive role of concepts in introspection, we can investigate some of the details of particular conceptual processes and their application to the under-standing of one’s own mental states. One of the more interesting and relevant conceptual structures to consider here is folk psychology. Folk psychology, as it is typically understood in philosophy and cognitive science, is the common conceptualization of human activity in terms of ordinary mental states, most notably beliefs and desires, which we utilize to understand one another in everyday life. If, for example, someone mistakes you for a clerk in a bookstore, asking you where to find the gardening section, you will quite likely understand this person’s behavior in terms of a mistaken belief that you work in the store (assuming you are not actually a bookstore clerk, of course). The concept of belief helps you understand the person’s question, in terms of a background psychological state as the reasonable internal cause of her behavior. In this chapter, I will argue that the folk psychological concepts that we use to understand each other in such contexts are applied recursively as well, to one’s own mental states, thereby consti-tuting a notable kind of conceptually mediated introspection. Moreover, in light of the conceptually mediated nature of folk psychology, we will see further how the introspective attribution of mental states to oneself is a fallible and opaque process, in contrast to the Cartesian transpar-ency and infallibility often assumed of introspective knowledge.

Many describe folk psychology in terms of a causal theory of beliefs and desires, the application of which enables the prediction of human behavior, just as a causal theory of gravity enables predictions concerning physical objects (e.g., Churchland, 1981 ; Fodor, 1987 ). So many follow this characterization of folk psychology as a causal theory, in fact, that

6 Understanding Our Own Beliefs and Desires


it is now often referred to simply as ‘theory of mind’ among main-stream cognitive scientists (e.g., Baron-Cohen, 1995 ). Others, however, offer alternative conceptions of folk psychology, and some doubt the cogency and utility of the idea itself (e.g., Gallagher, 2001 ; Ratcliffe, 2007 ). I will assume that folk psychology exists in at least some form or other, but will adopt a fairly neutral orientation to folk psychology, one that may be acceptable to, and perhaps even complimentary with the concerns of, those who are critical of the idea as it is conceived and utilized in mainstream circles. I will need to make a few selected stands concerning folk psychology (rejecting the first-person authority assumed by simulationist accounts of folk psychology, for instance), but I choose to be as neutral and non-committal as possible here because my purpose is not to establish an account of folk psychology itself but rather to consider how folk psychological concepts are at work in our introspective abilities, supplying a conceptual framework that we can apply interpretively to ourselves as well as to others. All I will assume is that we at least sometimes utilize a theory-like (in a loose sense, as will be explained) conceptual framework of beliefs, desires, and other kinds of mental states in our commonsense understanding of human experience and activity. I will, as much as possible, refrain from strong commitments regarding the origins and underlying mechanics of folk psychology and the extent to which it alone accounts for social cogni-tion in general.

To illustrate the neutrality of my use of folk psychology with regard to the more general and contentious issues concerning the nature of folk psychology itself, let us first look at some of the critiques that have been launched against the use of folk psychology as an explana-tory construct in philosophy and cognitive science, before considering its application to introspection in particular. Shaun Gallagher ( 2001 ) has criticized the understanding of social cognition in terms of folk psychological attributions, arguing for the primacy and prominence of body-based sensorimotor interaction in our social practices, over the abstract conceptualization of mental states commonly emphasized in accounts of folk psychology. Extending the kind of critique launched by Gallagher further, Matthew Ratcliffe advocates for the elimination of ‘folk psychology’, arguing that it is a misguided philosophical inven-tion rather than a genuine phenomenon in everyday commonsense experience and social life (2007). Both Gallagher and Ratcliffe reject mainstream explanatory appeals to folk psychology, critical of the idea that we regularly utilize a conceptual theory involving mental state attributions to understand each other. I think there are viable points

Understanding Our Own Beliefs and Desires 101

in their critiques that ought to be considered, and I would wager that much of what goes on in human social life falls outside the explana-tory purview of folk psychology per se, 1 but I also think that we need to be careful here about throwing out the baby with the bathwater. Even if the embodied personal interactions and external social structures aptly emphasized by Gallagher and Ratcliffe play a more substantial role in social cognition than folk psychology, this does not entail that folk psychological concepts have no significance in our general under-standing of human life.

It is interesting to note that both Gallagher and Ratcliffe qualify their critiques by allowing for rare or isolated uses of folk psychological concepts and theorizing. Gallagher says, for instance:

... I do not want to deny that we do develop capacities for both theo-retical interpretation and simulation [The two main competing accounts of folk psychology in mainstream work on the topic], and that in certain cases we do understand others by enacting just such theoretical attitudes or simulations. Such instances are rare, however, relative to the majority of our interactions. Theory theory and simulation theory, at best, explain a very narrow and specialized set of cognitive practices that we sometimes use to relate to others (this would constitute a weak pragmatic claim for theory of mind). Neither theoretical nor simulation strategies constitute the primary way in which we relate to, interact with or understand others (2001, p. 85).

Along similar lines, Ratcliffe says, ‘Propositional attitudes, principally beliefs and desires, may not be ubiquitous features of interpersonal understanding. However, they are surely central to cases where we explicitly understand or explain someone’s behavior in psychological terms (2007, p. 185)’. Indeed, it would be rather odd for Gallagher or Ratcliffe to deny the application of folk psychological concepts alto-gether in human life, given the prominent explicit theorizing about beliefs involved in their own work and careers as philosophers. So, even a strong critique of folk psychology can allow for the conceptual theori-zation of mental states in some limited contexts of human life, namely those contexts where we are explicitly taking an abstract theoretical stance in our understanding of human activity. This, in fact, is all that my application of folk psychology to introspection will require. I think that the use of folk psychological concepts like ‘belief’ and ‘desire’ is more prominent in everyday life than Gallagher or Ratcliffe let on, but


a minimal commitment to the human capacity to conceptually reflect upon people as beings with internal mental states like beliefs and desires, however rarely or frequently the capacity appears in everyday life, is sufficient to ground the introspective capacities I discuss here. 2

So, if folk psychology is applicable to introspection, how exactly is it that we understand the thoughts and actions of both ourselves and others in terms of beliefs and desires? One common and somewhat intuitive answer is that we develop an understanding of our own beliefs and desires through direct first-person experience and then inferen-tially apply this understanding to others. This answer meshes with commonsense conceptions of first-person privilege regarding intro-spection, as well as intellectual reliance upon introspection as a foun-dation for knowledge (Descartes’ cogito, for instance), however in what follows I will argue that it is mistaken. Drawing upon some revealing findings that bring first-person privilege into question, I will illustrate how our conceptual understanding of our own mental states is medi-ated by the same fallible commonsense understanding of beliefs and desires that we apply to others. In order to have an understanding of our own beliefs and desires as beliefs and desires, we cannot simply observe them within our minds but must rely upon a background interpretive conceptual scheme that gives meaning and content to the concepts of ‘belief’ and ‘desire’ as we apply them to the phenomena of our own mental lives. As Gallagher himself states, while emphasizing the abstract conceptual nature of a folk psychological theory of mind, ‘This conceptual recognition involves an element of abstractness. To discover a belief as an intentional state even in myself requires that I take up a second-order reflective stance and recognize that my cogni-tive action can be classified as a belief (2001, p. 91)’.

Folk psychology and first-person authority

Many people hold or simply assume that we have direct unmitigated access to our own mental processes, including what we believe, think, desire, and intend. As an indication of popular commitment to this perspective, we generally trust that people know what they are thinking and know why they are doing whatever it is that they are doing. For example, Kevin Falvey, speaking for the commonsense trust that we grant to first-person authority, states that ‘it is not uncommon, and frequently entirely appropriate to ask another how he knows what he claims about matters of fact in general. I assume that the singular inap-propriateness of this question in the face of a person’s avowal of his


belief indicates that we have a distinctive warrant for accepting such avowals as true’. (2000, p. 70). A person is generally thought to have direct incorrigible and authoritative knowledge of her own beliefs and desires on the simple basis that they are her own beliefs and desires, directly and transparently present to her first-person experience. From the common privilege we typically grant to self-knowledge, there is no need, or perhaps even possibility, of giving any further explana-tion or justification for one’s knowledge of one’s own beliefs, desires, and actions. They are immediately given, by their very nature as occur-rences in one’s own mind, and that is where the process simply bottoms out.

It should be noted that on this view, folk psychology is still often regarded as a theory. In fact, perhaps the most common formulation of the view under consideration here is that it is because we have direct access to our own actions originating from our own beliefs and desires that we can come to a theoretical understanding of other people’s actions on these terms. In other words, the idea is that folk psychological theory is the product of inference from our first-person experience of our own minds to our mediated third-person grasp of other minds. To consider a simple example, I know that when I go to a restaurant, it is usually because I have a desire for food and a belief that I can satisfy that desire at the restaurant, with the motivating belief/desire nexus presumed to be transparent to my experience of my own agency. In observing someone else going to a restaurant, however, I can only infer indirectly that the person has operational beliefs and desires behind the observed behavior, on the basis of my past experience and its connections with similar behavior. The attribution of a desire for food to the person in the restaurant is a theoretical inference, but one that seems rooted in my own prior first-person experience of analo-gous situations. From this perspective, my understanding of others is a matter of abstract conceptualization and theoretical reasoning, but my understanding of myself seems direct and unmediated. As plausible as this perspective may sound, it is mistaken. It seems intuitively obvious to many, but this is a case in which our intuitions lead us astray, giving us false confidence in the seeming transparency of our own minds.

The theory theory challenge to first-person authority

The common assumption of first-person authority is misguided because our understanding of our own beliefs and desires is not direct and immediate but rather mediated by a folk psychological theory. Instead of


having direct access to our own beliefs and desires as the causal sources of our own actions, we apply theoretical concepts in attributing such states to both others and ourselves. In other words, the ‘theory theory’ claim that folk psychology consists of a theory cannot be restricted to third-person ascriptions, but rather extends to our own first-person folk psychological attributions as well. Understanding one’s own beliefs and desires as beliefs and desires is not immediately given in experience. Instead, it involves the fallible theoretical application of folk psycho-logical concepts to one’s own mental states.

To illustrate the theory-mediated nature of folk psychological attri-butions, it is helpful to see how this theory theory conception of folk psychology is akin to the related claim that observation is theory-laden, as discussed in the philosophy of science. Consider, for instance, Alan Musgrave’s depiction of this idea:

... when we look in the direction of the table and declare ‘There is a table here’ we interpret certain sensory inputs as indicating the presence of a table. But if we did not previously know the meaning of the word ‘table’, or possess the concept of a table, then we could not formulate the observation-statement. A creature which did not possess the concept of a table could still see the table; but such a creature could not learn from this experience that there is a table here and could not formulate the observation report ‘Here is a table’. ... [This] gap between experiences and observation reports, a gap filled by language and concepts, is an important source of the fallibility of observation reports. (1993, p. 55)

In a similar fashion, our folk psychological understanding of human thoughts and actions, including our own thoughts and actions, requires the employment of mediating concepts, namely, the concepts of ‘belief’, ‘desire’, and the various other mental state concepts involved in our abstract understanding of the roles beliefs and desires play in human thought and action.

There is a crucial difference here between having or experiencing folk psychological phenomena (i.e., beliefs and desires) and having a conceptual understanding of such phenomena. One can have the former without the latter. For example, although there is some room for skepti-cism, I think it is quite plausible to think of animals and pre-linguistic/conceptually-naive children as having beliefs and desires. However, they may have little to no understanding of their own beliefs and desires as beliefs and desires. 3 This further understanding of beliefs and desires as


beliefs and desires requires a theoretical understanding of beliefs and desires that enables mediated inferences from the phenomena of beliefs and desires (e.g., the occurrence of having a belief or the experience of a desire) to a conceptual grasp of what these phenomena are and the roles they play in behavior. As Musgrave put it, there is a gap between an experience and the construction of an observation statement about that experience, a gap that is mediated by concepts. Analogously, there is a gap between the mere presence of beliefs and desires and an under-standing of such phenomena as mental states of certain types, and this gap is what makes our understanding of our own beliefs, desires, and actions a theory-mediated, opaque, and fallible phenomenon. So, contrary to the common assumption of first-person authority, we do not have a direct and immediate understanding of our own beliefs and desires as beliefs and desires. Rather, we understand such phenomena via a mediated conceptual framework regarding the nature of such phenomena.

So, what exactly is this folk psychological theory that we use to intro-spect, and where does it come from? In order to flesh out the terrain of the theory theory perspective as it pertains to introspection, I will briefly point to some possibilities here, without committing myself to any particular viewpoint. As already suggested by the analogy above, one view of the theory theory is that we acquire our theory of folk psychology in the same way that scientists construct theories. As expressed by Gopnik and Meltzoff, the basic idea here is that ‘the processes of cognitive development in children are similar to, indeed perhaps even identical with, the processes of cognitive development in scientists’. (1997, p. 3). From this perspective, we begin with a prelimi-nary hypothetical understanding of human action and go through a variety of revising and reformulating processes in light of subsequent experiences, ultimately culminating in our mature adult folk psycho-logical understanding. Another viewpoint, which has come to promi-nence via evolutionary psychology, gives much more emphasis to the idea of having an innate theory of mind. According to this account, our folk psychology is rooted in an adaptive cognitive module that provides a ‘built-in’ theory of human action (For one elaboration and defense of this idea, see Botterill & Carruthers, 1999 , pp. 77–103). In other words, it is a part of our nature, ‘hard-wired’ into our brains, to cognitively process human actions in terms of beliefs and desires. From yet another viewpoint, we acquire our folk psychological theory socially, through a cultural belief system. In illustration of this idea as it pertains to intro-spection, Lyons states that


As soon as our ‘introspecting’, or anything else for that matter, involves this web of intentional concepts [i.e., folk psychological theory], it is also inevitably involved in a certain stereotyped culture-based outlook. Thus our ‘introspection’ of our motives, desires, deliberations, plans, decisions, or intentions will be colored by and adapted from our own particular culture’s ‘rough-and-ready’ view of the mind and its operations. (1986, p. 126)

As I see it, determining which of these viewpoints is correct is an open empirical issue, which we need not have a stake in here. Any one of these views, or some modified combination across them, could be true without significant impact to my more narrow and focused claims here regarding folk psychology-mediated introspection. In fact, it is worth noting that some critical responses to theory theory-type accounts of folk psychology are compatible with my account of the role of folk psychology in intro-spection. For example, Daniel Hutto rejects theory theory accounts on the basis that folk psychology operates through learned narrative prac-tices rather than innate mental modules or hypothesized generalizations (2009), but the narrative practices that he emphasizes could equally serve as the cognitive medium that lies between our experiences and the fallible folk psychological interpretations we make of them. 4 The theory in ‘theory theory’ need not be a standardly-recognized kind of theory in order to make the more general point about mediation here. All that the present point requires is that our folk psychological attributions occur through conceptual structures that mediate between our experiences and the interpretations we make of them, whether that conceptual medium be a science-like theory, an innate computational language of thought, or the ordinary language we use in everyday life.

Empirical evidence of theory-mediated introspection

So, the key claim here is that our introspective understanding of our own folk psychological mental states (e.g., beliefs and desires) is mediated by an interpretive conceptual scheme that renders such introspection fallible. But why should we accept this mitigated picture of ourselves? What reasons do we have for accepting the conceptually mediated status of introspection I have described? As I see it, the most relevant and convincing considerations on the matter come from empirical evidence. If we rely upon a priori philosophical reflection alone, as many philosophers are inclined to do, we are presented with the apparent immediacy of our beliefs and desires and are likely to remain within


the paradigm of first-person authority I described earlier. In order to tease apart the seemingly transparent conception we have of ourselves and the third-person claim that we apply theoretical constructs when understanding ourselves, we must turn to the relevant empirical results in neuroscience, cognitive psychology, and developmental psychology. There is a large and growing body of empirical work pertaining to folk psychology, the vast bulk of which is aimed towards our understanding of each other rather than our introspective abilities. However, some noteworthy findings shed light on folk psychological introspection as well. In what follows, I will look at three different empirical studies, each of which demonstrates how our attributions of folk psychological states to ourselves can fall prey to the same kinds of errors we make when attributing mental states to others, thereby revealing the opaque fallible nature of folk psychological self-attributions.

Perhaps the most striking evidence comes from the now-famous split-brain studies of severe epileptics. Michael Gazzaniga, Roger Sperry, and other researchers have conducted a number of studies on people that have had their corpus callosum severed (as a remedy for severe epilepsy), with the two hemispheres of their brains thereby divided. One result of this procedure is that such people have two separate ‘sides’, or streams of experience that may be disassociated from one another. This allows researchers to expose the two ‘sides’ of a person to different stimuli, without each side knowing what the other side experiences. The results are quite interesting, as Gazzaniga explains in the following scenario:

One picture was shown exclusively to the left hemisphere and the other exclusively to the right. The patient was asked to choose from an array of pictures ones that were lateralized to the left and right sides of the brain. In one example, a picture of a chicken claw was flashed to the left hemisphere and a picture of a snow scene to the right hemisphere. Of the array of pictures placed in front of the subject, the obviously correct association was a chicken for the chicken claw and a shovel for the snow scene. One of the patients responded by choosing the shovel with his left hand and the chicken with his right. When asked why he chose these items, his left hemi-sphere replied, ‘Oh that’s simple. The chicken claw goes with the chicken, and you need a shovel to clean out the chicken shed’. In this case the left brain, observing the left hand’s response, interpreted that response in a context consistent with its sphere of knowledge – one that does not include information about the snow scene. What is amazing here is that the left hemisphere is perfectly capable of saying


something like, ‘Look, I have no idea why I picked the shovel – I had my brain split, don’t you remember? You probably presented some-thing to the half of my brain that can’t talk; this happens to me all the time. You know I can’t tell you why I picked the shovel. Quit asking me this stupid question’. But it doesn’t say this. The left brain weaves its story in order to convince itself and you that it is in full control. (Gazzaniga, 1998 , pp. 24–25)

For our purposes, the important point here is that the patient’s speaking left side/brain does not have access to the causal basis for his response but still readily comes up with an explanation for the answer he chose. Presumably, he (the right side) chose the shovel because he believed that it most appropriately fit with the snow scene picture. The patient’s left side was not aware of this and therefore could not appeal to this belief in explanation of his actions. However, as Gazzaniga emphasizes, this lack of access does not stop the patient from giving an explanation of his action. Without any hesitation, the patient confabulates a belief associating the shovel with the chicken, falsely but nevertheless confi-dently claiming that this belief is the source of his action.

How and why does he do this? The most plausible explanation is that the patient tacitly made an inference based upon his general under-standing of human psychology. In other words, he automatically (and unconsciously) drew upon his background understanding concerning what would be a reasonable response and then formed his self-attribution on the basis of the result. Why else would a person pick a picture of a shovel when shown a picture of a chicken, besides appealing to some associational belief about cleaning chicken sheds? This seems infer-entially plausible from a general folk understanding of human ration-ality and decision-making, coupled with a cultural understanding of the human use of chickens, of course. It is such an understanding that this patient is applying to himself, just as he would to someone else if interpreting the same response in another person. So, it appears that the human adult has an incredibly facile ability to reason about human actions via some sort of theoretical understanding regarding the causal basis of human actions, and that this ability is employed to understand not only other people’s actions, but also one’s own actions. This all happens quite automatically, such that it could easily give the appear-ance of direct access to beliefs and desires as the causal bases of one’s own thoughts and actions.

One might worry here that this result is merely a strange by-product of people with split brains, and not an accurate depiction of what happens


in the minds of normal people. After all, people with split brains might conceivably be thought of as containing two distinct and independent sets of mental processes, and therefore the above case may be more like a case of understanding someone else rather than oneself. However, this objection will not do because the same self-rationalizing process can be found in undivided folks as well. Consider, for instance, the following influential study conducted by Nisbett and Wilson:

On a busy Saturday morning Nisbett and I placed a sign on a display table that read: ‘Consumer Evaluation Survey – Which Is the Best Quality?’ We then made sure that four pairs of nylon panty hose were arranged neatly on the table and waited for the first passerby to stop and examine them. ... In an earlier version of the study, we noticed that people showed a marked preference for items on the right side of the display. We observed this same position effect in the panty-hose study. The panty hose were labeled A, B, C, and D, from left to right. Pair A was preferred by only 12 percent of the participants, pair B by 17 percent, pair C by 31 percent, and pair D by 40 percent, for a statistically significant position effect. We knew that this was a posi-tion effect and not that pair D had superior characteristics because in fact all the pairs of panty hose were identical – a fact that went unnoticed by almost all our participants. After people announced their choice, we asked them to explain why they had chosen the pair that they did. People typically pointed to an attribute of their preferred pair, such as its superior knit, sheerness, or elasticity. No one spontaneously mentioned that the position of the panty hose had anything to do with the preference. When we asked people directly whether they thought that the position of the panty hose had influenced their choice, all participants but one looked at us suspiciously and said of course not. (Wilson, 2002 , pp. 102–103; see also Nisbett & Wilson, 1977 )

The subjects of this study, just as the split-brain patient described earlier, employed a tacit folk psychological theory in their understanding of their choices. As Nisbett and Wilson stated in their original report of the study, ‘w hen people are asked to report how a particular stimulus influenced a particular response, they do so not by consulting a memory of the mediating process, but by applying or generating causal theories about the effects of that type of stimulus on that type of response’. (1977, p. 248). While they in fact had no good reason for choosing one pair of panty hose over another, the subjects of the study readily attributed


to themselves beliefs about desired features of the panty hose. The best explanation seems to be that these beliefs about the panty hose were retrospectively confabulated by the subjects, in accordance with their folk psychological understanding of why people make such choices. In other words, the subjects explained their actions not by directly accessing and remembering the sources of their actions but rather by inferentially attributing certain beliefs and desires to themselves that make rational sense of their actions. Ideally speaking, rational people choose prod-ucts on the basis of the products’ qualitative features as they relate to their desires and interests (though in reality, especially in a culture with prominent advertising and consumer manipulation, we may be neither so rational nor self-directed in our product ‘choices’). The subjects in this study intuitively understood this picture of human decision-making and applied it to make sense of their own actions when pressed to explain why they did what they did. This provides notable empirical support for the idea that we understand our own beliefs and desires via a tacit (but readily available) theoretical understanding of human psychology.

False-belief tests in developmental psychology provide even further evidence that reveals the theory-laden nature of self-attributions. Through a number of studies, developmental psychologists have reached a general consensus that the ability to represent false beliefs comes online roughly around the age of 3 ½ in children. For instance, 3-year-olds often fail to exhibit any understanding of false beliefs but by age 4 or 5, such an understanding appears to be firmly in place (Gopnik, 1993 ). There is much to discuss concerning the study of folk psychology through false- belief tasks, but for our purposes the most significant feature of these findings is that 3-year-olds generally make the same sorts of errors about themselves as they do about others. In one revealing case, Gopnik and Astington had children encounter a candy box that had been deceptively filled with pencils. The experimenters questioned children after they had discovered the box’s deceptive contents, asking them about both what other children will believe about the box and what they themselves believed about the box prior to discovering its contents. In both cases, the results were strikingly similar:

One-half to two-thirds of the 3-year-olds said they had originally thought there were pencils in the box. They apparently failed to remember their immediately previous false beliefs. Moreover, chil-dren’s ability to answer the false-belief question about their own belief was significantly correlated with their ability to answer the question about [another child’s] belief ... (1993, p. 7)


It is important to note that memory effects were controlled for in this study. The children’s inability to attribute prior false beliefs to them-selves was not simply a matter of forgetfulness, but rather an inability to comprehend having had a false belief. The children who failed in this task were missing some kind of understanding of the misrepresen-tational capacities of beliefs, and they equally failed both in regard to others and themselves. Furthermore, the children that were successful at the task showed a significant correlation between attributing false beliefs to themselves and to others. These results strongly suggest that one’s folk psychological understanding of oneself develops parallel to one’s understanding of others. Rather than having direct privileged understanding of our beliefs and desires, our representational grasp of our own mind appears to develop alongside our understanding of others, being prone to the very same sorts of errors and benchmarks concerning mental state attributions. It is reasonable to infer from such findings that our ability to understand our own mental states depends upon the same cognitive processes we utilize in understanding the mental states of others. As Gopnik herself states, this provides signifi-cant support for the theory theory account of folk psychological introspection:

The developmental evidence suggests that children construct a coherent, abstract account of the mind which enables them to explain and predict psychological phenomena. ... Moreover, the child’s theory of mind is equally applicable to the self and to others. (1993, p. 10)

So, we once again reach the conclusion that our grasp of our own psychology is mediated by some kind of theoretical understanding of human psychology in general. By considering the empirical findings I have outlined here, we can see that the seemingly direct access we have to our beliefs and desires, as presumed by those who maintain first-person authority in self-attributions, is illusory. Instead, our concep-tual grasp and understanding of our mental lives is theory-laden, being mediated through a background folk psychological conceptual structure of some sort or other. So, rather than having unquestioning confidence in our attributions of beliefs and desires to ourselves, under-stood in terms of direct and transparent access to the contents of our own minds, we ought to recognize that such attributions are only as explanatorily grounded as the abstract conceptual devices we employ in making them. This is not to say that we know about our own beliefs and desires in exactly the same way in which we know about the beliefs


and desires of others, but rather that both kinds of attributions are rooted in, and therefore epistemically dependent upon, the same sort of abstract conceptualization.

Mind-monitoring mechanisms and the phenomenology of propositional attitudes

Now that I have explained and defended the idea that our introspective grasp of our own beliefs and desires is mediated by folk psychological concepts, I will consider some potential objections and, in doing so, also elaborate upon some important features of my perspective on the topic. Nichols and Stich ( 2003 ) have recently criticized the theory theory on the basis that it cannot account for self-awareness of one’s own beliefs and desires. While they accept that we employ a theory of mind in understanding the beliefs, desires, and actions of others, and even when we reason about our own beliefs, desires, and actions, Nichols and Stich argue against the idea that our initial access to our own beliefs and desires is a result of employing a theory of mind. Instead, they posit the existence of a special dedicated ‘monitoring mechanism’ that provides individuals with the ability to detect their own beliefs and desires:

The basic facts are that when normal adults believe that p, they can quickly and accurately form the belief I believe that p ; when normal adults desire that p, they can quickly and accurately form the belief I desire that p ; and so on for the rest of the propositional attitudes. In order to implement this ability, no sophisticated Theory of Mind is required. All that is required is that there be a Monitoring Mechanism (MM) (or perhaps a set of mechanisms) that, when activated, takes the representation p in the Belief Box as input and produces the repre-sentation I believe that p as output. This mechanism would be trivial to implement. To produce representations of one’s own beliefs, the Monitoring Mechanism merely has to copy representations from the Belief Box, embed the copies in a representation schema of the form: I believe that ___ ., and then place the new representations back in the Belief Box. The proposed mechanism would work in much the same way to produce representations of one’s own desires, intentions, and imaginings. (2003, p. 160)

So, Nichols and Stich claim that there is a mechanism in the mind that transforms belief (and other mental) states into propositional represen-tations of themselves. On this view, that is all that is needed for the


detection of beliefs and desires. We do not need the application of a theory of mind to know what we believe and desire, but rather only a simple monitoring device.

I think that Nichols and Stich are correct to acknowledge that we do not necessarily access our own beliefs and desires in the same way that we access the beliefs and desires of others. As they point out, it would be absurd to think that we know of our own beliefs and desires solely by inferring them from our own externally observable behavior. In virtue of being constituted by our own conscious mental states, as explained earlier in Chapter 4 , we have first-person experiential knowledge that is not available to others. For instance, a person could understand herself as having a desire for food on the basis of an experiential feeling of hunger, prior to and irrespective of any resulting behavior that may ensue from that hunger. There are first-person bases for mental state attributions that are not accessible from a third-person perspective, and any post-behaviorist account of human psychology ought to acknowl-edge this. However, crucially, Nichols and Stich are too quick to jump from such uncontroversial claims about the data available to us from a first-person perspective to the idea that no theoretical structures are implicated in identifying our own beliefs and desires as beliefs and desires.

To see this, let us consider how Nichols and Stich’s proposed moni-toring mechanism can actually do its work, something which they say surprisingly little about. The claim is that some sort of mechan-ical process takes a belief (or other mental state) and creates a further belief regarding the belief-holder having the first-order belief fed into the mechanism as input. In order to do this, the proposed mechanism must, at a minimum, have some conception of what it is to have a belief, such that beliefs can be recognized and individuated from other mental phenomena. Otherwise, the proposed mechanism, however it might be constituted, simply could not identify a belief as a belief. Considering this, there must be some kind of tacit appeal to a conceptual theory of mind here. Although the concept of belief might be a relatively simple part of a theory of mind, it is nevertheless a theoretical posit, roughly on a par with other theoretical notions that play roles in our everyday understanding, such as the concept of an object or a force. Again, there is a significant difference between having a belief or desire and under-standing that mental state as a belief or desire. Arguably, a dog might have the belief that sitting up on its hind legs will result in getting a treat, without having any understanding at all that this state of belief is a belief. Such understanding is something quite different from the


presence of the belief itself. In order to have this further understanding, the dog would need to have some kind of cognitive grasp of what it means to have a belief, which is to say that it would need a conceptual theory of some kind or other that posits beliefs as a kind of mental entity, however this theory may be actually manifested in the mind/brain. By taking note of this, we can see that Nichols and Stich’s sugges-tion does not offer an alternative to the theory theory, but rather tacitly depends upon it for the identification of a belief as a belief. Whatever the merits of their proposed mechanism might be, it requires a tacit appeal to a theoretical conceptual structure that dictates what it means to have a belief, or other mental state. Without such a conceptual struc-ture residing somewhere within a person’s psychology (including the proposed monitoring mechanism), the idea of a person having a ‘Belief Box’ is simply incoherent. There must be some kind of theoretical understanding in order for a person to form beliefs about beliefs in an understandable and utilizable manner.

Another potential objection to the theory theory comes from Alvin Goldman’s phenomenal account of mental state concepts, as a part of his simulationist approach to folk psychology (1993; 2006). Goldman suggests that we identify our own mental states, including propositional attitudes like beliefs and desires, by phenomenal characteristics of types of mental states, and not by any functional role that such states play in a theory of mind. To this effect, he states that

... someone who had never experienced certain propositional atti-tudes, for example, doubt or disappointment, would learn new things on first undergoing these experiences. There is ‘something it is like’ to have these attitudes, just as much as there is ‘something it is like’ to see red. ... a plausible-looking hypothesis is that mental states are states having a phenomenology, or an intimate connection with phenomenological events. This points us again in the direction of identifying the attitudes in phenomenological terms. (1993, p. 24)

If Goldman is right here, then people can identify their own beliefs and desires without appealing to any folk psychological theory, by simply recognizing their phenomenal character. This would provide grounds for rejecting the theory theory as I have portrayed it, since it would exclude theory mediation in our first-person access to our own beliefs and desires.

But is Goldman right? I think not, for a number of reasons. First of all, I think Goldman’s ‘phenomenology’ is inaccurate, more likely


a rationalization of his own philosophical theory than an authentic description of human experience. The idea that there is a phenomenal property that uniquely picks out beliefs in terms of what they are like experientially is highly dubious. People have beliefs with all sorts of different phenomenal qualities, from dread to delight and from interest to apathy. Moreover, many of our beliefs fall outside the domain of phenomenal experience altogether ... tacit assumptions that do not make a direct appearance in conscious thought, for example. Considering both the variety of phenomenal qualities that may be associated with beliefs and the possibility of having a belief with no phenomenal quality at all, there is simply no identifiable qualitative experience that is specific to having a belief. Since Goldman’s account posits and requires that such a qualitative experience be manifestly apparent, I conclude that it fails. In contrast with Goldman, I am inclined to think that what typifies a belief as a belief is the role that it plays in one’s psychology. When we attribute beliefs to ourselves, we do so in virtue of there being some kind of representative state that can form the basis for an inference or provide reasons for action. In other words, beliefs are beliefs in virtue of what they do in our cognitive economy, and are attributed on the basis of their roles rather than any phenomenal properties they purportedly share. To adequately argue for this conception of beliefs would require much more than what I can plausibly do here, but, fortunately, my criti-cism here need not offer a positive conception of beliefs. Whatever it is that makes a belief a belief, such that it can be identified as a belief, it is clearly not a phenomenal property.

Moreover, even if we grant to Goldman that there is a phenomenal character to beliefs, his account is still fundamentally inadequate. A phenomenal experience alone is not sufficient for identifying your mental states as particular types of states, nor understanding why you do what you do on the basis those states. To do these things, you must be able to accommodate the experience into some kind of interpretive structure that conceptualizes the mental state as the kind of mental state it is. Let us consider the case of desire, which could much more plausibly be said to have a characteristic phenomenal state. Suppose that you are having a meal at an outdoor cafe. Just after your food is served, a rather skinny dog comes to your table and starts to whine. In this case, you would be safe to assume that the dog has a desire for your food. Doubts about animal consciousness aside, let us say that this dog has some basic qualitative experience typically associated with desire. But, could this dog understand its desire in the same way that you do? Could it identify its desire as such and understand the role the


desire is playing in determining its actions? There are good reasons to think not. In order to have this understanding of its desire, the dog would need to have access to some kind of conceptual structure that accounts for what desires are, at a minimum. This conceptual structure might include particular kinds of phenomenal experiences as identi-fying features of desires, but those phenomenal experiences in them-selves do not convey that a desire is a desire. Such conveyance requires a concept under which the phenomenal experience may be placed. In other words, desires are not conceptually self-intimating; they do not in themselves present themselves as desires. Understanding a desire as a desire requires some additional cognitive interpretation beyond the desire itself. So, having a phenomenal experience of a cognitive/folk psychological state does not necessarily entail that it is understood as such. Thus, the dog at your table, arguably lacking these higher-level conceptual abilities, would not know of its desire as a desire, however vivid and present the phenomenal quality of its desire may be. As I have emphasized throughout this chapter, there is a significant difference between having or experiencing a folk psychological state and under-standing the state in folk psychological terms. An understanding of a desire as a desire (or a belief as a belief, etc.) inherently requires not only the presence of a desire but also some kind of abstract concep-tion of what a desire is. So, again, I conclude that Goldman’s account simply falls short as an accurate portrayal of our folk psychological understanding of our own beliefs and desires. The theory theory must be brought into play in order for us to account for the manner by which we understand our own beliefs, desires, and actions.

Conclusion

In this chapter, I have addressed the relationship between our folk psychological understanding of ourselves and our understanding of other people, arguing that they both rely upon the same background interpretive structures. Our commonsense understanding of folk psychology provides the conceptual underpinnings necessary for under-standing beliefs, desires, and actions, even from our own first-person standpoint. So, our understanding of our own such mental states is mediated and fallible. We do not have a special privileged conceptual grasp of our own mental states, as many philosophers have maintained. If we know our own beliefs and desires better than others, it is only because we have phenomenal knowledge of what they are like experi-entially, through our first-person experience of being ourselves, and not


because we have some privileged conceptual grasp or understanding of ourselves. In fact, we can be just plain wrong about what we ourselves believe or desire and the roles our beliefs and desires play in why we do what we do. On this view, external, third-person perspectives (such as the experimental data I discussed) can override a first-person report regarding the content of one’s own mind or the reasons why an action was performed. While it certainly seems that we have direct, unmedi-ated access to our own beliefs, desires, and actions, the empirical facts indicate that this is illusory. Instead, we apply a tacit interpretational understanding when attributing mental state concepts to ourselves.

Zooming out to the bigger picture, it is important to recognize that folk psychological understanding of oneself is only one facet of intro-spection. We do have a privileged experiential knowledge of our own mental states, in terms of what it is like to be in them, but this knowledge must be distinguished from the conceptual knowledge we obtain (or fail to obtain) through the interpretational overlay of folk psychology. A lot of mistaken philosophy has been rooted in the conflation of these epistemologically distinct domains, such that the unique knowledge we have of our lived experiences is taken as a privileged conceptual foun-dation. Most centrally, consider Descartes’ cogito, whereby he infers an indubitable transparent grasp of his own thoughts from his existence as a conscious being. This Cartesian foundationalism mistakenly over-looks the unseen mediation of concepts between propositional thought and experience, constructing a false unity across differing facets of introspection. To be sure, there is an element of intuitive appeal in the idea of first-person introspective privilege and authority, but to have an accurate understanding of our introspective position here we must tease apart the epistemologically distinct domains involved, parsing out in what ways the epistemology of introspection is unique and in what ways it overlaps with epistemic properties of other domains of human life and cognition.

Moreover, although I have treated folk psychology itself as a unitary domain for the purposes of this chapter, it too might decompose into different elements that have distinct epistemic properties as they pertain to introspection. For example, with regard to the purportedly competing accounts of the nature of folk psychology that I surveyed briefly above (folk psychology as an innate adaptive module, science-like theorization, cultural framework, and/or learned narrative prac-tice), it may be the case that folk psychology itself is a heterogeneous phenomenon that draws upon some or all of these cognitive processes, each of which may have more nuanced characteristics that are relevant


to the nature and epistemic properties of introspection. There is room for much more investigation here, extending beyond the big-picture overview of the terrain established in this book. For now, however, we can look ahead to the next chapter as a case in point. As I briefly pointed out, one recent approach to folk psychology is to understand it in terms of the learned narrative practices embodied in everyday social interac-tion (Hutto, 2009 ). Presumably, ordinary spoken language is the vehicle of these practices, being the fabric through which the narratives we tell each other are spun. The next chapter looks at the role of language in introspection, detailing the roles and epistemic characteristics of inner speech in introspection. If this language use is also a mechanism of folk psychological understanding, such that the narratives through which we understand people’s beliefs and desires are applied recursively to oneself as well, then its epistemic characteristics (which, as we will see, are themselves somewhat diverse) are relevant and applicable to a nuanced understanding of folk psychological introspection.

119

People commonly report an internal stream of words constituting their own private thoughts. Reflections come and go in the form of linguistic expressions, whether we are thinking about dinner plans, mulling over a philosophical debate, or rehearsing a discussion with another person. This phenomenon is known as inner speech. Inner speech is the expe-rience of articulated natural language phenomena through internal auditory images and sub-vocal assertion, typically with phenomenal properties similar to one’s own voice, as if one is silently speaking to oneself. 1 In this chapter, I will discuss and analyze the nature of this inner speech and the roles it plays in introspection. I will argue that a significant portion of our thoughts (namely, our conscious proposi-tional thoughts) occur through inner speech and, consequently, that this vehicle for conscious thought provides the basis for substantial aspects of our awareness and understanding of our own minds. I will then discuss the epistemic status of inner speech, explaining how inner speech utterances exhibit a variety of different epistemic qualities, from utterances that self-constitutively determine their own truth values to the tendency of some linguistic phenomena to misguide and even deceive ourselves about the contents of our own minds.

Before beginning, however, it will be instructive to lay out some connections between this chapter and the previous two. In Chapters 5 and 6, I explained how much of our introspective self-knowledge is mediated by the higher-level cognitive capacities of human beings. In particular, I have analyzed the role of conceptual understanding in introspection, where we apply abstract conceptual structures, such as the belief and desire framework of folk psychology, to interpret, under-stand, and make sense of what is happening in our own minds. As will become apparent in this chapter, I believe that natural language

7 The Internal Monologue


plays an enabling role in this capacity to conceptually understand our own minds, insofar as such understanding is explicitly manifested in conscious propositional thought. In essence, my view is that language serves as a cognitive vehicle with which we engage in abstract conceptual thought, through which we consciously entertain propositional asser-tions. To be clear, this is not to say that language provides the sole basis for thought in general, but rather that it is one cognitive device, among perhaps many others, with which we think. In particular, language serves as a cognitive vehicle when we are engaged in conscious propo-sitional thought about something, thought that something is the case, providing the tangible experiential medium with which we experience our own such thoughts as they occur. From this perspective, language is an important ingredient in our ability to engage in the kinds of intro-spective thought that involve conscious conceptual representation and abstract reflection upon our own minds.

In other words, the introspective functions of language to be discussed in this chapter facilitate the implementation of the kinds of introspec-tive processes discussed in the previous two chapters. However, there is a reason why I have not emphasized the role of language until now. Although I believe that language plays an important role in introspective thought, the processes that I have discussed thus far may occur without language as well. From a purely analytic perspective, Chapters 5 and 6 can stand alone, without necessary dependence upon the further claims about introspective thought that I make in this chapter. After reading this chapter, I hope that you understand how these areas are, as a matter of fact, connected, but for those who remain dubious of the view that language can perform significant functions in cognition, I have arranged my analysis so that what I have said thus far does not stand or fall upon the roles I attribute to language here in this chapter. With this back-ground context in mind, let us now turn to consider the relationship between language and thought as exhibited in inner speech.

Knowing our own thoughts through language

As it is often understood, language appears to be a communicatory tool. We have a thought and then afterward, when we want to share it with others, we express it in words. In contrast with this understanding of language, however, some philosophers and scientists have argued that language is more than just a tool for expressing thoughts. It is, in addi-tion, a tool with which we think. In particular, it has been claimed that language, in the form of inner speech, is the basis of conscious

The Internal Monologue 121

thought (Dennettt, 1991; Carruthers, 1996). With some qualification and refinement, I will follow this cognitive conception of inner speech and explore its implications regarding the nature and epistemic proper-ties of introspection.

The identification of inner speech with thought has been around at least since Plato’s conception of thought as the ‘dialogue of the soul with itself’, and it can be found in various forms among many noteworthy theorists, some of whom are in profound disagreement with one another on other matters. As I have done in previous chapters, I will navigate between the lines on some broader areas of contention concerning the relation-ships among language, thought, and experience, carving out a somewhat neutral standpoint from which some understanding can be developed regarding introspection. To establish this neutrality, I will begin with a couple of quotes illustrating agreement across enemy lines. First, here is Frege on the role of linguistic symbols in conceptual thought:

I do not deny that even without symbols the perception of a thing can gather about itself a group of memory-images; but we could not pursue these further: a new perception would let these images sink into darkness and allow others to emerge. But if we produce the symbol of an idea which a perception has called to mind, we create in this way a firm, new focus about which ideas gather. We then select another [idea] from these in order to elicit its symbol. Thus we penetrate step by step into the inner world of our ideas and move about there at will, using the realm of sensibles itself to free ourselves from its constraint. Symbols have the same importance for thought that discovering how to use the wind had for navigation. Thus let no one despise symbols! A great deal depends upon choosing them properly. And their value is not diminished by the fact that, after long practice, we need no longer speak out loud in order to think; for we think in words nevertheless, and if not in words, then in mathematical or other symbols. Also, without symbols we would scarcely lift ourselves to conceptual thinking. Thus, in applying the same symbol to different but similar things, we actually no longer symbolize the individual thing, but rather what [the similars] have in common: the concept. This concept is first gained by symbol-izing it; for since it is, in itself, imperceptible, it requires a perceptible representative in order to appear to us (1882/1972, p. 83).

From this perspective, linguistic symbols provide the tangible sensory vehicles of our conceptual thoughts, enabling us to engage in concep-tual reflection upon what we experience. There is more to Frege’s overall


account here, of course, from the imperceptible and mind-independent Platonic account of concepts that he suggests to the rigorous artificial symbol systems he goes on to propose to remedy the shortcomings he identifies with natural language, but we need not follow him here on these further matters. The key points, for our purposes, are that language serves as the sensible vehicle of conceptual thought and that this function can be internalized without external expression.

A similar account can be found across the analytic/continental divide in philosophy, in the very different phenomenological perspective of Merleau-Ponty:

If speech presupposed thought, if talking were primarily a matter of meeting the object through a cognitive intention or through a representation, we could not understand why thought tends towards expression as towards its completion, why the most familiar thing appears indeterminate as long as we have not recalled its name, why the thinking subject himself is in a kind of ignorance of his thoughts so long as he has not formulated them for himself, or even spoken and written them, as is shown by the example of so many writers who begin a book without knowing exactly what they are going to put into it. A thought limited to existing for itself, independently of the constraints of speech and communication, would no sooner appear than it would sink into the unconscious, which means that it would not exist even for itself. ... we present our thought to ourselves through internal or external speech (1945/2002, p. 206).

With this we again see the identification of language with the activity of conscious thought. Like Frege (despite their vast philosophical differences overall), Merleau-Ponty emphasizes how language brings our thoughts to consciousness, providing the medium by which our thoughts are present in experience. Moreover (and here is where Frege and Merleau-Ponty begin to part ways), there is a sense in which the thoughts themselves emerge in the activity of speech. It is not as though the thoughts pre-exist, either within or outside the mind, but rather are created as they unfold in linguistic expression. Although I will later argue that this picture of self-generated thought does not gener-alize across all instances of inner speech, allowing for the existence of pre-linguistic thoughts that can be either accurately or inaccurately represented in inner speech, self-generated verbal thought can function as an identifiable kind of introspective process, which Merleau-Ponty’s expressivist take on language helps bring to light.


A comprehensive survey of perspectives that link language to conscious thought would require much more discussion, but the examples of Frege and Merleau-Ponty sufficiently highlight the key idea and the fact that it can cut across differing philosophical frameworks. With this as our backdrop, I will now turn to more recent work on the topic, developing our way toward a nuanced understanding of the roles the cognitive functions of language can play in introspection. Peter Carruthers has been at the forefront of this topic in contemporary philosophy, so I will begin by building upon his work here.

In his extensive analysis Language, Thought, and Consciousness: An Essay in Philosophical Psychology , Carruthers proposes that natural language is constitutively involved, as a matter of natural necessity, in conscious thought: conscious thought simply consists in the tokening and mental manipulation of sentences in natural language (1996, p. 38). It is important to qualify that Carruthers’s version of the cogni-tive conception of language is confined to conscious thought. He believes (as do I) that thought in general is conceptually independent from language and grants the existence of thought without language. To illustrate this, Carruthers points to the thought-based behavior of animals and pre-linguistic humans (Ibid, p. 16). Such creatures exhibit the presence of thought without natural language, through various elementary reasoning skills exhibited in their behavior. However, Carruthers believes that conscious thought requires natural language, which leads to his controversial view that animals lacking language are not conscious creatures. I will shortly make a further qualification that brings this general claim about consciousness into question. First, however, I want to focus on the claim that the conscious thought of humans with language is constituted by their use of language, which, as we will see, can be distinguished from the broader claim about consciousness in general.

Carruthers, like both Frege and Merleau-Ponty, grounds this claim in the phenomenological observation that we find within our conscious experience the presence of inner speech. Our thoughts often, perhaps even typically, occur to us in linguistic form. Based upon this obser-vation, Carruthers argues that such internal linguistic occurrences constitute the thoughts we have. The words in inner speech simply are conscious thoughts. In support of this, Carruthers notes that we typically do not find any thought that is separate from or prior to an item of inner speech in our introspective awareness (Ibid, p. 50). There are simply the linguistic utterances of inner speech themselves, as the experienced tokens of conscious thought. From this basis, Carruthers


goes on to argue for a dispositional higher-order thought theory of consciousness, such that consciousness itself is linked to mental content being available to language-based reflection.

Although I agree with much of what he establishes regarding language and conscious thought, I do not follow Carruthers’ broader theory of consciousness in general. This topic is far too complicated to address here, however, so I will carve out the relevant content with a qualifica-tion: As I see it, the plausible route is to limit Carruthers’ account to conscious propositional thought, rather than to make a claim about all conscious thought in general. While we may indeed have conscious thoughts in various forms in the absence of language, our conscious propositional thoughts form a subset of language-dependent conscious cogitations. In accordance with standard philosophical usage, I use the term ‘propositional thought’ here to refer to those thoughts that make a structured assertion about something, as opposed to thoughts that are simply of things. For instance, suppose that you are looking at an apple. The mere awareness that you have of the apple is not a propositional thought, and can occur entirely outside any linguistic usage. However, if you have a thought that the apple is ripe, this would be a proposi-tional thought. It contains propositional content that makes an asser-tion about the apple, predicating a property to it and thereby taking your thought beyond perceptual awareness of the apple to a structured propositional assertion about it.

With this distinction, we can limit the domain of Carruthers’s claim. Rather than saying that language is constitutive of conscious thought across the board, we can say that it is constitutive of conscious propo-sitional thought. With this more refined and limited claim, one can grant that basic perceptual awareness and other modes of conscious-ness (e.g., emotions, memories) occur outside of language. Only our conscious propositional thoughts, through which we reflect upon our experiences to think that such and such is or is not the case, essentially involve language. In other words, inner speech is a kind of conscious thought, among perhaps a variety of others, that enables us to entertain propositional assertions.

In a 1998 article, Carruthers does in fact defend the claim that language is essentially involved in conscious propositional thought (1998). 2 Although he does not explicitly say so, I suspect that it is some-thing like the above qualification that motivated Carruthers to weaken his view in this manner. Regardless, I find this to be a much more defen-sible viewpoint than the initial, broader claim that language is required for conscious thought in general. So, let us consider Carruthers’s defense


of this weakened version of his cognitive conception of language. According to Carruthers, if we do have conscious propositional thought at all, it must be through natural language. The only alternative is to be an eliminativist about conscious propositional thought, claiming that it does not exist at all (1998, pp. 460–461). Carruthers bases his argu-ment here on the idea that to be a conscious object requires direct, non-interpretive access. In other words, the point is simply that a conscious thought is itself immediately present to awareness. But, if our inner speech is merely expressive of propositional thought, rather than being the thought itself, then propositional thought is medi-ated through an interpretative mechanism and, therefore, not itself a conscious object. So, if our inner speech is not constitutive of proposi-tional thought, then propositional thought itself cannot be a conscious object. The implication is that propositional thought, by default, lies outside our conscious awareness if it is not itself constituted by the inner speech that we directly experience. So, if we want to be realists about conscious propositional thoughts, taking our introspective awareness of inner speech as an awareness of our own conscious thoughts, it must be conceded that such thought is itself constituted by inner speech. More simply, the point is that the thoughts that consciously occur to us in inner speech must be composed of the inner speech itself; otherwise, we are not aware of our conscious thoughts at all.

A simple example will help illustrate the plausibility of this perspec-tive. In consciously noting that there is a computer screen in front of me as I write this, I am aware of the linguistic label ‘computer screen’ when the thought actually occurs to me. This inner speech act is intrinsic to my conscious awareness of the thought, and nothing else is ‘there’ in the thought besides the inner speech utterance itself. To be clear, this is not at all to say that my awareness of the computer screen is constituted by the linguistic tokening of the words ‘computer screen’ in my mind. I can be consciously aware of its shape, colors, location, etc. as perceptual objects, without uttering anything to myself. But, to think about the computer screen propositionally, to consciously think that the object I perceive is a computer screen, there are linguistic items that are intrinsic to my conscious awareness of the thought, namely the assertoric label ‘computer screen’. Moreover, such inner speech seems especially promi-nent when consciously reasoning about the computer screen. When I note that there is a smudged fingerprint in the bottom left corner of the screen and consciously infer that I must have touched the screen at some point, these very words resound in my immediate awareness, and nothing else is present in my conscious entertainment of the thought,


as an interpretive propositional inference about what I see. I simply cannot imagine such a line of reasoning, as a conscious object in my own awareness, without such linguistic occurrences. Indeed, the only image of the thought available to me is the phenomenology of the inner speech act itself. So, when I reason, think, reflect, or speculate, whether in regard to the ‘external’ world or the contents of my own mind, the cognitive activity seems essentially embodied in language. I think in and with words.

I am guessing that you, too, engage in inner speech as the embodi-ment of your conscious thoughts. If not, though, bear with me. I will get to some further considerations and potential objections shortly. For now, however, I find the above considerations sufficient to conclude with Carruthers that the words embodying our conscious propositional thoughts simply are the thoughts themselves and, consequently, that our awareness of our own inner speech constitutes our awareness of our own thoughts. It is when we reflect and reason about the world through formulating propositions about it that language becomes a central component of conscious thought, and it is through the occur-rence of such language in inner speech that we become consciously aware of what we are thinking.

In addition, conscious propositional thought through inner speech can enable conscious thoughts about thoughts. This point brings us to one of the most interesting and, for our purposes, important aspects of the rela-tionship between language and conscious propositional thought, where language serves a recursive function and provides a basis for introspective thought through reflective reasoning. Daniel Dennett offers an intriguing ‘Just So’ story that can serve to introduce this function of language:

Consider a time in the history of early Homo sapiens when language – or perhaps we should call it proto-language – was just beginning to develop. ... Now it sometimes happened, we may speculate, that when one of these hominids was stymied on a project, it would ‘ask for help’, and in particular, it would ‘ask for information’. ... Then one fine day (in this rational reconstruction), one of these hominids ‘mistakenly’ asked for help when there was no helpful audience within earshot – except itself! When it heard its own request, the stimulation provoked just the sort of other-helping utterance production that the request from another would have caused. And to the creature’s delight, it found that it had just provoked itself into answering its own ques-tion. What I am trying to justify by this deliberately oversimplified thought experiment is the claim that the practice of asking oneself


questions could arise as a natural side effect of asking questions of others, and its utility would be similar: it would be a behavior that could be recognized to enhance one’s prospects by promoting better-informed action-guidance. (Dennett, 1991 , pp. 194–195)

Although this is admittedly a highly speculative account, it illustrates what could very plausibly be a functional (or perhaps exaptationally co-opted) role served by inner speech in human conscious thought. On this story, language emerged as a means of communication. However, by turning language ‘inward’, subjects could communicate with themselves, eliciting their own thoughts into conscious delibera-tion. Language thus becomes a means of questioning and answering oneself and acquires the role of bringing one’s own thoughts to the fore, as consciously accessible objects. In this process, which Dennett calls ‘autostimulation’, the mind obtains the capacity to interact with, and reflect upon, itself without appeal to any dubious internal perceptual mechanisms, like an introspective ‘mind’s eye’ (Ibid., p. 196). Through the linguistic utterance of our own thoughts in inner speech, we become consciously aware of our own thoughts, and, in turn, we can consciously think about these thoughts. As such, inner speech provides the basis for self-reflective thought and reasoning and thus can be regarded as a central cognitive function behind our ability to introspect.

The linguist and cognitive scientist Ray Jackendoff has argued along similar lines, maintaining that language can serve the purpose of bringing thoughts, including thoughts about thoughts, to conscious awareness (Jackendoff, 1996 ). Language gives conscious form to what Jackendoff calls the ‘valuation of percepts’ (Ibid., p. 25–27). The valua-tion of a percept involves noting and labeling phenomenal characteris-tics of a perceptual experience, such as noting that a particular percept is novel or familiar. According to Jackendoff, having a language that can give conscious form to these valuations, in the form of linguistic utter-ances, enables us to consciously attend to them. For instance, uttering to oneself ‘haven’t seen that before’ or ‘that’s new’ with a novel visual experience brings that evaluative thought to the fore of one’s conscious experience, with the thought occurring in the form of the linguistic utterance itself. The experience of novelty may itself occur without any such accompanying thought, but it is only through the accompanying reflective utterance that the experience of novelty itself can become the object of thought within one’s conscious experience.

Language-based valuations can also be attended to in themselves, providing the basis for further recursive thought. For instance, I could


consciously recognize my thought about the novelty of my visual expe-rience as a thought, uttering something like ‘Interesting, I thought that was new but it’s not’. or perhaps wondering ‘Why did I think that’s new?’ to myself, thereby making the thought itself an object of thought. Jackendoff suggests that recursive linguistic valuations like this are what enable us to reason about our own reasoning. Reasoning, through the consciously attentive valuation enabled by language, can itself become an object of reasoning. In this manner, then, it is through language that we can reflect upon our own thoughts, making thoughts about our own thoughts consciously available. Consideration of this discursive process, coupled with Dennett’s illustration of the ‘auto-stimulation’ function of inner speech, highlights how language can be seen as an important constitutive element in our capacity to know our own thoughts. By labeling and conceptualizing our own thoughts through the consciously graspable medium of linguistic utterances in inner speech, we can consciously engage in thoughts about our own thoughts.

So, the linguistic utterances that occur to us in inner speech are the vehicles with which we engage in conscious propositional thought and which enable further introspective thought about those thoughts. Thinking to ourselves through the medium of language enables us to be aware of our own thoughts as they occur, and this in itself consti-tutes a kind introspective awareness, in that our own thoughts are manifested in a consciously accessible form. As argued by Carruthers, this is the only possible explanation for our being consciously aware of our own propositional thoughts; if the thoughts themselves are some-thing other than the linguistic occurrences in inner speech, then we are not consciously aware of the thoughts themselves. So, presuming that we are consciously aware of the thoughts that occur to us in inner speech, inner speech is itself the constitutive medium with which we think, and, simultaneously, with which we know our own thoughts. Moreover, following Dennett and Jackendoff, inner speech can serve a recursive function, allowing us to not only be aware of our own propo-sitional thoughts, but to also engage in conscious thought about those thoughts. As I will explain later, these cognitive roles of language enable several different types of introspective self-knowledge.

While I agree with Carruthers, Dennett, and Jackendoff in granting these constitutive roles to language in the production of conscious thought, I think that the further inference that language provides the basis for consciousness itself, a claim all three authors make in various ways, is misguided. It seems undeniable to me that there are conscious


experiences that are fundamentally non-linguistic in nature. However, in order to have conscious propositional thoughts about such percep-tions, and thereby to engage in conscious reasoning about them, I think that we utilize language as a vehicle to do so. Within this narrower scope, I think that Carruthers, Dennett, and Jackendoff are right on target. Language is indeed a central component of human conscious thought, and taking notice of the roles it plays provides substantial insight into how we know (and, as we will see later, fail to know) our own thoughts. This is a very significant point in understanding the nature of introspection, particularly the higher-level sorts of introspec-tion that human beings are capable of engaging. While we arguably share many cognitive abilities with other creatures, we are (apparently, at least) unique in having a natural language with enough express-ible complexity and recursive compositionality to produce an endless variety of assertions. It is this capacity that enables us to consciously think and reason in a propositional manner, and, in turn, to cogni-tively grasp the contents and events of our own minds.

To wrap up this section, here is a nice passage from Andy Clark, whose conception of language as a cognition-extending artifact helps illumi-nate the introspective functions of language I have outlined: 3

Perhaps it is public language which is responsible for a complex of rather distinctive features of human thought viz, our ability to display second order cognitive dynamics . By second order cognitive dynamics I mean a cluster of powerful capacities involving self-evaluation, self-criticism and finely honed remedial responses. Examples would include: recognizing a flaw in our own plan or argument, and dedi-cating further cognitive efforts to fixing it; reflecting on the unreli-ability of our own initial judgments in certain types of situations and proceeding with special caution as a result; coming to see why we reached a particular conclusion by appreciating the logical tran-sitions in our own thought; thinking about the conditions under which we think best and trying to bring them about. The list could be continued, but the pattern should be clear. In all these cases, we are effectively thinking about our own cognitive profiles or about specific thoughts. This “thinking about thinking,” is a good candi-date for a distinctively human capacity – one not evidently shared by the other, non-language-using animals who share our planet. As such, it is natural to wonder whether this might be an entire species of thought in which language plays the generative role – a species of thought which is not just reflected in, or extended by, our use of


words but is directly dependent upon language for its very existence. Public language and the inner rehearsal of sentences would, on this model, act like the aerial roots of the Mangrove tree – the words would serve as fixed points capable of attracting and positioning additional intellectual matter, creating the islands of second-order thought so characteristic of the cognitive landscape of homo sapiens (Clark, 1998 ).

Objections and clarifications

One common objection to the view that we think with language begins with the idea that the same thought can be expressed in different languages, or differently in the same language. For example, it seems that if a person utters to herself ‘Ich liebe dich’, she is having the same thought as she would if she had uttered ‘I love you’ in the same context. However, if, as I have argued, conscious propositional thought is consti-tuted by natural language, then it seems to have the mistaken conse-quence that different linguistic items constitute different thoughts. Tokenings of ‘I love you’ and ‘Ich liebe dich’ are, in essence, different thoughts. Some might take this to be a reductio ad absurdum of the view I defended above.

While I am committed to different linguistic utterances being different thoughts, I do not think that this is as problematic as the above objection suggests. In response, I think it is important to note that what makes different linguistic tokens seem to express the same thought is that they share semantic content. Different thoughts can be semantically related or even synonymous, while still being individually different thoughts. In other words, the individual utterances of ‘I love you’ and ‘Ich liebe dich’ can each be regarded as a single token thought, while still recognizing that they both fall under the same utterance type, in regard to their shared semantic content as a thought expressing love for another. Taking note of this token/type distinction adequately accounts for the understandable, but mistaken, notion that different linguistic items can express the same thought, rendering the fact that different linguistic items constitute different thoughts unproblematic. Quite simply, ‘I love you’ and ‘Ich liebe dich’ are two thought tokens of one thought type.

Moreover, it is worth reiterating the fact that there is likely more going on in one’s conscious awareness than the conscious propositional thoughts that one entertains. In the case of expressing love to someone,


the person likely has some emotion through which that love is experi-enced, and this can happen without any linguistic events taking place at all. This conscious emotion may even have a role in the semantic content of the person’s thoughts, but it itself does not provide the prop-ositional structure of a conscious propositional thought, such as ‘I love you’. On my view, it is only when the person has the conscious proposi-tional thought that she loves the other person, whereby she consciously entertains a proposition about her experience via a particular linguistic utterance, that language plays a constitutive role in the thought. In other words, to say that the conscious propositional thought ‘I love you’ is cognitively manifested through the linguistic utterance of those very words is not to suggest that the conscious experience of that love, or the full semantic content of the thought, is reducible to the linguistic utter-ance alone. I suspect that many who oppose cognitive conceptions of language do so because they reject the reduction of human experience and mental content to linguistic phenomena alone. By offering my somewhat more nuanced account, whereby only specific domains of thought are identified with language, such opposition can be set aside, so that the genuine cognitive contribution of language, especially as it pertains to our introspective abilities, can be acknowledged.

Another potential objection is concerned with the appeal to intro-spective, first-person experience as evidence of how we actually think. The line of reasoning I have given above, from the initial perspectives of Frege and Merleau-Ponty through Carruthers’s argumentation and my elaboration upon it, rests to a large degree upon what appears in our own first-person conscious experience. However, it might be ques-tioned, what basis do we have for trusting such first-person experi-ence as evidence for what is really going on in our minds? Could we not just be wrong or misguided in the introspective assessment that we think through inner speech? Given the fallibility of conceptual thought in general, it seems to be at least conceptually possible that the inner speech I have argued to be constitutive of conscious propo-sitional thought is really just a surface phenomenon (epiphenomenal, if you will), a reflection of deeper unconscious thought processes in the human mind. As I understand him, Jerry Fodor holds something like this view, maintaining that thought is semantically prior to natural language, via a more basic, innate language of thought (Fodor, 2001 ). The key idea here is that thought does occur in a language, but not natural language. Natural language is a tool that expresses or communi-cates thoughts, not something that is essential to them. If this is right, then the introspective claim that inner speech constitutes propositional


thought is illusory. Our inner speech is not constitutive of propositional thought; it is merely an expression or reflection of it. To put this point in the context of Carruthers’s anti-eliminativist argument for conscious propositional thought, it is an acceptance of the idea that we may not be conscious of our own propositional thoughts as they actually occur in our minds. Perhaps in our inner speech we are only conscious of a mediated production of our propositional thoughts, and not the thoughts themselves.

This is a possibility that I cannot rule out with conceptual certainty. However, I also cannot rule out with certainty that I am really just a brain in a vat, connected to a virtual reality machine that merely simu-lates my body and my experiences. In terms of pure conceptual anal-ysis, these are possibilities that we cannot eliminate, as epistemologists in the fallibilist tradition have come to understand. However, beyond acceptance of fallibility, there are some further considerations that can illustrate the plausibility of my viewpoint and save it from rejection via the somewhat skeptical stance described above. Let us grant (for the sake of argument, if for no other reason) that there are proposi-tional thoughts that occur in our minds through an innate language of thought, outside of our conscious awareness. Even if this is the case, we can still plausibly regard the linguistic utterances that occur through inner speech as propositional thoughts as well.

First of all, there is no reason to think that these two kinds of propo-sitional thought processes could not coexist, and perhaps even interact with one another across the full spectrum of human thought. The human mind is a complicated, multi-faceted device, and can arguably contain multiple modes of thinking, both conscious and unconscious. Considering this, we need not draw a mutually exclusive dichotomy between these two kinds of thought. We could have both unconscious and conscious propositional thoughts occurring in our minds, through different mediums, via our massively complex and parallel-processing minds/brains.

Secondly, and more importantly, it would be very implausible to think that the linguistic occurrences in inner speech are not them-selves thoughts. Acceptance of such an idea would undermine our general sense of conscious rational agency, whereby we explicitly engage in conscious thought and deliberation. To see this, think back for a moment upon the various thoughts you have had while reading this text. Presumably, you have been consciously thinking about what you are reading, and these thoughts of yours have been occurring through inner speech, whereby you utter thoughts to yourself in linguistic form.


Now, if the eliminativist viewpoint described in the objection above is correct, all of this has to be wrong. Perhaps you have been thinking, but those thoughts would have occurred in a medium of which you are entirely unaware, and that which appears to you as your thoughts in inner speech is merely an inert by-product, a shadowy reflection of your genuine thoughts. In other words, despite your conscious efforts to attend to and think about what you are reading, if the objection under question is correct then there are simply no conscious thoughts happening in your mind. All actual thought is going on behind the scenes, and what you ‘think’ you are thinking, via the words you encounter in your seemingly conscious thought, is really not thought at all (I hope this is starting to sound preposterous to you, dear conscious, thinking reader!). From this perspective, what are we to make of all of those words that have been appearing to you in your inner speech? They are not thoughts, and they are not performing any other discern-able function. So, why are they there at all? To keep the conscious ‘you’ entertained while your subconscious mind deliberates? I see no plau-sible way to account for these events, without rejecting the eliminativist perspective and accepting what seem to be our thoughts as thoughts. While these concerns do not produce a decisive rejection of the elimi-nativist viewpoint, I think they do show that it would be rather diffi-cult to really believe that the ‘thoughts’ that occur to us through inner speech are illusory, while all the real cognitive work is being done behind the scenes, so to speak, in an unconscious language of thought. To be clear, I am not denying that some, or perhaps even most, of the cognitive mechanics behind human thought happens on an uncon-scious, or even subpersonal, level, nor am I suggesting that the appear-ance of thoughts in inner speech provides us with entirely transparent and infallible awareness of our own thought processes (as I will explain in the next section, there is plenty of room for error here). All I am saying is that the thoughts that seem to consciously occur to us in inner speech are indeed thoughts, and that the conscious accessibility of this inner speech provides us with the ability to know what we are thinking, at least under certain appropriate circumstances.

But, even granting that inner speech is not illusory, some may still have a problem with the introspective observation that our conscious propositional thoughts are themselves linguistic. For example, in his recent essay ‘You Don’t Know How You Think: Introspection and Language of Thought’, Edouard Machery claims that ‘The introspective fact of inner speech cannot be evidence that our conscious thoughts are linguistic’ (2005, p. 473). Machery bases his view on a distinction


between vehicles of thoughts and contents of thoughts, claiming that inner speech is an imagistic representation of the content of our thoughts, and is not necessarily the vehicle of our thoughts. He illus-trates this point through an analogy with visual images:

Nowadays, not many people are willing to infer any property of the vehicles of these visual images from the properties of their content. For example, it is widely recognized that the visual image of a red apple does not have to be red: true, the visual image represents the property red , but that fact does not support the claim that the vehicle of this image also has that property. ... In short, the sentences that are uttered in inner speech are represented by conscious images. They are part of the content of these thoughts. Now, generally, the fact that the content of some thoughts possesses a property P does not license per se the attribution of P to their vehicles. Hence, the linguistic nature of the sentences uttered in inner speech cannot be attributed to the thoughts themselves. (Machery, 2005 , pp. 475–477)

So, the basic claim is that inner speech is like the phenomenal experi-ence of a color, in that it is an auditory image, and just as the experience of the color red does not reveal to us how we actually perceive color, the experience of inner speech tells us nothing about how we actually think.

This is an intriguing and initially compelling argument, but it fails to dismiss the view that we are consciously aware of our own thoughts through the language presented in inner speech. Now, it is true that inner speech involves auditory images. I take this as an obvious phenom-enologically observable fact, but, as Machery points out, there are also neurological studies showing that inner speech is manifested in the same brain areas that process language in inter-personal communication, where we physically hear speech through auditory perception (2005, p. 476). However, there are important differences between the imagistic experience of language and the perceptual experience of color images, such that Machery’s analogy between these two phenomena does not hold up. In the case of language, the content is encoded in the vehicle in a much different way than occurs in our ordinary sensory modali-ties. A simple example should sufficiently illustrate this: I can convey my propositional thoughts to you through language, but there is no vehicle with which I can convey to you my experience of red. Somewhat remarkably, we can grasp the propositional thoughts of other people through language, but there simply are no analogous cases regarding our


imagistic sensations, such as our experiences of sound and color. The transference of thought through language could not be possible unless the content of a thought is encoded in a discernible vehicle, which, of course, consists of the particular linguistic utterances that we under-stand when grasping a propositional thought through language. So, the vehicle and the content are intimately and constitutively connected together in the case of language, unlike perhaps anything else in our experience. For this reason, Machery’s analogy between language and color perception breaks down, and the distinction between vehicle and content that he draws from general imagistic perception does not apply to language. In the case of language, we can justifiably identify partic-ular linguistic occurrences as the vehicles that encode the contents of thoughts. Otherwise, we could not feasibly account for our ability to understand the content of a linguistic utterance, whether it be a token of inner speech or speech from another person. So, we are consciously aware of our propositional thoughts through inner speech because the language of inner speech is a content-encoding vehicle. It is worth adding that this does not imply that we know the cognitive mechanics of how such thoughts happen, which, arguably, is a legitimate target of Machery’s argument. There are many complicated processes at work in language comprehension, most of which fall under the radar of our conscious awareness, but the point that we know what we are thinking via our inner speech utterances still stands.

Another potential objection to the view that language is constitu-tive of conscious propositional thought stems from concerns about explaining the intentional character of language and thought. For instance, John Heil has argued that the various positions typically held by philosophers on the relationship between language and thought are explanatorily defunct. Whether language or thought is given priority, Heil argues, the intentionality of both is left rather mysterious and unexplained, the intentional character of one being merely shifted over to the other (1988). On these grounds, someone like Heil might complain of my account that it really does not get us anywhere in understanding how it is that we think about things, including our own thoughts. It is the intentional character of language and thought, the fact that they are about things that needs explanation, so claims simply about the relationship between language and thought tell us nothing truly informative.

I must admit that I am somewhat sympathetic to Heil’s complaint here. However, explaining the intentionality of thought and language is not the task that I have undertaken here. My purpose has been to


show that language is constitutively involved in conscious proposi-tional thought, and to illustrate how this facilitates certain kinds of introspection, rather than explaining what language and/or thought are in themselves, qua their intentional character. In response to Heil, and in illustration of my sympathies to his position, I want to show how I see my account as being complementary to Heil’s proposed endeavor of explaining the intentionality of language and thought together in naturalistic terms. As I have mentioned, I see language and conscious propositional thought as essentially related elements of human cogni-tion. Rather than one taking priority over the other, I think that they are, in essence, elements of one and the same thing. Conscious prop-ositional thoughts just are the tokenings of linguistic items in inner speech. Considering this, the possibility of explaining the intention-ality of language and thought together becomes a viable enterprise, which is exactly what Heil calls for. So, while my account admittedly has nothing to offer in explaining the intentionality of language and thought, I think it does pave the way for such an explanation to be given. It allows us to group important kinds of language and thought together, so that an explanation of their intentionality can be inves-tigated in terms of explanatory factors that are external to both. Of course, this is a project that extends well beyond the concerns I have addressed here. Nevertheless, the accord that my position has with such possible avenues of explanation and understanding should not be over-looked. Ultimately, I think that these views could potentially coalesce into a powerful and broad-reaching understanding of language and mental phenomena together. For the time being, however, the point to take home is that language and conscious propositional thought, however they may instantiate intentional content in the mind/brain/world nexus, can be understood as intrinsically united aspects of our ability to consciously think about things, including our own minds. Identifying the constitutive link between language, in the form of inner speech, and conscious propositional thought provides us with an insightful vantage point from which we can understand how we know our own thoughts and think about those thoughts.

A variety of additional objections and concerns could be raised regarding the relationship between language and thought, but I simply cannot address them all here. However, I would like to highlight some general points about my account that might allay some additional concerns that I have not explicitly addressed. Again, it is important to note the fairly narrow domain with which I am concerned. I have not made any general claims about the nature of language and thought


altogether, but have made the more specific claim that natural language provides a basis for conscious propositional thought. There are argu-ably many other domains of human thought to which my claims do not apply. Moreover, even within this narrowed context, the claim I make is somewhat weak, in comparison to what might be said about the matter. Essentially, all that I have claimed is that when we are engaged in conscious propositional thought, the linguistic utterances occur-ring in our inner speech constitute the thoughts we have. As such, this mode of thought enables us to be consciously aware of our own propo-sitional thoughts, and when recursively applied, to consciously think about our own thoughts. This does not exclude the possibility of other kinds of thoughts feeding into the content of these conscious proposi-tional thoughts (such as memories, emotional experiences, and so on across the broad palette of human thought and experience), nor does it exclude the possibility of various additional cognitive processes under-lying our ability to engage in such thought. Considering these things, my view should be acceptable to all but the most adamant opponents of granting a cognitive role to natural language. In any case, even if you do not concede to the view I offer, you may still be able to draw upon the further claims I make about the nature of introspection, by replacing the functions I grant to language with whatever cognitive mechanisms you think actually serve in those capacities (e.g., a Fodorian language of thought). With that said, let us turn to consider the epistemic attributes of inner speech, as they pertain to our abilities to know (and fail to know) our own minds.

Kinds of self-knowledge enabled by inner speech

As I see it, the introspective knowledge enabled by inner speech is multi-faceted and forms the basis of three distinguishable types of self-knowledge. First, the experience itself of undergoing inner speech provides an immediate kind of phenomenal knowledge of our thought processes, which we acquire in virtue of the experience itself. There is a qualitative character to the experience of inner speech, such that one knows what it is like through its occurrence. As discussed earlier in Chapter 4 , this type of knowledge is unique to the first-person perspec-tive, in virtue of the experiencing subject embodying that which is known. Conceptually, this may seem like a rather odd sort of knowl-edge, since it is phenomenal rather than propositional in nature and yet that which is known is itself a propositional thought. One might regard this as contradictory, but it is crucial to note that this is only one type of


self-knowledge made available by inner speech, which simultaneously overlaps with other types of self-knowledge that have propositional content. With the understanding that self-knowledge is multi-faceted, noting the additional kinds of self-knowledge described below that accommodate propositional content, the seemingly contradictory nature of this kind of self-knowledge can be rendered unproblematic.

Before we move on to the propositional kinds of self-knowledge, however, some further observations regarding the phenomenal knowl-edge of inner speech are in order. The phenomenal experience of inner speech consists of auditory imagery, which is perhaps its most frequently noted phenomenal quality, but it also includes the experience of inten-tional expressive articulation, in the act of saying something. We saw this earlier in the passage from Merleau-Ponty, where he describes language as an expressive activity wherein we present our thoughts to ourselves in speech (1945/2002, p. 206). We experience not only the auditory image of hearing one’s voice but also the act of producing speech. It is interesting and revealing to note here that subtle sub-vocal muscle movements in the mouth and throat correspond to the activity of inner speech, as if one were actually saying something aloud in an attenuated manner. In fact, interestingly, the detection of the subtle body movements produced during inner speech is involved in ongoing research toward new ‘silent speech’ interface technologies that may serve a variety of communica-tory functions, from speaking into a cell phone without saying anything aloud to enabling people with certain speech pathologies to communicate more effectively (Denby et al., 2009 ). Considering the phenomenal expe-rience of expression coupled with the embodied movement of speech, it is appropriate to understand inner speech as something we enact, rather than something that we merely observe, in a passive manner.

Noting the expressive character of inner speech is important because it helps us understand again how introspection is not a kind of inner perception, as previously discussed in the first few chapters. We should not be tempted to regard inner speech as a second-order inner sense mechanism, dedicated to the task of detecting one’s own mental states, analogous to the inner eye we have already poked out. The experience of inner speech is not the perception of one’s voice in one’s head, such that an inner ear somehow detects one’s thoughts and renders them observ-able as objects. There is auditory phenomenology that is coupled with the active experience of saying something, of course, but this auditory imagery is a first-order experience of saying something with one’s voice rather than a second-order perception of an inner event. This point is backed up further by the fact that the brain areas involved in ordinary


external speech perception appear to be the same areas active with inner speech as well (e.g., McGuire et al., 1996 ). Thus both phenomenology and neuroscience converge to show that the introspective phenomenal knowledge we obtain through inner speech is acquired in virtue of embodying the expressive act of saying one’s thoughts, rather than an internally-directed perception of one’s thoughts.

Moving into knowledge with propositional content, the second kind of self-knowledge provided by inner speech is produced through the second-order function of taking our own experiences and other mental phenomena as objects of thought, such that they can be conceptually grasped in propositional terms. I leave it as an open possibility that there could be other modes of conceptual thought besides thought based in language, but nevertheless language significantly contributes to our ability to engage in conscious conceptual representation and is the medium with which we consciously conceptualize our own mental states in propositional terms. For example, suppose that while on a diet you attend a birthday party where your favorite kind of cake is being served. Upon seeing the cake, you experience a desire to eat a piece of the cake, but then think to yourself, ‘I really want a piece of that cake, but that would go against my diet’. In having this thought as an occur-rence in inner speech, you utilize the phrase ‘want a piece of cake’ to conceptualize your experience of desire, thereby making it an object of conscious propositional thought (which could, in turn, play a role in you overcoming the desire and accordingly controlling your behavior). It is through linguistic utterances like this in our everyday inner speech that we consciously engage in thought about our own mental lives. We use language as a tool to conceptually categorize, understand, and think about what is happening in our own minds, thereby developing a second-order domain of self-knowledge in which we can consciously know about our own minds in a propositional manner (e.g., knowing that we are in mental state X). This is fundamental to our capacity to obtain introspective knowledge of our own mental states.

It is worth noting that this is where language-based introspection meets up with the conceptually mediated modes of introspection discussed in the past two chapters. The conceptualization of our experi-ences occurs through the labeling and structuring of the experiences into propositional content, which occurs through the medium of language. In this respect, inner speech embodies the application of interpretive conceptual structures that help us understand our own mental states as particular mental states of some type or other. For example, in the case of reflecting upon a desire for cake, the inner speech locution of ‘want a


piece of cake’ embodies the folk psychological conceptualization of the desire as a desire (‘want’) and enables one to consciously reason about that desire as it pertains to one’s beliefs and other desires, resulting in the reflective capacity to, for instance, think about how it conflicts with one’s desire to lose weight, vis- à -vis the background belief that the cake is full of empty calories, etc. Considering this, inner speech may be plausibly regarded as a medium for folk psychological introspection, perhaps along the lines of Hutto’s treatment of folk psychology in terms of narrative practices (2009), as mentioned in the previous chapter.

Kicking the processes up another level of reflection, the third type of self-knowledge enabled by language pertains to the recursive capacity to consciously think about our own conscious thoughts. Not only can we conceptualize our mental states, as described above, but we can also conceptualize those conceptualizations, such that propositional thoughts themselves become the object of propositional thought. This can generate a kind of recursive self-knowledge of one’s own conscious thoughts themselves. Continuing with the cake example, suppose that your thoughts about your conflicted desires (wanting the piece of cake vs. wanting to lose weight) were to prompt you to further deliberate about your broader goals and values in life, as second-order beliefs and desires about your beliefs and desires. You think, for instance, some-thing along the lines of ‘But what’s more important, my desire to lose weight or my desire to enjoy life in the moment?’ thereby engaging in further higher-level reflection concerning your overall beliefs and desires about how to live life, perhaps deciding to eat the piece of cake after all. Through this example, we can see how inner speech locu-tions are utilized to engage in conscious propositional thought about thought, bringing us to the significant capacity of the human mind to engage in reflective deliberation about its own states, whereby we can know about ourselves through abstract representation of and conscious thought about our own thoughts. As I have suggested, inner speech is the cognitive basis for extending introspective knowledge into these multi-level and recursive domains of conscious thought. Through such language-enabled thought, we can consciously reflect upon the content, processes, and nature of our own minds, thereby developing a multi-faceted conceptual understanding of who and what we are.

From self-determined truth to self-deception and back

In light of these types of self-knowledge enabled by inner speech, we can now consider some more specific epistemic qualities of inner


speech, in terms of what propensities it has to produce true or false beliefs about ourselves. As I see it, the epistemic territory is quite diverse here, ranging from utterances that determine their own truth values, simply in virtue of their structure and content, to self-deceptive narra-tive rationalizations that produce beliefs about oneself.

Let us begin with the kinds of utterances that determine their own truth values. Consider the simple event of thinking ‘I am thinking’ to oneself. Doubts about the existence of a self as the referent of ‘I’ aside, it is reasonable to regard this thought as self-determinably true. The thought simply makes itself true, in virtue of the fact that it itself is a thought that satisfies its truth conditions. There are philosophers who have attempted to get some conceptual heavy-lifting out of such thoughts, perhaps most notably Descartes’ cogito, arguing that privi-leged first-person authority and/or infallibility is exhibited in such cases. However, as you might guess, I am skeptical of such claims. To see why, compare ‘I am thinking’ with the similarly self-referential ‘This is a sentence’. Asserting ‘This is a sentence’ is likewise self-determinably true. Through its assertion, the sentence is made true. A speaker is infallible in asserting it, but this is merely a relic of its self-referential structure and, crucially, not an indication of privileged epistemic status on the part of introspection. Considering this, the construction of self-referential and self-affirming assertions about thought do not necessarily establish a privileged epistemic position in introspection. 4

There are other types of utterances of a self-determining nature that arguably reveal a more interesting epistemic character regarding intro-spection. Suppose, for example, that when you decided to eat the piece of cake discussed earlier, you said something like ‘Ah, to Hell with it, I oughta just eat it and enjoy myself!’ as the culmination of your inner speech deliberation. In doing so, the occurrence of that thought is itself constitutive of the content it expresses, as the embodied articulation of your attitude toward the event at that moment. But what is the truth value of the sentence? Is it true or false? A case could be made for either option, but I am inclined to say that it does not have a determinable truth value at all. In saying that you ought to eat the cake, you are expressing an attitude, representing a normative judgment that may or (more likely) may not be in your best interest, but there is no clear deter-minable fact of the matter. The self-determining nature of the inner speech utterance in this case is not primarily a matter of saying some-thing true or false (though as a proposition, it structurally suggests that it is) but rather is an instance of expressive self-constitution. It is the self-determining event of an agent thinking out her normative attitude


toward the situation, rather making an assertion about it. Considering this, we can see that some utterances in inner speech exhibit an epis-temic character that does not match up with the standard epistemic paradigms used to evaluate typical propositional knowledge, i.e., in terms of evidential support, reliable perceptual mechanisms, and so on. When you utter to yourself, ‘I oughta just eat it’, you are not uttering a proposition that is true in virtue of mapping on to some state of affairs that independently exists apart from the agent’s normative attitude at the moment. Instead, the utterance is an expression that itself embodies the agent’s normative judgment.

This leads us into the murky territorial overlap between human agency and the epistemology of self-knowledge. Insofar as we have some kind of genuine agency through which we actively engage in the determination of our own mental states, knowing ourselves is less like factual discovery and more like a self-constitutive, constructive process performed by an intentional agent. This expressivist aspect of language has been taken up elsewhere in more developed and nuanced ways than I can offer here, so I will not delve into the details. 5 Suffice it to say that in a certain range of cases, the conscious thoughts that we have about our own minds may exhibit unique epistemic qualities that set them apart from other statements with propositional structure.

However, acknowledgement of this sort of self-determining intro-spective process is not to say that we are uniquely privileged in the various propositional thoughts we think about ourselves. In fact, as we have seen in the case of self-applied folk psychology, our propositional assertions about ourselves arguably extend into fallible truth-evaluable content, such that uttering a false propositional statement about oneself is at least possible in principle. As we have seen in the previous two chapters, the kind of self-knowledge where we engage in concep-tual representation of our own mental states is mediated by processes that could conceivably go wrong, thereby making room for epistemic error. When we conceptually represent our own mental states as being a certain way through specific linguistic descriptions, there is a gap between that which is represented and the representation of that thing, such that the possibility of misrepresentation arises. Suppose, for example, that you are one of the unwitting subjects in Nisbett and Wilson’s study on self-reports ( 1977 ), as discussed in the previous chapter. You are out at a shopping mall and come across (what appears to be) a panty hose customer survey booth. You try to simply walk past it, but the booth attendant insists that you look over four pair of panty hose laid out on a table and pick which pair you like the best. You casually inspect


each pair, from left to right, and while doing so you note various char-acteristics, thinking such things as ‘feels soft’ and ‘seems very elastic’. You pause for a moment after inspecting all four pair and say, ‘I like D the best’, pointing to the pair on your far right. When the attendant asks you why you chose that pair, you think about the desirability of its smooth texture and then respond, ‘Because it’s the smoothest’. The attendant then asks you whether the position of the panty hose had any impact on your choice. You think for a moment, ‘Hmm the posi-tion ... I only remember the smoothness’ and then answer, ‘No’. In this case, your conscious thoughts in inner speech represent your choice as being based upon qualitative differences among the pairs of panty hose. However, in this particular case, the panty hose are all qualitatively the same. Moreover, as Nisbett and Wilson found, there was a significant position effect, such that a statistically significant number (40 per cent) of the participants chose pair D on their far right (and another 30 per cent chose pair C, in the mid-right position). So, it is quite plausible to think, from an objective third-person perspective, that your choice was produced through some tendency to choose items on the right. Perhaps, for instance, you are right-handed and have a propensity to be drawn toward things to your right, despite a lack of conscious recog-nition of this propensity. The point here is not necessarily that this was in fact the causal basis of your choice, but rather only that it is conceivable that such is the case. It is quite possible that your conscious representation of your choice, through the inner speech you engaged in while going through the process of choosing a pair of panty hose, was mistaken and did not accurately reflect the mental states that actually produced your choice. In short, your conscious propositional thoughts about your choice, as embodied in your inner speech, can be fallible confabulations that do not accurately reflect the actual cognitive proc-esses at work in your decision.

Evidence of such fallibility can be inferred from outside the phenomena of inner speech and self-knowledge, among studies of the accuracy of verbal reports in general. Psychologists have identified a phenomenon called ‘verbal overshadowing’, in which the soliciting of verbal reports can significantly decrease memory accuracy (Schooler & Engstler-Schooler, 1990 ; Meissner & Memon, 2002 ). For instance, Schooler and Engstler-Schooler found that observers of a video robbery scene who were subsequently asked to write down detailed descriptions of the robber’s face were much less likely to accurately identify the robber’s face at a later time than were observers who did not engage in any overt verbalization of the event (with 38 per cent accuracy among


the verbalizers and 64 per cent accuracy among the non-verbalizers). Similar results have been found in numerous other studies, providing strong evidence that the use of language can actually impede our ability to accurately access our own mental states, and may actually impede the cognitive processes involved in the formation of our mental states, such as our memories. Although the subjects in such studies engage in overt verbalization, it is reasonable to assume that similar effects can occur in our inner speech commentaries upon the contents and attributes of our own minds, implicating the potential fallibility of inner speech in its introspective abilities.

With these considerations, it should be sufficiently clear that inner speech can lead us to make mistakes about the nature and content of our own minds. It should be added, however, that it is not yet clear to what extent the verbalization of our mental states and processes has a tendency to produce inaccurate results. Although there are fairly robust results demonstrating decreased accuracy in mental state attributions among verbalizing subjects, there is also some significant evidence suggesting that verbal reports can be reasonably relied upon in some contexts. Most notably, Herbert Simon and K. Anders Ericsson have developed protocol analysis methods designed to avoid the misguiding tendencies of verbalization (Ericsson & Simon, 1984 ; Ericsson, 2003 ). Setting aside some of the nuanced details, the main idea is to solicit verbal reports with very specific directions to verbally state one’s thoughts as they occur in real time. Under such circumstances, Ericsson states that:

subjects can think aloud without any systematic changes to the sequential structure of their thought processes. The fact that subjects must already possess the necessary skills for efficient verbalization of thoughts is consistent with extensive evidence on the acquisition of self-regulatory private speech during childhood and on the spon-taneous vocalization of inner speech by adults, especially in noisy environments. (Ericsson, 2003 )

Ericsson and Simon’s empirical work on their protocol analysis method has confirmed it as a valid experimental approach, which has become widely acknowledged and utilized by mainstream psychological research. So, although verbal thought is fallible, there is some reason to think it can be trusted in certain contexts.

This leaves us with a significant need for more work in identifying and assessing the epistemic qualities of inner speech. Considering the


current status of empirical data and conceptual distinctions that pertain to the epistemology of inner speech, as I have outlined above, it is clear that it does not admit of any singular characterization. I regard the acknowledgment of this point as progress, but it is really just a begin-ning, identifying a need for a broad range of comparative philosophical and scientific inquiry into the epistemic strengths and weaknesses of our diverse introspective capacities enabled through language, as well as the implications that they have in our lives. I hope that the work I have provided here may serve as a foundational framework for this sort of inquiry, such that our understanding, and misunderstanding, of ourselves through the incessant monologues we find in our minds can itself be more adequately understood.

Before ending, however, I want to offer some more speculative consid-erations that extend the epistemic dynamics of inner speech further. In regard to the fallibility of our inner speech, we can extend this further toward the possibility that inner speech can, in some cases, lead us into self-deceptive states, where we engage in narrative rationalization of our own mental states in such a way that we not only confabulate false attributions but also mitigate or cover up aspects of ourselves that could otherwise be overtly apparent. Once again, this leads into complex territory that I can only hazily outline. 6 Nevertheless, it is still worth considering here the possibility of self-deception through inner speech, which I will illustrate with an example. Imagine the case of a married man who becomes friends with an attractive co-worker. Upon meeting her as a new employee on the job, he finds that they get along quite well. She is very nice to him and shares many of his interests, and over a period of a few weeks they get to know each other in a very friendly manner. Through this interaction, however, suppose the man develops a romantic attraction for the woman and finds himself thinking about her often. Suppose further that this also brings about a sense of fear in the man. He begins to feel threatened by the woman, in the sense that his attraction to her conflicts with his desire for marriage fidelity. At this point, he starts to have defensive thoughts that question the woman’s character, telling himself things like ‘There’s something devilish about her. She seems deceitful’. Suppose that, in reality, the woman is quite nice and has no ill intentions toward the man, having come to think of him as a good friend. The man, however, through the construction of nega-tive thoughts about the woman in his inner speech, eventually convinces himself that he does not like the woman. In this case, it seems reasonable to me to regard the man as self-deceived. In reality, he has established the beginnings of friendship with the woman. She is friendly to him,


and on an intuitive, visceral level that he experiences and knows in the phenomenal sense discussed earlier, he is drawn to interact with her. But, through confabulatory thoughts in inner speech, he convinces himself that he does not like her, deceiving himself about his feelings and rela-tionship. With this being a conceivable situation, we can see that our inner speech could possibly lead us to believe false propositions about our own mental states that would otherwise be readily acknowledged. In this respect, inner speech can be a mechanism for self-deception.

However, the story may not end there. We can imagine that over time he may actually grow to genuinely dislike the woman. As the man continues to engage in negative inner speech about the woman, his feelings of camaraderie and attraction will gradually subside, and he could even become less interested in the things they initially held in common. So, eventually, the beliefs that the man confabulated in inner speech may actually become true, such that he really genuinely does not like the woman and consistently behaves accordingly. With this extension of the story, we can see how what was first a case of self-deception through inner speech can become a case of self-fulfilling prophecy regarding one’s character.

In interpretation of this possibility, I suggest that the epistemic falli-bility of inner speech, as it pertains to knowledge of our own minds, is not necessarily a straightforward case of constructing false beliefs. Now, as illustrated in the various examples I have discussed, we can clearly come to have false beliefs about ourselves, but to leave the impact of inner speech on self-knowledge at that would be to miss something. Through our inner speech, we not only can come to believe false things about our own minds (or, of course, believe true things about our minds, when we get it right) but also can construct, shape, and impact our own mentality. Perhaps we partially create our own nature over time through the things we tell ourselves in inner speech. In other words, through weaving narrative stories and theories about ourselves, we may be, to some extent, creating ourselves. Some have made similar points in regard to understanding the nature of the self (e.g., Dennett, 1989 ; 1992 ), but I want to separate this point from the attempt to explain what the self is, per se. Whatever we might be, in terms of the under-lying mechanics and metaphysical nature of the self, we do think about ourselves through language in inner speech, and in consideration of examples like that above, what we tell ourselves about ourselves can impact what is true of our own mental states. In short, perhaps we can not only know and fail to know our own minds through inner speech, but also at least partially determine what there is to know about our


minds, in virtue of the fact that the thoughts we have about ourselves in inner speech can themselves become constituents of what we seek to obtain in the ever-changing pursuit of self-knowledge.

Conclusion

In this chapter, I have explained and defended the view that language, in the form of inner speech, is constitutive of conscious propositional thought. Stated succinctly, the idea is that we think in and with words. One central aspect of this view is that language brings objects of thought to conscious attention in the form of propositional language, enabling us to be consciously aware of our own propositional thoughts as they occur to us. Moreover, the language of inner speech serves a recursive function in human thought, allowing us to consciously think about our own thoughts. I have disagreed with the view that this recursive func-tion of language is the basis of consciousness itself but have defended the weakened view that it serves as the basis for conscious propositional thought and self-reflection upon such thought, which are important facets of our introspective abilities.

I have suggested that the self-knowledge enabled by inner speech consists of three distinct types. First, there is the phenomenal knowl-edge of having a conscious propositional thought, experienced as the utterances that occur in inner speech. Second, we obtain propositional knowledge or our own mental states through the use of language to conceptually represent aspects of our own minds in propositional form. Finally, we can further use language to conceptually represent our propo-sitional thoughts themselves, thereby enabling recursive self-knowledge of our own propositional thoughts about our own minds.

Through further analysis of more specific epistemic qualities of inner speech, I have noted that the terrain here is quite varied and in need of further inquiry. In some cases, our utterances in inner speech self-determine their own truth value, while in other cases, we may come to have false beliefs about ourselves. Moreover, in some cases we may even deceive ourselves through inner speech, which could in turn have a self-fulfilling impact on our mental states, such that we actually take on the mental characteristics that we initially deceived ourselves about. Considering these things, further empirical inquiry and theoretical analysis of the epistemic qualities of inner speech are clearly called for and could provide further substantial insight into how we can know, fail to know, and perhaps also create our own mental states through our internal monologues.

148

Philosophical conceptions of introspection and self-knowledge typically emphasize first-person access to one’s own mind, an orientation which we have analyzed in various ways throughout this book. More recent philosophical and scientific work emphasizes third-person objectivity, which I have also brought into the mix to provide a more accurate and well-rounded understanding of introspection, as well as to mitigate some of the over-inflated epistemic privilege that has been granted to the first-person perspective, especially in Cartesian approaches to the mind. However, self-knowledge involves second-person interper-sonal processes as well, through social interactions in which epistemic agents learn about themselves from others. The second-person stance is often not considered in philosophical and scientific work on the mind, though some have made efforts to give it more attention (e.g., Thompson, 2001 ; Ratcliffe, 2007 , pp. 152–185). This chapter estab-lishes a framework for bringing second-person social interaction into our understanding of introspection and self-knowledge. In particular, I will look at how substantial components of self-knowledge are acquired through social navigation and interpersonal testimony, via an agent’s ability to navigate her position in social environments through what others say to her. There is more to our interpersonal relationships than the social navigation and testimonial processes I will discuss here, but these aspects of social life help draw out some important kinds of infor-mation that are particularly salient to a complete understanding of the nature and limits of what we know about ourselves. In particular, considering these aspects of social cognition will help us understand and analyze how it can be rational for a person to defer to the testimony of others when forming beliefs about her own social characteristics. Sometimes the first-person perspective must give way to the second-

8 On the Social Side of Self-Knowledge

On the Social Side of Self-Knowledge 149

person, though exactly when, when not, how, and why certainly isn’t always clear. Nevertheless, considering the interpersonal dimension of self-knowledge will help round out the pluralist framework surveyed throughout this book, providing a final vantage point from the social side of things.

Social self-knowledge

As social creatures, we often come to know ourselves through the social factors that surround us in our concrete lives as human beings. We learn about various human traits through social interactions with other people, on the basis of which we can consider the presence or lack of those qualities in ourselves. More significantly, many of our traits arguably cannot be conceptualized without recourse to social factors, such as the relationships we hold to others (e.g., parent, peer, friend) and the social character traits that we may or may not have in those relationships (e.g., cooperativeness, honesty, kindness). In at least some facets of human life, social factors are fundamental. In fact, some of the cognitive processes that we have already discussed, especially language and folk psychology, are dependent in various ways upon social proc-esses. Of course, it has been long recognized in both the philosophical and scientific investigation of human nature that we are social crea-tures. We are inextricably intertwined with other people in our various endeavors in life, from immediate interpersonal relationships with family and friends to participation in larger-scale social phenomena such as the multiplicity of factors involved in economics and politics. Given the social nature of human life, then, it only seems reasonable that we ought to consider the social context of our individual lives if we are to obtain an accurate and comprehensive understanding of how we come to know ourselves.

Unfortunately, however, the philosophers and scientists who have addressed the social side of self-knowledge have tended to over-state the case, leading toward implausible forms of behaviorism and social constructionism that reduce the self to a mere confluence of social factors. Consider, for example, what George Herbert Mead said about self-knowledge:

The individual experiences himself as such, not directly, but only indi-rectly, from the particular standpoints of other individual members of the same social group, or from the generalized standpoint of the social group as a whole to which he belongs. For he enters his own


experience as a self or individual, not directly or immediately, not by becoming a subject to himself, but only insofar as he first becomes an object to himself just as other individuals are objects to him or in his experience; and he becomes an object to himself only by taking the attitudes of other individuals towards himself within a social environment or context of experience and behavior in which both he and they are involved (Mead, 1934 , p. 138).

On this account, the first-person perspective is itself dependent upon social processes, processes that are held to be the foundation upon which we come to have experience of ourselves. Self-knowledge is thus conceived as a social product, created over time through communicative exchanges with others in which we become objects to ourselves. While it is true that some important aspects of self-knowledge depend upon social phenomena, Mead’s reduction of the first-person to the social is hyperbole, errone-ously suggesting that people can only know themselves as others know them. At a minimum, it overlooks the immediate primacy of phenomenal consciousness, the uniquely first-person ‘what it’s like’ to be conscious that we considered in depth earlier. While our conceptual grasp of phenom-enal consciousness may rely upon various social factors, the experiential knowledge we have of consciousness itself is arguably best regarded as a fundamentally first-person phenomenon, with respect to our conscious experience itself and irrespective of the socially mediated concepts and labels we may apply to it. The first-person nature of such knowledge is simply left off the map with purely social accounts of self-knowledge.

This leads to an unnecessary divide in philosophical approaches to mind and self, between those who latch onto the first-person nature of consciousness and those who favor primarily social accounts of mind and self. Unfortunately, it is only the latter camp that has thus far attended to the social dimensions of self-knowledge, extending Mead’s hyperbolic reduction of self-knowledge to purely social processes. At the endpoint of this trajectory is the emergence of social constructionist accounts of self and self-knowledge. Consider, for example, the social constructionist Kenneth Gergen’s characterization of self-knowledge:

... in the case of self-knowledge, there is virtually nothing to know. That is, there is little in the way of factual data that lies waiting to be discovered, explored, or assessed. Rather than factual data, we are confronted with a series of social rules or intelligibility systems employed in the process of personal explanation (Gergen, 1991 , p. 384).


This denial of factual self-knowledge makes social approaches to self-knowledge seem more untenable than it needs to be. While there are social processes at work in the construction of self-knowledge, Mead’s and Gergen’s claims that our knowledge of ourselves is entirely constituted by those social processes themselves leads towards an erro-neous mitigation of the unique first-person orientation we hold toward ourselves.

It is important here to recognize that self-knowledge need not be reduced to either extreme, as a singular state that is either a private first-person phenomenon or a by-product of social processes. Self-knowledge is most accurately regarded as complex and multi-dimensional in nature, containing both private and social elements, as well as further rela-tional elements that emerge out of interactions between the two (folk psychological attributions of beliefs to oneself, for example). Indeed, as will be more apparent in a moment, there are genuine social facts about oneself that, despite being dependent upon the background social factors at work in human lives, play important roles in the obtain-ment of accurate self-knowledge. Whatever selves may be, they exist in social environments and any comprehensive account of self-knowledge must recognize the existence of facts about how one is situated amidst a variety of social factors. Due to the immense complexity of social structures, this chapter will not be able to fully flesh out this terrain but a few key forays into the social environment of the self will help us outline the social side of self-knowledge with enough content to see some basic ways in which we come to know ourselves through our rela-tionships with others.

Navigating one’s position in social space

As social creatures, we live in complex social environments composed of diverse interpersonal relationships. We exist in families, as children, siblings, parents, and spouses. We participate in a variety of different economic and political interactions, related to others as employees, co-workers, and bosses, as leaders, followers, compatriots, and enemies. In order to thrive in these complex and varied environments, we must of course know where we are within them. As a simple example, consider the fact that in order to be aware of oneself as a parent one must at a minimum know that one has a child and know who it is to whom one bears this relation. The social nature of parenthood is intrinsic to the knowledge of oneself as a parent, requiring recognition of oneself as being situated within a particular kind of social matrix.


Clearly, this leads us into one social facet of self-knowledge – knowing where oneself is, socially speaking. How, then, do we obtain this kind of self-knowledge?

To get an initial grasp on this issue, let us flesh out the spatial analogies hinted at above. We can think of the complex webs of social relations and structures as spatially representable environments, with individ-uals situated in spatial locations defined by their relationships to others vis- à -vis particular social domains. This conception of social structure follows the visual representations of social relationships generated by social scientists involved in social network analysis, where individuals are represented as nodes spread across a spatial array and their various relationships are depicted by linear connections between them (e.g., D’Andrea et al., 2010 ). Within this type of representational framework, the question of self-knowledge becomes: How does one know where one is located within a social environment?

Notice that this calls for a kind of indexical knowledge, such that a person’s position is pointed out to oneself in some way or another. Just as a physical organism needs to be able to indexically track its position in a physical environment in order to successfully navigate through that environment, a social being needs to be able to locate itself indexically as an entity in a particular social environment in order to become an engaged agent in that environment (Ismael, 2007 ). This is a very basic kind of self-knowledge that is not acquired through purely first-person means but rather through interaction with the external social envi-ronment via one’s interpersonal relationships. To see this, consider the analogous case of locating oneself in physical space. Suppose, for example, that by looking out a window and seeing a sunset you realize that you are on the west side of a building. Analogously, just as you know where you are physically located through perceptual observation of your surrounding environment (and not by internal monitoring of your physical body), you can come to know your position in society by observing features of your social environment, and not by reflecting upon your internal mental states. For example, a person might acquire insight into her character as a friend when someone confides in her about a personal matter he is facing, another might learn some-thing about his emotional disposition toward others when his spouse points out to him that he appears irritated around certain people, an instructor might realize that she is operating as an effective educator when a group of students demonstrate new understanding of a topic she is teaching, and an employee might come to know her relative position and status in a business by being told to do a particular task by another.


All of these cases of self-knowledge are socially acquired. They exhibit a genuine type of self-knowledge that is intrinsically social in nature, emerging out of social interactions that reveal one’s location within social environments.

This type of self-knowledge is not a merely subjective interpretation constructed by projecting the third-person gaze of others onto oneself, nor is it something that one can know about oneself through concep-tual reflection. It is knowledge that exists in virtue of our concrete factual existence as social beings and, moreover, which may or may not be acquired, depending upon whether or not the epistemic agent is able to accurately pick it up from the surrounding environment. There are, for instance, facts about whether or not a person is operating as an effective educator or sympathetic friend on a particular occasion, objective facts that may or may not come to be known by the person to which they apply. An instructor might, for instance, be mistakenly feeding her students false information as profound truth, in which case she would be epistemically amiss if she were to believe that she is acting as an effective educator on that occasion. Alternatively, however, she might be genuinely conveying cogent insight to her students in a detectable manner, in which case she could be epistemically warranted in believing that she is an effective educator. Consider also familial rela-tions, such as being a spouse, or economic circumstances, such as being in poverty . There are facts about such things, facts that we can come to know about ourselves by navigating through our social engagements with others. Such knowledge constitutes a distinctly social type of self-knowledge, knowledge of facts about one’s own social positions and characteristics. This is a component of self-knowledge that has been ignored by some and epistemically mitigated by others, but which must be included in any accurate and comprehensive understanding of the nature of self-knowledge.

Direct and indirect testimony about oneself

A complete explication and defense of the existence of social facts and their legitimacy as objects of self-knowledge cannot be provided here, but the above considerations sufficiently illustrate the vast domain of social facts that people can come to know through the navigation of their positions in society. Let us turn now to consider the source of this kind of self-knowledge, fleshing out in some detail the kind of informa-tion that enables its acquisition. There are arguably multiple types of information that we derive from our social environment that can play


roles in the acquisition of social self-knowledge, but a central kind of social information at work here is the testimony we receive through language. As speech act theory and other linguistic analyses have revealed, language is the medium of many social phenomena, particu-larly in the domain of interpersonal social relations. Relatedly, a vast portion of the information we obtain about society (and much else as well) comes from the testimony of others. Language is thus a major vehicle for self-knowledge, both in terms of the processes we considered in the previous chapter and as we come to know ourselves within the social environments described by, and sometimes even created by (the speech act of wedding vows, for instance), linguistic phenomena. Let us consider then how exactly language can play a fundamental role in the obtainment of self-knowledge, focusing on the phenomena of interper-sonal testimony in particular.

First, consider a case of direct testimony from another person about oneself. Suppose that a travel companion tells you that you often fail to check your blind spot when changing lanes while driving. This may very well be a fact about you, a fact that you would do well to know if you have an interest in being a good driver and travel companion. Assuming that your companion’s utterance bears the conditions required for successful testimonial transference of knowledge (e.g., it is a sincere expression of a reliably formed true belief), you may indeed come to know this fact about yourself via the testimonial utterance of your companion. In this case, you have derived self-knowledge from the direct testimony of another, such that the information provided to you through language constitutes knowledge you acquire about your-self, within the particular social matrix of being a driver. Testimony, then, can be the basis for social self-knowledge. This is, of course, only one example among innumerable possibilities, but it sufficiently illus-trates that we can come to know facts about ourselves through what other people directly tell us about ourselves.

Importantly, similar types of self-knowledge can be derived from indi-rect testimony as well. Indirect testimony will be understood here as testimony that is not directly about oneself but nonetheless can reveal things about oneself either implicitly or through some type of further inferential process. Arguably, this kind of information is implicit in much of human social life. Consider again the example of an instructor. If in the midst of a discussion with a student an instructor hears the student express successful understanding of some important concept on the topic at hand, the instructor could take this as testimonial evidence of success in her abilities as an educator. Again, assuming the


conditions of testimonial knowledge are met (e.g., the student’s utter-ance reflects a genuine grasp of insight), this testimony could form the basis for the acquisition of self-knowledge on the part of the instructor. In this case, however, the testimony does not contain direct informa-tion about that which comes to be known. How exactly does this work? How can testimony from another person provide you with knowledge about yourself when the testimony itself contains no direct reference to you and your qualities?

To understand this, it will be helpful to return to the spatial analogies discussed earlier. In cases of physical perception, we can obtain knowl-edge of our location without any direct representation of our location in our experience. Consider walking across a room, for instance. As you walk across a room, the changes you perceive in your visual field reveal to you your position in the room, without that position itself being overtly represented (as it might be represented on a map, for instance). Your perception in this case reveals what ecological psychologist J. J. Gibson calls ‘self-specifying information’, information that implicitly reveals properties about its recipient to its recipient by the nature of the infor-mation itself, without explicit representation of the recipient (1986). In an analogous fashion, we obtain similar types of self-specifying infor-mation about ourselves amidst the various testimonial utterances we receive from others. When a student says something that reveals her newly acquired grasp of a subject, her utterance provides self-specifying information about the conversation in which it is embedded, thereby revealing the success of what brought it about. It thus conveys informa-tion about the instructor vis- à -vis her social position as an educator, even though she may not be a direct referent of the utterance. In this way, we obtain important kinds of social self-knowledge that reveal to us who and where we are as social beings, kinds of self-knowledge that are irreducibly social in nature due to their dependency upon the social factors involved in human communication.

Arguably, this kind of indirect self-knowledge is pervasive across social phenomena, forming the backdrop through which we navigate our social position, just as we tacitly pick up our physical location via the spatio-temporal flux of phenomena in our perceptual experience, without needing to explicitly note where we are. In other words, inter-personal testimony forms a rich body of information through which people can obtain important kinds of self-knowledge, knowledge of social facts about their various positions and traits within a particular social environment, even when the testimony itself does not contain direct content about these facts. The facts can come to be known through


testimony when its content itself reveals things about the person in question, of course, but they can also be known when the testimony indirectly reveals information about the testifier, the recipient, and their relationship, thereby providing a basis by which people can infer their social traits and positions from what others say to them.

Bias and the rationality of testimonial deference

The above considerations demonstrate that there is a clearly discern-ible social side to self-knowledge, a genuine aspect of self-knowledge through which we derive vital kinds of information about ourselves from other people. We are thereby epistemically dependent upon others in knowing ourselves. This result runs counter to common conceptions of self-knowledge as a privileged and uniquely first-person state, and thus warrants some further clarification and analysis here. In particular, it calls for explanation of the particular epistemic qualities of testimony about ourselves. To what extent are the testimonial utterances of others trustworthy sources of knowledge? What are we to do when testimony is at odds with other potential sources of self-knowledge, particularly our private first-person experience? Are there cases in which our own first-person standpoint should trump the information that we receive from others? Are there cases in which we should defer to testimony over our private first-person thoughts about ourselves? Do the epistemic qualities of social self-knowledge exhibit the same conditions as other commonly recognized types of knowledge, or is it a unique domain unto itself? This chapter cannot adequately address all of these issues, but pursuing some answers to these questions will help delineate the nature and status of socially derived self-knowledge. In particular, it will be instructive to look at some relevant empirical data on bias and some recent work in episte-mology on testimonial deference, both of which will help clarify our epis-temic position with regard to the testimony of others about ourselves.

Significantly, multiple studies have revealed that people have a tendency to be erroneously biased toward themselves. These studies, like others we have already seen, indicate that we are fallible with regard to at least some of our beliefs about ourselves, opening the door for possible correction of our beliefs by appropriate external informa-tion. Perhaps the most well-known is the so-called Lake Wobegon or Better-Than-Average effect, in reference to the tendency of people to rate themselves as above average with regard to various positive traits (Alicke & Govorun, 2005 ). People regularly rate themselves as above-


average drivers, for instance, when in fact it is impossible for most people to be above average. There are also empirical findings that reveal bias in our understanding of our interpersonal relationships. For example, a social network analysis of friendship ties among a group of employees revealed a significant tendency for individuals to locate themselves closer to the center of the group than did the others’ collective represen-tations of that same person’s position in the network (Kumbasar et al., 1994 ). These over-inflated egocentric judgments suggested that

each actor within the network perceives himself or herself to be close to the center, surrounded by friends, with a dense net of ties among them, and perceives all others to be positioned further away from the actor, with sparser ties among them (p. 499).

Interestingly, however, statistical analyses of the data revealed that, while each individual exhibited an inflated assessment of their own position in the group, the aggregated reports of those same individuals produced a fairly consistent and reliable picture of the global network as a whole, suggesting that people are better at assessing the relation-ships of others than they are at assessing their own relationships. Similar results have been found in a variety of other domains as well, providing fairly robust support for the conclusions that 1) the biased position we hold toward ourselves can prevent us from accurately knowing our own social positions and characteristics and 2) other people can be better positioned to assess our own social traits than we are.

How are we to make epistemic sense of this situation? Is it rational to accept that our capacity for self-knowledge is significantly flawed and thereby defer to what others tell us about ourselves? Yes, at least in certain cases. Insofar as other people have a less biased and more informed position about our own various social traits, we can ration-ally defer to what they tell us about ourselves through their testimony. Consider, for example, the case of the travel companion discussed earlier. If a generally trustworthy person tells you that you have a dangerous tendency to ignore your blind spot while changing lanes, it is rational for you to take that information into consideration and to appropriately change your driving habits. Obstinate disregard of your companion’s testimony, perhaps based in a deep-seated first-person confidence in one’s own driving abilities, would be both epistemically negligent and pragmatically dangerous. In cases such as this, then, it is appropriate to defer to the testimony of another.


Although not focused on the particular case of self-knowledge, Elizabeth Fricker’s recent work on testimonial deference is apt here. In explaining the rationality of testimonial deference, she states that

... it is rational to accept another’s word on a topic, and even to allow her expressed judgement to override one’s own prior opinion, when one knows that she is strongly placed epistemically, and better placed than oneself, regarding the matter in question. For each of us, her appreciation of her own circumscribed and feeble epistemic powers and small position in the larger scheme of things, together with her grasp of folk psychology, including where applicable appre-ciation of others’ superior expertise and epistemically more advanta-geous position, entails that deference to others’ opinions is rational, in these circumstances. Lack of such appreciation of one’s limited powers and others’ superior ones, and an accompanying refusal to bow to others’ judgement or advice even when they are clearly relatively expert, is pig-headed irrationality, not epistemic virtue or strength (2006, p. 239).

More formally, Fricker adopts the following ‘Testimony Deferential Acceptance Principle’:

One properly accepts that P on the basis of trust in another’s testi-mony that P – her word that P – just if she speaks sincerely, and she is epistemically well enough placed with respect to P so that were she to have, or make a judgement to form, a conscious belief regarding whether P, her belief would almost certainly be knowledge; and she is better epistemically placed with respect to P than oneself; and one recognizes all these things to be so; and one is not aware of signifi-cant contrary testimony regarding P (2006, p. 232).

Consider how this principle could apply to the findings of the social network study described above. Suppose, for instance, that P is the testi-mony of person A about person B’s position in the network. B thought that he was centrally located in the network, but A sincerely reports that B is only closely connected to a handful of peripheral members of the group. What is B to think in this case? Should he stand by his judgment or should he defer to B and revise his understanding of his position in the network? The epistemic status of A’s testimony is crucial here. In keeping with the results of the study, suppose that A’s testimony meshes well with the aggregate reports of the other members of the network,


giving it strong statistical support. B’s self-focused assessment, however, is an outlier from the rest of the data, indicating centrally located network ties that are not reciprocated by the reports of the group.

If all of this were made transparent to B, the rational choice is for B to revise his beliefs. A’s testimony is a sincere judgment that is demon-strably reliable (via its correspondence with others’ reports and the statistical analyses of the study) and thus quite likely an instance of genuine knowledge. If B is made aware of these conditions, and there are no further viable considerations to suggest otherwise, then he ought to defer to A’s testimony as an instance of knowledge about his own social position. More colloquially, if someone is in a good position to know something about you and can thereby offer justifiable informa-tion about your social position or characteristics that is not mitigated by any additional factors, then it is epistemically feasible for you to accept their testimony about you.

Of course, this does not mean that one must always defer to others’ testimony. Fricker’s conditions for testimonial deference are fairly demanding, and quite rightly so. We should only defer to another’s testimony when we have some reason to think that it provides accurate information about ourselves, either directly or indirectly, such that we can reliably trust it as an instance of knowledge. As depicted in the various examples covered here, this does indeed happen with some of the personal testimony we receive from others. Other people are often in a position in which they can provide one with insight into one’s own traits, from one’s professional performance and status to one’s emotional responses and relationships with other people. To the extent that such information can be trusted, it can be the basis of significant socially derived self-knowledge, including revisionary knowledge that can override one’s prior beliefs about oneself. The testimony of others is not guaranteed to provide this knowledge, however. As Carter and Dunning ( 2008 ) note in their survey of faulty self-assessment, we can fail to know ourselves precisely because of the faulty or impoverished data available from others. So, even though it may be sometimes rational to defer to the testimony of others, we are definitely not epistemically obliged to always do so.

Just when and when not to defer to others is a delicate matter and does not admit of any simple generalities that can be offered here. This should not be too surprising, however, considering the complex status of epistemic deference in general. Appropriate deference to authorities, for example, is a multifarious issue full of contextual nuance concerning the conditions under which one may reliably trust an authority. For


instance, there is no universal generalization concerning whether or not one should heed the advice of a medical doctor. If the doctor is offering advice concerning a medical condition about which she is a trained expert, following her advice is most likely justified. If, on the other hand, she is offering advice concerning religious belief in the face of death, the epistemic status of her advice is likely much more ques-tionable. Analogously, while the considerations offered here show that we do indeed come to know ourselves from the information we get from other people, the balance of what testimony we accept or do not accept in forming beliefs about ourselves can only be determined in light of the concrete circumstances at hand in particular cases.

Conclusion

This chapter describes ways in which we obtain self-knowledge through our social interactions with others. More specifically, I have argued that, as social beings, we come to know our social positions and qualities via direct and indirect testimony from other people in our social environ-ments. While we do indeed have a uniquely first-person orientation toward ourselves that is no doubt a central facet of self-knowledge, the considerations offered here show that this simply is not all there is to self-knowledge. We learn significant things about ourselves from what others tell us, whose words can both directly and indirectly reveal to us our social positions and other socially dependent traits and proper-ties. Coupled with empirical work concerning bias and social judgment, along with epistemological assessment concerning rational deference to testimony, these considerations illuminate our epistemic dependence upon others in the perennial quest for self-knowledge.

161

In this book, I have provided a pluralist framework for understanding introspection and self-knowledge. Introspection is not a single faculty of the mind, as is commonly thought, but rather a heterogeneous collec-tion of different kinds of first-person experiences and cognitive proc-esses, with a variety of different epistemic qualities. In closing, I will summarize the main features of this account.

First of all, I have shown that the widespread conception of intro-spection as inner perception cannot be taken literally, involving instead the projection of perceptual metaphors onto our knowledge and under-standing of our own minds. Among the most salient reasons for rejecting this viewpoint are the lack of physiological processes that can be identi-fied with mental perception, the lack of mental objects with perceptu-ally identifiable properties, and the inability of a perceptual model to account for the recursive nature of introspection. Taken together, these reasons provide a strong case against the literal construal of introspec-tion as a kind of perception, thus enabling a more pluralistic framework for understanding the nature of introspection to emerge. This pluralism frames the positive approach of the book, as a survey of the major factors and processes that underlie our capacity to know our own minds.

The most basic feature of our knowledge of our own minds is the experiential self-knowledge that is intrinsic to being in a phenomenally conscious mental state. By the very fact that we enjoy conscious experi-ences, we acquire a kind of non-propositional knowledge such that we know what it is like, phenomenally speaking, to be in the conscious states we embody in life. This kind of knowledge is unique to our first-person experience of our own mentality, and is quite different in nature from the other kinds of knowledge typically recognized in the theory of knowledge. The unique nature of this kind of knowledge accounts for

Conclusion: Is That All There Is?


some common intuitions regarding introspection, such as the privileged status we take ourselves to have with regard to our own minds and the often misapplied idea that we infallibly know our own conscious mental states. However, it must also be acknowledged that this aspect of intro-spective self-knowledge is somewhat limited, pertaining only to the phenomenal character of a conscious state. This experiential knowledge, contrary to what many have inferred, does not in itself tell us anything propositional about the nature of the mental states that we experience, including whether or not they can be understood in terms of physical processes. To obtain any propositional understanding of our own minds, whereby we can make propositional claims about their nature and their properties, we must resort to additional cognitive processes.

This brings us into the domain of mental representation, conceptuali-zation, and attention. Although there are other cognitive processes that play roles in our introspective abilities, these three fundamental compo-nents of human cognition are particularly important in understanding how we are capable of knowing our own minds. The representational capacities of the mind enable us to represent our own mental states, such that our own mental states can become the objects of thought and reasoning, just as representing aspects of the external world enables us to think and reason about them. Overlapping with this is the ability to engage in conceptual thought, which allows us to place the various phenomena we experience into coherent theoretical frameworks that make sense of them. So, just as we utilize the concept of gravity to understand certain events in the world around us, we can utilize concepts of mental phenomena (such as the concepts of pain, emotion, and belief)to understand what is happening in our own minds. In addi-tion, attention enables us to guide these cognitive resources towards various phenomena, such that we can actively reflect upon and think about our own mental lives. Taken together, these cognitive processes provide a powerful explanatory framework that can account for many of our introspective abilities and accommodate them within cognitive processes that we already understand, independent of any particularly introspective function they may perform. Although these processes provide the basis for significant aspects of introspective self-knowledge, it is crucial to realize that they also open the door for epistemic errors in introspection. Intrinsic to the mediated nature of representation and conceptualization is the possibility of misrepresenting a mental state as something other than what it really is. Thus, by applying mistaken concepts to various aspects of our own mental lives, we can sometimes fail to know our own minds.

Conclusion 163

The possibility of error is particularly pertinent to our understanding of our own beliefs and desires. While it may seem that we immedi-ately know what we ourselves believe and desire, I have shown how our understanding of these aspects of our own minds involves the interpre-tive application of abstract theory-laden concepts that may or may not accurately reflect the actual contents of our own minds. To support this somewhat counterintuitive idea, I have described several different avenues of empirical research that indicate that our conceptual under-standing of our own minds is in fact fallible. Just as we inferentially apply folk psychological concepts when understanding the behavior of others, we use this understanding to infer from our first-person experi-ences the presence of folk psychological states in our own minds. And just as we can be wrong about another person’s beliefs and desires, so also can we be wrong about our own. Of course, we do sometimes, and perhaps even usually, get it right, but this accuracy is not directly given in our experience. It is dependent upon the explanatory adequacy and applicability of the particular concepts and interpretive strategies at work in our folk psychological self-attributions.

Overlapping with the various cognitive processes that enable our ability to introspect is the phenomenon of inner speech. I have suggested that the language of inner speech is a vehicle for conscious propositional thought and, moreover, a means of engaging in recur-sive conscious thought about our own thoughts. It is important to note that this is somewhat weaker than many other cognitive conceptions of language. I do not claim that natural language is the basis for thought, but rather that it is only one cognitive device, among possibly many others, with which we think. Additionally, I do not claim that natural language is the basis for consciousness in general, but rather only the basis for conscious awareness of our own propositional thoughts. Considering these cognitive abilities enabled by natural language, we can plausibly regard inner speech and the linguistic content it embodies as an important facet of introspective cognition.

This inner speech exhibits a variety of different epistemic properties. In some cases, what we utter to ourselves in inner speech constitutes a self-constitutive truth, such that the utterance is rendered true by its very occurrence. However, in most cases the utterances of inner speech are significantly fallible, and leave open the possibility of making erro-neous statements about our own minds. In this regard, inner speech exhibits the contingent fallibility of the representational and concep-tual processes mentioned earlier. At the far end of this spectrum lies the capacity to deceive ourselves, such that what we utter in inner speech


can at times confabulate false beliefs and cover up aspects of ourselves that would otherwise be readily apparent. Over time, however, such self-deception can sometimes become self-fulfilling, when what we initially deceive ourselves about becomes integrated into our thoughts and behavior in such a way that it eventually becomes true. So, even within the domain of inner speech, we can see that the epistemic char-acteristics of introspection are quite varied, providing further support for a heterogeneous characterization of introspection. Introspection is complex and multi-faceted, much like the minds it seeks to know.

Finally, it is important to understand that the pursuit and acquisition of self-knowledge is not a purely first-person process, often involving the diverse second-person social relationships we hold to others. We are dependent upon other people in knowing ourselves, sometimes warranting the relinquishment of first-person judgments in deference to the direct and indirect testimony of others. There is thus an impor-tant social side to self-knowledge that must be acknowledged and inves-tigated in a comprehensive understanding of introspection and the limits of our own first-person standpoint.

After considering the facets of introspection and self-knowledge that I have brought to light here, one may be left to wonder in the end, ‘Is that all there is?’ Do the dimensions of introspection and self-knowl-edge I portray comprehensively account for the nature, scope, and limits of what we know about ourselves? Yes but, as the song goes, let’s keep dancing. The pluralist framework I have established outlines all the main avenues of introspection, providing a comprehensive bird’s eye overview of the terrain, but there is much more to investigate and develop in the details, both in regard to the particular concrete dynamics of people’s lives as well as the more detailed and nuanced philosophical and scientific considerations that may be brought to bear across the various domains I have surveyed in broad scope. In these respects, there is certainly much more to say about introspection, and I hope I have left you with a plausible framework for doing so.

165

1 Introspection as Inner Perception

1 . Or, perhaps more accurately, they disagreed about the interpretation of results. See Monson and Hurlburt ( 1993 ) for an interesting reevaluation of the debate.

2 Poking Out the Inner Eye

1 . Of course, this is not an entirely original claim on my part. Multiple philoso-phers have criticized and/or rejected the perceptual account of introspection in various ways. See Lyons (1986), Dennett (1991), and Shoemaker (1994a and b) for a few noteworthy examples. This previous work has helped unearth problems with the perceptual account and, in part, inspires some of the reasons I give in this chapter for rejecting the view that introspection is a kind of perception.

2 . There is significant debate in the philosophy of perception concerning the relationship between perception and sensation (see e.g., Schwartz, 2004), from those who identify perception with computational processes, distinct from the sensory stimulations that serve as their input (e.g., Marr, 1982 ; Gregory, 1998), to those who understand perception itself in sensorimotor terms (e.g., Gibson, 1986; No ë , 2004), but we need not enter that territory here. All that the present point requires is that there is a dependent causal relationship between sensation and perception (even if they may be discern-ible as distinct processes), which ought to be a modest enough requirement to accommodate most accounts of perception.

3 . Exactly what kinds of objects we perceive in ordinary perceptual experi-ence is an interesting and not easily answerable issue (see e.g., Gendler & Hawthorne, 2006; Hawley & Macpherson, 2011), but I will be as non-committal as possible on the topic here. For our immediate purposes in this section, the term ‘object’ should be understood generically, as whatever it is we perceive in ordinary experience (whether they be things or events, sense data or mind-independent material things, need not be settled here).

4 . See my ‘Scan Thyself’ for more on Philip K. Dick and self-knowledge, as explored through his work and life (Butler, 2011b).

5 . This transition from direct realism to representationalism is merely rhetor-ical, with no intention of being an accurate description of the nature of mental states and their relationship to brain states.

6 . For what it is worth, I find Ruth Millikan’s biosemantic approach to mental content to be the most plausible functionalist account aired on the topic thus far, understanding the representational content of beliefs and other mental states in terms of evolutionarily-adaptive relational func-

Notes

166 Notes

tions vis- à -vis biological organisms and their environmental interactions (Millikan, 1993).

7 . This is, of course, related to the previous question concerning perceptual objects. However, it is a distinctly different question. In theory at least, one could agree that there are no perceptual objects in introspection but answer affirmatively to the question of qualia, maintaining that while we do not perceive actual objects in introspection, distinctive phenomenal qualities of mental states appear to us in a perceptual or perception-like manner. As an analogy, consider pain. One could argue that pains are not perceived as objects, despite the fact that they have distinctive phenomenal qualities through which they may be identified. This possibility is one reason to treat these two domains separately.

8 . It is important to note that I do not mean ‘unified functional process’ in the sense that there is only one identifiable process involved. It is now clear that there are distinguishable informational pathways involved in specific kinds of visual information processing. There are, for instance, separate dedicated areas of the brain for detecting color, shape, motion, faces, and perhaps other aspects of vision (Gregory, 1998; Farah, 2000; Gaulin & McBurney, 2001, pp. 106–109; Palmer & Palmer, 2002, pp. 60–65; Jacob & Jeannerod, 2003). My point here is that these individual processes, while they each constitute a functional unity in themselves, can also be said to more broadly constitute a functional unity in terms of general visual perception. They all work together to integrate visual environmental information into the various purposes of the organisms that have them.

9 . See Byrne and Whiten (1988) for an interesting account that emphasizes strategic manipulation and deception in explaining the evolution of human intelligence.

4 Knowing Our Own Consciousness

1 . For example, Colin McGinn has recently argued that consciousness should itself be understood as a kind of knowingness (2008).

2 . One might wonder how this could be achieved in actuality. Perhaps her capacity to sense red has been disabled somehow. Since this is only a thought experiment, however, we can set aside the real-world requirements of Mary’s circumstances.

3 . It is worth noting that Jackson too has more recently accepted that the knowledge argument fails to dismiss physicalism. See his Forward and Postscripts in Ludlow et al. (2004).

4 . For an influential earlier development of the concept of acquaintance, see Russell (1910). It is also worth noting that acquaintance-like characterization of introspection appears in the recent methodological approach in the scien-tific study of consciousness proposed by Petitmengin and Bitbol (2009). See my Butler (2011a) for an analysis and critique of their reliance upon acquaintance.

Notes 167

5 Introspection through Cognition

1 . Excluding cases of misrepresentation, of course, such as the fasci-nating phantom limb cases studied by Ramachandran and others. See Ramachandran and Blakeslee (1998).

2 . I will return to the topic of language in Chapter 7 , detailing the introspec-tive roles of language in inner speech, but without making commitments there to the issues surrounding introspection and external cognition I have touched on here, for the reason that the claims I make about inner speech need not depend upon the more controversial extended mind perspective. For those who follow Clark (1998; 2008) and others in regarding language as an external cognitive artifact, however, the introspective dimensions of external cognition and inner speech may be profitably brought together.

6 Understanding Our Own Beliefs and Desires

1 . I will return to some of this territory in the final chapter, addressing social cognition more broadly and the roles our social relationships with others can play in our understanding of ourselves.

2 . In fact, the rarity of folk psychological theorizing claimed by Gallagher and Ratcliffe may dovetail nicely with the rarity and relative difficulty of intro-spection I emphasized earlier in rejecting the idea of a dedicated introspec-tive faculty in the mind.

3 . To be clear, I am not saying that this is the case, but rather only that it is possible. See Heyes (1998) for a critical review of this topic.

4 . Furthermore, Hutto’s narrative practice account is a plausible bridge between the claims I am making here about folk psychology and the claims I will make in the next chapter about language. For the immediate purposes of this chapter, however, such a connection is unnecessary.

7 The Internal Monologue

1 . See Hurlburt (1990) regarding reports of inner speech. For evidence that inner speech utilizes the same brain areas as vocalized speech and external speech perception, see McGuire et al. (1996), who conclude that ‘silent articulation of sentences involves activity in an area concerned with speech generation, while imagining speech is associated with additional activity in regions associated with speech perception (p. 29).’ I will focus on auditory inner speech, due to the commonality of auditory language imagery and my own experience with it, but it is interesting to consider how comparable language imagery may occur in non-auditory sensory modalities. What experiential form do introspective propositional thoughts take for the deaf, for instance? Do thoughts occur through visual or tactile signs of language? Does a deaf person fluent in a gestural sign language feel her own thoughts through gesture images, for instance? These are fascinating and worthwhile issues to explore, but I must set them aside here. For an interesting window into the experience of sign language as a voice, see Sacks (1989).

168 Notes

2 . He has further refined his views in subsequent publications, though increas-ingly tied to his work defending modularity and dual systems theory, which need not concern us here.

3 . To be clear, the cognitive functions of language described here are not necessarily tied to the conception of language as an external artifact. I am sympathetic with Clark’s extended mind approach to cognition, as discussed earlier in Chapter 5 , but the introspective functions of language are not dependent on an external conception of language. For a sympathetic chal-lenge to the externality of Clark’s account of language, see Wheeler (2004).

4 . For more on self-reference and related phenomena, I recommend the unique and entertaining work of Douglas Hofstadter (1979; 2007), who makes much more of it than I do here.

5 . For recent work on this topic, see Moran (2001), Finkelstein (2003), and Bar-On (2004).

6 . For entry into the literature on self-deception, see McLoughlin and Rorty (1988) and Dupuy (1998).

169

Alicke , M . and Govorun , O . 2005 . ‘The Better-Than-Average Effect’. In Alicke, M., Dunning, D., and Krueger, J . (eds) The Self in Social Judgment . New York: Psychology Press: 85–108

Alter , T . and Walter , S . (eds) 2006 . Phenomenal Concepts and Phenomenal Knowledge: New Essays on Consciousness and Physicalism . Oxford: Oxford University Press.

Armstrong , D. M . 1968 . A Materialist Theory of the Mind . London: Routledge and Kegan Paul.

Armstrong, D. M . 1997 . ‘What is Consciousness?’ in Block, N., Flanagan, O., and Güzeldere, G. (eds) The Nature of Consciousness: Philosophical Debates . Cambridge, MA: MIT Press: 721–728.

Baars , B. J . 1997 . In the Theater of Consciousness: The Workspace of the Mind . Oxford: Oxford University Press.

Baron-Cohen , S . 1995 . Mindblindness: An Essay on Autism and Theory of Mind. Cambridge, MA: MIT Press.

Bar-On , D . 2004 . Speaking My Mind: Expression and Self-Knowledge . Oxford: Oxford University Press.

Berm ú dez , J. L . 1998 . The Paradox of Self-Consciousness . Cambridge: MIT Press. Bigelow , J. C . and Pargetter , R . 1990 . ‘Acquaintance with Qualia’. Theoria, 61 (3):

129–147. Boring , E. G . 1953 . ‘A History of Introspection’. Psychological Bulletin, 50 (3):

169– 189. Botterill , G . and Carruthers , P . 1999 . The Philosophy of Psychology . Cambridge:

Cambridge University Press. Butler , J . 2011a . ‘Introspective Knowledge of Experience and Its Role in

Consciousness Studies’. Journal of Consciousness Studies, 18 (2): 128–145. Butler, J . 2011b . ‘Scan Thyself’. In Wittkower, D.E. (ed.) Philip K. Dick and

Philosophy: Do Androids Have Kindred Spirits? Peru, IL: Open Court Publishing Company, pp. 77–87.

Byrne , R. W . and Whiten , A . (eds) 1988 . Machiavellian Intelligence: Social Expertise and the Evolution of Intellect in Monkeys, Apes, and Humans. Oxford: Clarendon Press.

Carruthers , P . 1996 . Language, Thought and Consciousness: An Essay in Philosophical Psychology. Cambridge: Cambridge University Press.

Carruthers , P . and Smith , P. K . (eds) 1996 . Theories of Theories of Mind . Cambridge: Cambridge University Press.

Carruthers , P . 1998 . ‘Conscious Thinking: Language or Elimination?’ Mind and Language, 13 (4): 457–476.

Carter , T . and Dunning , D . 2008 . ‘Faulty Self-Assessment: Why Evaluating One’s Own Competence Is an Intrinsically Difficult Task’. Social and Personality Psychology Compass, 2 (1): 346–360.

Chalmers , D. J . 1996 . The Conscious Mind: In Search of a Fundamental Theory . Oxford: Oxford University Press.

References

170 References

Chalmers , D . 2003 . ‘The Content and Epistemology of Phenomenal Belief’. In Smith, Q. and Jokic, A. (eds) Consciousness: New Philosophical Perspectives . Oxford: Oxford University Press. http://consc.net/papers/belief.html

Churchland , P. M . 1981 . ‘Eliminative Materialism and the Propositional Attitudes’. Journal of Philosophy, 78 (2): 67–90.

Churchland, P. M . 1984 . Matter and Consciousness: A Contemporary Introduction to the Philosophy of Mind . Cambridge, MA: MIT Press.

Churchland, P. M . 1985 . ‘Reduction, Qualia, and the Direct Introspection of Brain States’. Journal of Philosophy, 82 (1): 8–28.

Churchland , P. S . 2002 . Brain-Wise: Studies in Neurophilosophy . Cambridge, MA: MIT Press.

Clark , A . 1998 . ‘Magic Words: How Language Augments Human Computation’. In Carruthers, P. and Boucher, J. (eds) Language and Thought: Interdisciplinary Themes . Cambridge: Cambridge University Press, pp. 162–183.

Clark, A . and Chalmers , D. J . 1998 . ‘The Extended Mind’. Analysis, 58 (1): 7–19.

Clark, A . 2008 . Supersizing the Mind: Embodiment, Action, and Cognitive Extension. Oxford: Oxford University Press.

Conee , E . 1994 . ‘Phenomenal Knowledge’. Australasian Journal of Philosophy, 72 (2): 136–150.

Crick , F . 1994 . The Astonishing Hypothesis: The Scientific Search for the Soul . New York: Touchstone.

Csikszentmihalyi , M . 1997 . Finding Flow: The Psychology of Engagement with Everyday Life . New York: Basic Books.

Damasio , A . 1999 . The Feeling of What Happens: Body and Emotion in the Making of Consciousness . New York: Harcourt Brace and Company.

Danziger , K . 1980 . ‘The History of Introspection Reconsidered’. Journal of the History of the Behavioral Sciences, 16: 241–262.

D ’ Andrea , A ., Ferri , F ., and Grifoni , P . 2010 . ‘An Overview of Methods for Virtual Social Network Analysis’. In Abraham, A., Hassanien, A. E., and Sn áš el, V. (eds) Computational Social Network Analysis: Trends, Tools and Research Advances. New York: Springer, pp. 3–25.

deCharms , C . 1998 . Two Views of Mind: Abhidharma and Brain Science . Ithaca, NY: Snow Lion Publications.

Denby , B ., Schultz , T ., Honda , K ., Hueber , T ., Gilbert , J. M ., and Brumbery , J. S . 2009 . ‘Silent Speech Interfaces’. Speech Communication, 52 (4): 270–287.

Dennett , D. C . 1989 . ‘The Origins of Selves’. Cogito, 3: 163–173. Dennett, D. C . 1991 . Consciousness Explained . Boston: Little, Brown, and Co. Dennett, D. C . 1992 . ‘The Self as a Center of Narrative Gravity’. In Kessel, F.,

Cole, P., and Johnson, D. (eds) Self and Consciousness: Multiple Perspectives . Hillsdale, NJ: Erlbaum.

Descartes , R . 1637 . Discourse on the Method of Rightly Conducting One’s Reason and Seeking Truth in the Sciences, trans. Bennett, J. 2007. www.earlymoderntexts.com .

Dretske , F . 1995 . Naturalizing the Mind . Cambridge, MA: MIT Press. Dupuy , J . (ed.) 1998 . Self-Deception and Paradoxes of Rationality . Stanford: Center

for the Study of Language and Information Publications. Ericsson , K. A . and Simon , H. A . 1984 . Protocol Analysis: Verbal Reports as Data .

Cambridge, MA: MIT Press.

References 171

Ericsson, K. A . 2003 . ‘Valid and Non-Reactive Verbalization of Thoughts During Performance of Tasks’. In Jack, A. and Roepstorff, A. (eds): 1–18.

Falvey , K . 2000 . ‘The Basis of First-Person Authority’. Philosophical Topics, 28 (2): 69–99.

Farah , M. J . 2000 . The Cognitive Neuroscience of Vision . Malden, MA: Blackwell. Finkelstein , D. H . 2003 . Expression and the Inner . Cambridge, MA: Harvard

University Press. Flanagan , O . 1992 . Consciousness Reconsidered. Cambridge, MA: MIT Press. Flanagan, O . 2000 . Dreaming Souls: Sleep, Dreams, and the Evolution of the Conscious

Mind . Oxford: Oxford University Press. Fodor , J. A . 1987 . Psychosemantics . Cambridge, MA: MIT Press. Fodor, J. A . 1998 . Concepts: Where Cognitive Science Went Wrong. Oxford: Oxford

University Press. Fodor, J. A . 2001 . ‘Language, Thought and Compositionality’. Mind and Language,

16 (1): 1–15. Frege , G . 1882 /1972. Conceptual Notation and Related Articles . trans. Bynum, T.W.

Oxford: Clarendon Press. Fricker , E . 2006 . ‘Testimony and Epistemic Authority’. In Lackey, J. and Sosa, E.

(eds ) The Epistemology of Testimony . Oxford: Oxford University Press: 225–250. Fuster , J. M . 2002 . ‘Physiology of Executive Functions: The Perception-Action

Cycle’. In Stuss, D. and Knight, R. (eds ) Principles of Frontal Lobe Function . Oxford: Oxford University Press: 96–108.

Gallagher , S . 2001 . ‘The Practice of Mind: Theory, Simulation, or Interaction?’ Journal of Consciousness Studies, 8 (5–7): 83–107.

Gallagher, S . 2008 . Brainstorming: Views and Interviews on the Mind . Exeter: Imprint Academic.

Gallagher, S . and Zahavi , D . 2008 . The Phenomenological Mind: An Introduction to Philosophy of Mind and Cognitive Science. New York: Routledge.

Gaulin , S. J. C . and McBurney , D. H . 2001 . Psychology: An Evolutionary Approach. Upper Saddle River, NJ: Prentice-Hall, Inc.

Gazzaniga , M . 1998 . The Mind’s Past . Los Angeles, CA: University of California Press.

Gendler , T. S . and Hawthorne , J . (eds) 2006 . Perceptual Experience . Oxford: Oxford University Press.

Gergen , K. J . 1991 . ‘The Social Construction of Self-Knowledge’. In Kolak, D. and Martin, R. (eds) Self and Identity: Contemporary Philosophical Issues. New York: Macmillan.

Gertler , B . 2011 . Self-Knowledge . New York: Routledge. Gibson , J. J . 1986 . The Ecological Approach to Visual Perception. Hillsdale, NJ:

Lawrence Erlbaum Associates, Inc. Goldman , A . 1993 . ‘The Psychology of Folk Psychology’. Behavioral and Brain

Sciences, 16: 15–28. Goldman , A . 2006 . Simulating Minds: The Philosophy, Psychology, and Neuroscience

of Mindreading . Oxford: Oxford University Press. Gopnik , A . 1993 . ‘How We Know Our Own Minds: The Illusion of First-Person

Knowledge of Intentionality’. Behavioral and Brain Sciences, 16: 1–14. Gopnik, A . and Meltzoff , A . 1997 . Words, Thoughts, and Theories . Cambridge,

MA: MIT Press.

172 References

Gregory , R. L . 1998 . Eye and Brain: The Psychology of Seeing . 5th edition. Oxford: Oxford University Press.

G ü zeldere , G . 1997 . ‘Is Consciousness the Perception of What Passes in One’s Own Mind?’ In Block, N., Flanagan, O., and G ü zeldere, G. (eds ) The Nature of Consciousness: Philosophical Debates . Cambridge, MA: MIT Press: 789–806.

Hardcastle , V. G . 1998 . ‘The Puzzle of Attention, the Importance of Metaphors’. Philosophical Psychology, 11 (3): 331–351.

Harman , G . 1990 . ‘The Intrinsic Quality of Experience’. In Block, N., Flanagan, O., and G ü zeldere, G. (eds ) The Nature of Consciousness: Philosophical Debates . Cambridge, MA: MIT Press: 663–675.

Hawley , K . and Macpherson , F. ( eds) 2011 . The Admissible Contents of Experience . Malden, MA: Wiley-Blackwell.

Heil , J . 1988 . ‘Talk and Thought’. Philosophical Papers, 17 (3): 153–170. Heyes , C. M . 1998 . ‘Theory of Mind in Nonhuman Primates’. Behavioral and

Brain Sciences, 21: 101–114. Hofstadter , D. R . 1979 . G ö del, Escher, Bach: an Eternal Golden Braid . New York:

Basic Books. Hofstadter, D. R . 2007 . I am a Strange Loop . New York: Basic Books. Humphrey , N . 1986 . The Inner Eye . Oxford: Oxford University Press. Hurlburt , R . 1990 . Sampling Normal and Schizophrenic Inner Experience . New York:

Plenum Press. Hurlburt, R . and Schwitzgebel , E . 2007 . Describing Inner Experience? Proponent

Meets Skeptic. Cambridge, MA: MIT Press. Hutto , D . 2009 . ‘Folk Psychology as Narrative Practice’. Journal of Consciousness

Studies, 16 (6–8): 9–39. Ismael , J. T . 2007 . The Situated Self . Oxford: Oxford University Press. Itti , L ., Rees , G ., and Tsotsos , J. K . (eds) 2005 . Neurobiology of Attention . Burlington,

MA: Elsevier Academic Press. Jackendoff , R . 1996 . ‘How Language Helps Us Think’. Pragmatics and Cognition,

4 (1): 1–34. Jack , A . and Roepstorff , A. (eds) 2003 . Trusting the Subject?: The Use of Introspective

Evidence in Cognitive Science. Vol. 1. Charlottesville, VA: Imprint Academic. Jackson , F . 1982 . ‘Epiphenomenal Qualia’. The Philosophical Quarterly, 32:

127–136. Jackson, F . 1986 . ‘What Mary Didn’t Know’. The Journal of Philosophy, 83 (5):

291–295. Jacob , P . and Jeannerod , M . 2003 . Ways of Seeing: The Scope and Limits of Visual

Cognition . Oxford: Oxford University Press. James , W . 1890 . The Principles of Psychology . Vols. I and II. Cambridge, MA:

Harvard University Press. Kant , I . 1781 /1929. Critique of Pure Reason . Trans. N. K. Smith. New York: Palgrave

Macmillan. Kornblith , H . 1998 . ‘What Is It Like to be Me?’ Australasian Journal of Philosophy,

76 (1): 48–60. Kornblith, H . 2002 . Knowledge and Its Place in Nature . Oxford: Oxford University

Press. Kumbasar , E ., Romney , A. K ., and Batchelder , W. H . 1994 . ‘Systematic Biases in

Social Perception’. American Journal of Sociology, 100 (2): 477–505.

References 173

LaBerge , D . 1995 . Attentional Processing: The Brain’s Art of Mindfulness. Cambridge: Harvard University Press.

Lakoff , G . and Johnson , M . 1980 . Metaphors We Live By . Chicago: University of Chicago Press.

Lakoff, G . and Johnson , M . 1999 . Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Thought . New York: Basic Books.

Lewis , D . 1988 . ‘What Experience Teaches’. Proceedings of the Russellian Society, 13: 29–57.

Locke , J . 1689 /1975. An Essay Concerning Human Understanding. Nidditch, P. H. (ed .) Oxford: Clarendon Press.

Ludlow , P ., Nagasawa , Y ., and Stoljar , D . (eds) 2004 . There’s Something About Mary: Essays on Phenomenal Consciousness and Frank Jackson’s Knowledge Argument . Cambridge, MA: MIT Press.

Lycan , W. G . 1987 . Consciousness . Cambridge, MA: MIT Press. Lycan, W. G . 1996 . Consciousness and Experience . Cambridge, MA: MIT Press. Lyons , W . G. 1986 . The Disappearance of Introspection. Cambridge, MA: MIT

Press. Machery , E . 2005 . ‘You Don’t Know How You Think: Introspection and Language

of Thought’. British Journal of the Philosophy of Science, 56: 469–485. Maiese , M . 2011 . Embodiment, Emotion, and Cognition . Basingstoke, England:

Palgrave Macmillan. Margolis , E . and Laurence , S . (eds) 1999 . Concepts: Core Readings . Cambridge, MA:

MIT Press. Marr , D . 1982 . Vision . New York: W. H. Freeman. McGinn , C . 2008 . ‘Consciousness as Knowingness’. The Monist, 91: 237–249. McGuire , P. K ., Silbersweig , D. A ., Murray , R. M ., David , A. S ., Frackowiak , R. S .,

and Frith , C. D . 1996 . ‘Functional Anatomy of Inner Speech and Auditory Verbal Imagery’. Psychological Medicine, 26 (1): 29–38.

McLaughlin , B. P . and Rorty , A. O . (eds) 1988 . Perspectives on Self-Deception . Berkeley: University of California Press.

Mead , G. H . 1934 . Mind, Self, and Society. Chicago: University of Chicago Press. Meissner , C. A . and Memon , A . 2002 . ‘Verbal Overshadowing: A Special Issue

Exploring Theoretical and Applied Issues’. Applied Cognitive Psychology, 16: 869–872.

Menary , R . 2007 . Cognitive Integration: Mind and Cognition Unbounded . Basingstoke, England: Palgrave Macmillan.

Merleau-Ponty , M . 1945 /2002. Phenomenology of Perception . Trans. 1958. Smith, C. New York: Routledge.

Millikan , R . 1993 . White Queen Psychology and Other Essays for Alice . Cambridge, MA: MIT Press.

Monson , C. K . and Hurlburt , R. T . 1993 . ‘A Comment to Suspend the Introspection Controversy: Introspecting Subjects Did Agree About Imageless Thought’. In Hurlburt, R. T. (ed.) Sampling Inner Experience in Disturbed Affect. New York: Plenum.

Moore , G. E . 1903 . ‘The Refutation of Idealism’. Mind, 12: 433–53. Moran , R . 2001 . Authority and Estrangement: An Essay on Self-Knowledge . Princeton,

NJ: Princeton University Press. Musgrave , A . 1993 . Common Sense, Science, and Scepticism . New York: Cambridge

University Press.

174 References

Nagel , T . 1974 . ‘What Is It Like to be a Bat?’ The Philosophical Review, 83 (4): 435–50.

Neisser , U . (ed.) 1993 . The Perceived Self: Ecological and Interpersonal Sources of Self-Knowledge . Cambridge: Cambridge University Press.

Nichols , S . and Stich , S . 2003 . Mindreading: An Integrated Account of Pretence, Self-awareness, and Understanding Other Minds . Oxford: Oxford University Press.

Nida-R ü melin , M . 2010 . ‘Qualia: The Knowledge Argument’. The Stanford Encyclopedia of Philosophy. Zalta, E. N. (ed.) http://plato.stanford.edu/archives/sum2010/entries/qualia-knowledge/

Nisbett , R . and Wilson , T. D . 1977 . ‘Telling More than We Can Know: Verbal Reports on Mental Processes’. Psychological Review, 84: 231–259.

No ë , A . 2004 . Action in Perception. Cambridge, MA: MIT Press. Palmer , J. A . and Palmer , L. K . 2002 . Evolutionary Psychology: The Ultimate Origins

of Human Behavior . Boston: Allyn and Bacon. Pashler , H. E . 1998 . The Psychology of Attention . Cambridge, MA: MIT Press. Petitmengin , C . (ed.) 2009 . ‘Ten Years of Viewing from Within: The Legacy of

F. J. Varela’. Journal of Consciousness Studies , 16 (10–12 ). Petitmengin , C . and Bitbol , M . 2009 . ‘The Validity of First-Person Descriptions

as Authenticity and Coherence’. Journal of Consciousness Studies, 16 (10–12): 363–404.

Picton , T. W ., Alain , C ., and McIntosh , A . 2002 . ‘The Theatre of the Mind: Physiological Studies of the Human Frontal Lobes’. In Stuss, D. T. and Knight, R. T. (eds ) Principles of Frontal Lobe Function . Oxford: Oxford University Press: 109–126.

Powers , J . 1995 . Introduction to Tibetan Buddhism . Ithaca, NY: Snow Lion Publications.

Premack , D . and Woodruff , G . 1978 . ‘Does the Chimpanzee have a Theory of Mind?’ The Behavioral and Brain Sciences, 4: 515–526.

Ramachandran , V. S . and Blakeslee , S . 1998 . Phantoms in the Brain: Probing the Mysteries of the Human Mind . New York: HarperCollins Publishers.

Ratcliffe , M . 2007 . Rethinking Commonsense Psychology: A Critique of Folk Psychology, Theory of Mind, and Simulation. Basingstoke, England: Palgrave Macmillan.

Ristau , C. A . 1998 . ‘Cognitive Ethology: The Minds of Children and Animals’. In Cummins, D. D. and Allen, C. (eds ) The Evolution of Mind . Oxford: Oxford University Press: 127–159.

Rowlands , M . 2010 . The New Science of the Mind: From Extended Mind to Embodied Phenomenology . Cambridge, MA: MIT Press.

Russell , B . 1910 . ‘Knowledge by Acquaintance and Knowledge by Description’. Aristotelian Society Proceedings, 11: 108–128.

Russell, B . 1956 . ‘Mind and Matter’. In Portraits from Memory . Nottingham: Spokesman.

Ryle , G . 1949 . The Concept of Mind . Chicago: The University of Chicago Press. Sacks , O . 1970 . The Man Who Mistook His Wife for a Hat . New York: Touchstone. Sacks, O . 1989 . Seeing Voices: A Journey into the World of the Deaf . Berkeley:

University of California Press. Schooler , J. W . and Engstler-Schooler , T. Y . 1990 . ‘Verbal Overshadowing of Visual

Memories: Some Things are Better Left Unsaid’. Cognitive Psychology, 22: 36–71. Schwartz , R . (ed.) 2004 . Perception . Malden, MA: Blackwell Publishing.

References 175

Schwitzgebel , E . 2004 . ‘Introspective Training Apprehensively Defended: Reflections on Titchener’s Lab Manual’. Journal of Consciousness Studies, 11 (7–8): 58–76.

Searle , J . 1992 . The Rediscovery of the Mind . Cambridge, MA: MIT Press. Shoemaker , S . 1994a . ‘Self-Knowledge and ‘Inner Sense’: Lecture I: The Object

Perception Model’. Philosophy and Phenomenological Research, 54 (2): 249–269. Shoemaker, S . 1994b . ‘Self-Knowledge and ‘Inner Sense’: Lecture II: The Broad

Perceptual Model’. Philosophy and Phenomenological Research, 54 (2): 271–290. Shoemaker, S . 1994c . ‘Self-Knowledge and ‘Inner Sense’: Lecture III: The

Phenomenal Character of Experience’. Philosophy and Phenomenological Research 54 (2): 291–314.

Silvia , P. J ., and Gendolla , G. H. E . 2001 . ‘On Introspection and Self-Perception: Does Self-Focused Attention Enable Accurate Self-Knowledge?’ Review of General Psychology, 5 (3): 241–269.

Sperber , D . (ed.) 2000 . Metarepresentations: A Multidisciplinary Perspective . Oxford: Oxford University Press.

Sterelny , K . 1998 . ‘Intentional Agency and the Metarepresentation Hypothesis’. Mind and Language, 13 (1): 11–28.

Taylor , S . and Brown , J . 1988 . ‘Illusions and Well-Being: A Social Psychological Perspective on Mental Health’. Psychological Bulletin, 103: 193–210.

Thera , N . 1962 . The Heart of Buddhist Meditation . New York: Samuel Weiser, Inc. Thompson , E . (ed.) 2001 . Between Ourselves: Second-Person Issues in the Study of

Consciousness . Charlottesville, VA: Imprint Academic. Thompson , E . (ed.) 2003 . The Problem of Consciousness: New Essays in

Phenomenological Philosophy of Mind. Calgary: University of Calgary Press. Tye , M . 2009 . Consciousness Revisited: Materialism Without Phenomenal Concepts .

Cambridge, MA: MIT Press. Varela , F. J . and Shear , J . (eds) 1999 . The View From Within: First-person Approaches

to the Study of Consciousness . Bowling Green, OH: Imprint Academic. Wallace , B. A . 2000 . The Taboo of Subjectivity: Toward a New Science of Consciousness.

Oxford: Oxford University Press. Weiskrantz , L . 1986 . Blindsight: A Case Study and Implications. Oxford: Oxford

University Press. Wheeler , M . 2004 . ‘Is Language the Ultimate Artifact?’ Language Sciences, 26:

693–715. Wheeler, M . 2005 . Reconstructing the Cognitive World: The Next Step . Cambridge,

MA: MIT Press. Williams , B . 1970 . ‘The Self and the Future’. The Philosophical Review, 79 (2):

161–180. Wilson , T. D . 2002 . Strangers to Ourselves: Discovering the Adaptive Unconscious .

Cambridge, MA: The Belknap Press of Harvard University Press. Zahavi , D . 2002 . ‘First-Person Thoughts and Embodied Self-Awareness: Some

Reflections on the Relation Between Recent Analytical Philosophy and Phenomenology’. Phenomenology and the Cognitive Sciences, 1: 7–26.

177

Index

agency, 103, 132, 142appeal to ignorance, 20Armstrong, D., 12–13, 14, 18–19, 27attention, 2, 24, 26, 34–5, 37–40, 48,

75, 76, 90–4, 96, 98, 147, 162beliefs, 1, 2, 23–6, 27, 29–30, 33, 34,

49, 66–7, 71, 73, 74, 81, 82, 87, 99–118, 140–1, 146, 147, 148, 151, 156, 159, 160, 163, 164

understanding false, 34, 110–11see also folk psychology; theory of

mind

Bermúdez, J. L., 44, 68–9, 81Better-Than-Average (Lake Wobegon)

effect, 156–7Bigelow, J.C., 58blindsight, 57brains, 13, 17–20, 22–3, 27, 32,

37, 38, 39, 54, 57, 79, 81, 91, 92, 105, 107–9, 114, 132, 134, 136, 138–9

attention processes in, 38, 39

Carruthers, P., 33, 105, 123–6, 128, 129, 131, 132

Chalmers, D., 29, 58, 76, 95Churchland, P., 13–14, 27, 80, 99Clark, A., 64, 76, 95, 97, 129–30cognition, 29, 42–3, 46, 48, 64,

75–98, 120, 129, 136, 162, 1634EA (embodied, embedded,

extended, enactive, and affective), 38, 64; see also embodiment; emotions; extended mind/cognition

social, 32–4, 100–1, 148–60concepts, 2–3, 13, 26, 42, 43, 60, 61,

72, 73, 77, 82, 84–90, 99, 104–5, 122, 150, 162, 163

folk psychological, 99–118plurality/diversity of, 75

and theories, 87–9Conee, E., 58confabulation, 108, 110, 143, 145–6,

164consciousness, 12–14, 27, 32, 37, 38–9,

40, 45, 51–74, 77, 91, 92–3, 115, 150, 163

and language, 120–9, 147Csikszentmihalyi, M., 34–5

deference to others, 148–9, 156–60

Dennett, D., 38, 126–9, 146Descartes, R./Cartesianism, 5, 7–8, 9,

10, 22, 38–9, 49, 69, 99, 102, 117, 141, 148

developmental psychology, 34, 107, 110–11

Dretske, F., 76–80, 82–3, 85

embodiment, 2, 5, 17, 19, 38, 42–3, 48, 51, 53, 60, 62–5, 66, 68–9, 72, 77, 78, 80, 81, 95, 101, 118, 126, 137, 138–9, 141, 142, 143, 161, 163

emotions, 23–5, 26, 28, 30, 43, 44, 54, 61, 62, 70, 74, 84, 87, 90, 94, 124, 131, 137, 152, 159, 162

epistemology, 3, 51, 52, 56, 72, 94, 117, 142, 145, 156

see knowledge; introspectionEricsson, K. A., 144evolution, 4, 5, 16, 31–6, 97, 105,

126–7existential constitution model of

phenomenal knowledge, 2, 52, 55, 56, 62–5, 67, 68, 69, 71, 72, 73, 80

explanation, unifying, 17, 41, 48extended mind/cognition, 38, 64, 76,

94–8, 129–30

178 Index

Falvey, K., 102–3first-person authority, 100,

102–12, 141Flanagan, O., 4, 97Fodor, J., 77, 86, 99, 131, 137folk psychology, 2, 14, 45, 76,

88–9, 99–118, 119–20, 140, 142, 149, 158

simulation theory of, 100–1, 114theory theory of, 101, 103–16see also theory of mind

Frege, G., 121–3, 131Fricker, E., 158–9functionalism/functionality/

functional explanation/functional roles, 16, 18, 19, 24–5, 31–2, 39, 55, 79–80, 86, 97, 114, 127

Gallagher, S., 4, 52, 64, 100–1, 102

Gazzaniga, M., 107–8Gergen, K., 150–1Gibson, J.J., 68, 155Goldman, A., 114–16Gopnik, A., 34, 88, 105, 110–11Güzeldere, G., 27

Heidegger, M., 64Heil, J., 135–6Hofstadter, D., 44homunculus problem, 38–9Hume, D., 87Humphrey, N., 11, 32–3, 34Husserl, E., 21, 25, 64Hutto, D., 106, 118, 140

inner speech, 3, 49, 118, 119–47, 163, 164

intentionality, 135–6introspection

cognitively enabled/mediated, 75–98; see also folk psychology enabled/mediated; language enabled/mediated under introspection

defined, 46difficulty of, 34–5

diversity/heterogeneity/pluralist conception of, 1, 2, 3–4, 6, 40, 41, 47–50, 67–8, 70, 83, 91, 137–40, 149, 151, 161, 164

fallibility of, 3–4, 8, 11, 13, 49, 60, 69–73, 84, 96, 97, 99, 102, 104, 105, 106–12, 116–18, 131–2, 133, 142–7, 156–60, 163

first-person nature of, 1, 2, 3, 17, 23, 41, 46–7, 48, 63, 64, 68, 69, 71–2, 90, 92, 113, 137, 148, 150–1, 152, 156, 160, 161, 164

folk psychology enabled/mediated, 99–118

infallibility of, 63–4, 141, 162intrinsic to phenomenal

consciousness, 51–74language enabled/mediated, 96–7,

118, 119–47as a metaphor, 2, 6, 10, 17, 40–50,

91, 161non-perceptual nature of, 16–40perceptual accounts of, 6–15recursive capacity of, 2, 16, 36–7,

82, 94, 99, 118, 126–9, 137, 140, 147, 161, 163

scepticism of, 69–73introspectionism in psychology,

10–11Ismael, J. T., 152

Jackendoff, R., 127–9Jackson, F., 52, 54–5, 56, 59, 60James, W., 9–10, 11, 14, 40, 86, 91Johnson, M., 42–4, 46

Kafka, F., 16Kant, I., 9, 12, 14, 40Katha Upanishad, 16knowledge

ability, 55, 57, 58, 59, 62, 64acquaintance, 55, 57–61, 63, 64conceived through perceptual

metaphors, 42–4indexical, 152–3kinds of, 51, 55–61, 66phenomenal, 2, 48, 51–74, 78, 92,

116, 137–9, 147

Index 179

propositional, 55–64, 71, 142, 147; see also introspection, language enabled/mediated

testimonial, 148, 153–60, 164 Kornblith, H., 69–73

Lakoff, G., 42–4, 46Locke, J., 8–9, 12, 14, 27, 40Lycan, W., 14, 27, 37–40Lyons, W., 7, 10, 22, 45, 76,

105–6

Machery, E., 133–5Maiese, M., 64–5materialism, 13, 19, 38–9

eliminative, 13–14, 45–6Mead, G. H., 149–51meditation, 15, 35, 93Meltzoff, A., 34, 88, 105memory, 36, 57, 75, 76, 96, 109,

111, 143Menary, R., 38, 64, 75, 76, 95–6mental states

as objects of conceptual representation, 26–7, 61, 81–8

as objects of perception, 23–7knowledge of others’, 33–4, 103,

107, 157Merleau-Ponty, M., 21, 64, 122–3,

131, 138metaphysics, of consciousness, 51–2,

54–5, 59–60metarepresentation, 76, 81–4, 85,

87, 93misrepresentation, 76, 84, 111, 142Musgrave, A., 104–5

Nagel, T., 52–5, 59, 60, 68, 70, 79narrative processes/practices, 3,

49, 106, 117–18, 140, 141, 145, 146

naturalism, 4–5, 17, 19, 60, 136Neisser, U., 11Nichols, S., 112–14Nisbett, R., 72, 109–10, 142–3Noë, A., 21, 64

observation, theory-laden, 104–5

Pargetter, R., 58perception, 1, 6–40, 68–9, 82, 92,

129, 134–5, 155displaced, 83–4introspection as a kind of, 6–15,

45, 87, 92, 138–9, 161metaphorical concepts of,

41–4, 48objects of, 21–7 and perceptual constancy, 23–4

phenomenal properties, 21, 28–31, 51–74, 78–80, 114–16, 119, 134, 150

phenomenology, 4, 10, 16, 21–31, 36, 64, 112, 114–16, 122–3, 126, 134, 138–9

Plato/Platonism, 86, 121, 122pluralism, methodological, 1, 4–5

see also introspection, diversity/heterogeneity/pluralist conception of; cognition, plurality/diversity of

proprioception/bodily perception, 18, 68–9, 80–1

propositional attitudes, see folk psychology; knowledge, propositional

qualia, see phenomenal properties

Ratcliffe, M., 100–1, 148representations and representational

abilities/properties, 2, 3, 23, 33, 39, 76–87, 90, 93, 94, 96, 98, 111, 112, 120, 122, 134, 139, 140, 142, 143, 152, 155, 157, 162, 163

Russell, B., 60Ryle, G., 5, 38

Sartre, J. P., 64Searle, J., 25–6self

concepts/representations of, 69, 82, 89

as subject of experience, 25–6

self-deception, 3, 140, 145–7, 164self-perception in psychology, 11

180 Index

Simon, H., 144social network analysis,

152, 157Sperry, R., 107split-brain epileptics, 107–9Stich, S., 112–14

theory of mind, 33, 87–9, 99–118

evolutionary role of, 33–4see also folk psychology

Thompson, E., 52, 148

transparency of experience/mental states, 30, 94, 99, 103, 107, 111, 117, 133

Tye, M., 59

verbal overshadowing, 143–4

Wallace, B. A., 7, 19–20Wheeler, M., 7, 38, 64Wilson, T., 72, 109–10, 142–3

Zahavi, D., 25, 52, 64

rethinking introspection: a pluralist approach to the first-person perspective

Documents